Past Presentations

The following presenters have agreed to share their presentations with the MaMLIC group

Past Presentations

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

Past Event Page

Presenter:

Sangwook Kim, PhD Student
University of Toronto

Abstract:

Deep learning-based automated treatment planning cansignificantly enhance the efficiency and accuracy of radiotherapy. However, current planning approaches often depend on manually generated contours, limiting their efficiency. To address this, Sangwook and his colleagues implemented a multi-task learning framework that integrates automated contouring with voxel-based dose prediction, reducing the need for manual input and streamlining the planning process. Using two datasets – an in-house prostate cancer dataset and the publicly available OpenKBP head and neck cancer dataset – they developed a system that performs simultaneous segmentation and dose prediction.

Compared to conventional methods, the new framework improved the average absolute difference in dose volume histogram metrics by 2.90% for prostate cancer and 13.12% for head and neck cancer. Additionally, it enhanced dose prediction performance while maintaining high segmentation accuracy, with dice score coefficients of 0.824 for prostate and 0.716 for head and neck, compared to baseline scores of 0.818 and 0.674, respectively. These improvements can lead to more precise treatment plans and better patient outcomes.

The multi-task learning framework is not only generalizable to other anatomical sites and conditions but also holds promise for significantly reducing clinical workload and enhancing radiotherapy efficiency. By integrating automated contouring and dose prediction, this newapproach minimizes the need for sequential steps in the planning process, potentially allowing clinics to handle higher patient volumes with greater consistency and accuracy. This work illustrates the potential of AI to enable fully automated, efficient radiotherapy planning, supporting broader adoption of AI-driven tools in clinical practice.

Applications of Automated Treatment Planning in Adaptive Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Applications of Automated Treatment Planning in Adaptive Radiotherapy

Past Event Page

Presenter:

Aly Khalifa, PhD Student
University of Toronto

Abstract:

Technological advancements in radiotherapy have significantly enhanced treatment precision and patient safety, but they have also introduced a greater burden of manual, time-intensive tasks for clinicians to manage these innovations effectively. Machine Learning (ML) offers a promising solution to streamline these processes and optimize treatment outcomes.

The work of Aly and his colleagues explores applications of automated treatment planning in adaptive radiotherapy procedures. ML is used to predict an ideal radiation dose distribution based on the position of the tumour during treatment to reduce unnecessary radiation exposure to healthy tissues. In comparison to existing clinical methods, the ML automated technique requires no human intervention during the planning process. This removes the reliance on human skill to drive the treatment process.

The work of Aly and his colleagues demonstrates that ML improves the quality of treatment by reducing the radiation dose delivered to healthy tissues, compared to current clinical adaptation techniques. These findings suggest that automated ML-based approaches could substantially improve clinical workflows and patient outcomes in adaptive radiotherapy.

Chatbots as Mental Healthcare Proxies: Possibilities and Limitations

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Chatbots as Mental Healthcare Proxies: Possibilities and Limitations

Past Event Page

Presenter:

Catherine Stinson, PhD
Assistant Professor, Queen’s University

Abstract:

There is a great need for more affordable, accessible mental health treatment options, especially labour-intensive talk therapy. Given the impressive abilities of a new generation of chatbots like ChatGPT to mimic human conversational skills, there is hope that they might prove useful as proxies for human psychotherapists. In particular, there is hope that for communities facing barriers to mental health care, chatbots might fill the gap. We look in detail at the current generation of chatbots to understand what they do well, and what their limitations are, with support from empirical work in natural language processing. Where these tools perform best is on formulaic language tasks, in domains well covered in training corpora, using standard English. Unfortunately among the under-served communities are migrant and minority groups who may not communicate in standard English, and are not well represented in training corpora. For some psychotherapeutic interactions, particularly formulaic ones, the capacities of chatbots may be a good match. However, for for interactions where an empathetic relationship is essential, the current generation of therapy chatbots are potentially dangerous. While there is some room for chatbots to act as proxies for human psychotherapists, we should not overestimate their abilities.

Disease:

  • Mental Health

Other:

  • Health Equity

Healthcare Horizons: AI to Navigate the Evolving Landscape

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Healthcare Horizons: AI to Navigate the Evolving Landscape

Past Event Page

Presenter:

Elham Dolatabadi, PhD
Assistant Professor, York University

Abstract:

As healthcare evolves, data expands, and learning algorithms become increasingly sophisticated, the integration of artificial intelligence (AI) emerges as a transformative force, reshaping the healthcare landscape. In this talk, Elham will explore how AI goes beyond being a mere tool, driving revolutionary changes across the field. From continuous monitoring for specific single tasks to complex multimodal decision-making and conversational AI, the potential applications are vast. However, this journey is not without its challenges. Key hurdles include robust evaluation processes, mitigating biases, and cultivating the right mindset and knowledge to ensure that AI innovations are both effective and equitable. These challenges, and the strategies to overcome them, will be the focus of this talk.

Other:

  • Health Equity

AI-Driven Multimodal Computational Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Faisal Mahmood

Abstract:

This talk will discuss the use of AI in pathology and methods of training digitized images through the utilization of smaller patches to make the data efficient. Models have been developed to identify the most important images patches. In addition, the development of a breast cancer lymph node metastasis model will be discussed, showing how different imaging approaches were used. Models predicting histology have been developed for cancer applications.

Modality:

  • CT
  • Optical/microscopy
  • Histopathology
  • Genomics

Organ

  • Breast
  • Cardiac
  • Prostate

Disease

  • Cancer

AI-Based Precision Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Mattias Rantalainen

Abstract:

This talk will discuss precision medicine and its use in the development of AI in histopathology for the development of outcome prediction and treatment response prediction models. The development of a risk stratification model for breast cancer histological assessment will be discussed and the development of a medical device.

Modality:

  • Optical/microscopy
  • Gene expression
  • Histopathology

Organ

  • Breast
  • Lung
  • Prostate
  • Skin
  • Colorectal

Disease

  • Cancer

Other

  • Commercialization

Machine learning to predict disease progression

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Predicting Disease Progression

Past Event Page

Presenter:

Sun-In Lee

Abstract:

This talk will describe the use of text and phenotypic data in medical records for predicting patients’ clinical course, and potential uses for medical management. The presentation will be followed by questions and answers about the talk, and discussion focused on lessons learned, opportunities to collaborate, and sharing of data.

Modality:

  • Gene expression

Organ

  • Brain
  • Hematopathology
  • Lung
  • Skin

Disease

  • Cancer
  • Kidney and Liver Disease
  • Neurological Disease
  • Stroke and Cardiovascular
  • COVID

Other

  • Therapeutics
  • Personalized medicine

Segment Anything in Medical Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

202312-MedSAM

Open Forum on Enforcing Geometry in Machine Learning for Computational Neuroimaging

Past Event Page

Presenter:

Jun Ma

Abstract:

Medical imaging plays an indispensable role in clinical practice. Accurate and efficient medical image segmentation provides a means of delineating regions of interest and quantifying various clinical metrics. However, building customized segmentation models for each medical imaging task can be a daunting and time-consuming process, limiting the widespread adoption in clinical practice. In this talk, Jun will introduce MedSAM, a segmentation foundation model that enables universal segmentation across a wide range of medical imaging tasks and modalities. MedSAM achieved remarkable improvements in 30 segmentation tasks, surpassing the existing segmentation foundation model by a large margin. MedSAM also demonstrated zero-shot and few-shot capabilities to segment unseen tumor types and adapt to new imaging modalities with minimal effort. The results validate the versatility of MedSAM compared to existing customized segmentation models, emphasizing its potential to transform medical image segmentation and enhance clinical practice.

Disease

  • Cancer

Other

  • Resource Sharing
  • Reproducibility
Download Presentation PDF

Artificial intelligence augmented ultrasound detection of hip dysplasia

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2023-04-21Jaremko

Open Forum on AI and Precision Health Initiatives at the University of Alberta

Past Event Page

Presenter:

Jacob Jaremko

Abstract:

These talks will present new efforts underway at the University of Alberta in Edmonton AB to develop and translate AI technology into clinical practice. The College of Health Sciences at UAlberta is leading a new initiative in AI for precision health. These efforts are conducted in collaboration with the Alberta Machine Intelligence Institute (Amii), one of three national AI institutes supported through the Pan-Canadian AI Strategy. Amii is located in Edmonton and supports 40 professors conducting basic and applied AI research. Precision Health is the largest application area within Amii.

The UAlberta College of Health Sciences also collaborates extensively with Alberta Health Services (AHS), Canada’s first and largest province-wide, fully integrated health system. AHS provides health services to over 4.4 million people. For the past 20 years they have developed and maintained a broadly integrated data warehouse to collect and protect health data everywhere they provide services. This allows construction of large population-level datasets for precision health research.

These presentations will discuss some of the projects currently underway, and look ahead at new efforts to expand AI and Precision Health research at UAlberta and beyond.

Modality:

  • US

Organ:

  • Musculoskeletal

Disease

  • Musculoskeletal Disease

Other

  • Commercialization
Download Presentation PDF

Sensitivity of convolutional neural networks to common imaging parameters, perturbations and artifacts in MRI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

lyndon_malmic_2021

Open Forum on Alzheimer’s and Small Vessel Disease Imaging

Past Event Page

Presenter:

Lyndon Boone

Abstract:

A number of studies have shown that deep learning methods are capable of achieving near-human-level performance on neuroimaging segmentation tasks. With that said, most of the results quoted in the literature supporting this statement are in the context of test sets drawn from the same overarching dataset as the training data. Clinically-deployed models faced with out-of-distribution data (i.e. data that does not resemble the training set) may severely underperform relative to the standards set in the literature if they aren’t designed specifically with robustness to out-of-distribution data in mind. In this talk, Lyndon highlighted the sensitivity of modern CNN-based architectures to image corruptions, artifacts, and post-processing transforms commonly found in MRI. He then presented a methodology for benchmarking different architectures and models on the basis of robustness to out-of-distribution data, inspired by similar work in the computer vision literature.

Modality:

  • MR

Organ:

  • Brain
Download Presentation PDF

Deep Learning for Automated Segmentation of Left Ventricle Myocardium and Myocardial Scar From 3-D MR Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-06-18Zabihollahy

Open Forum on Machine Learning in Cardiac Imaging

Past Event Page

Presenter:

Fatemeh Zabihollahy

Abstract:

Deep learning has demonstrated promise for various cardiac imaging applications. However, the performance is usually degraded when the models are trained with small and under-annotated training datasets and tested on previously unseen domains, limiting the potential for broad clinical use. In this talk, Fumin presented his recent work on combining deep learning and machine learning models for cardiac MRI segmentation, where smaller datasets and fewer annotations are required for algorithm training. He also provided examples of integrating the segmentation tools for myocardial infarct heterogeneity quantification in contrast enhancement MRI in the context of MRI-guided cardiac arrhythmia treatment.

Modality:

  • MR

Organ:

  • Cardiac

Disease

  • Heart Disease
  • Stroke and Cardiovascular
Download Presentation PDF

Deep learning with uncertainty quantification in MRI-guided radiation therapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Hemsley

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Matt Hemsley

Abstract:

Deep learning methods are able to match or surpass the performance of ​​classical methods as well as human experts in a growing number of medical imaging related tasks. However, most deep learning methods are unable to quantify uncertainty, hindering clinical translation. In this talk, Matt presented an overview of techniques used to model uncertainty on network output, demonstrated the connection between uncertainty quantification and reproducibility, and presented examples where uncertainty estimates were obtained from networks that performed clinical tasks related to real-time MRI-guided adaptive radiation therapy.

Modality:

  • CT
  • MR

Organ:

  • Brain
  • Head and Neck

Disease

  • Cancer
Download Presentation PDF

Design, conduct, and reporting of radiomic analyses: Let’s not reinvent the wheel

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Moskowitz

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Chaya Moskowitz

Abstract:

The number of published papers reporting on radiomic analyses has been growing exponentially. Very few of these paper produce results that are translated into clinical practice at least in part because of methodological flaws in the work. Although complex methods for producing radiomic signatures are increasingly available and user-friendly and progress has been made in radiomic biomarker taxonomy and standardization, fundamental elements of study design, rigorous statistical analysis, and quality of reporting methods and results are frequently overlooked. In this talk, Chaya highlighted common pitfalls encountered in radiomic studies that could be avoided by knowledge of existing methods and adherence to existing reporting standards.

Other:

  • Reproducibility
Download Presentation PDF

Improving Peproducibility in Machine Learning Research

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Haibe-Kains

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Benjamin Haibe-Kains

Abstract:

While biostatistics and machine learning are essential to analyze biomedical data, researchers are facing multiple challenges around research reproducibility and transparency. It is essential for independent researchers to be able to scrutinize and reproduce the results of a study using its materials, and build upon them in future studies.

Computational reproducibility is achievable when the data can easily be shared and the required computational resources are relatively common. However, the complexity of current algorithms and their implementation, the need for specific computer hardware and the use of sensitive biomedical data represent major obstacles in healthy-related research. In this talk, Benjamin described the various aspects of a typical biomedical study that are necessary for reproducibility and the platforms that exist for sharing these materials with the scientific community.

Other:

  • Resource Sharing
  • Reproducibility
Download Presentation PDF

Introduction to the Ontario Health Data Platform

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

OHDP2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenter:

Amber Simpson

Abstract:

The Ontario Health Data Platform (OHDP) is a secure, federated high-performance computing environment built for linkage and analysis of large health data sets. The platform was built to aid Ontario in their fight against COVID-19. This talk described the platform, available data, and the potential for future data such as genomic and imaging data repositories.

Other:

  • Data source/database
Download Presentation PDF

Brain-CODE

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Brain-CODE2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenters:

Tom Mikkelsen, Kirk Nylen, Brendan Behan

Abstract:

The Ontario Brain Institute (OBI) provided an overview of our Brain-CODE platform, its data holdings, and the opportunity to access and analyze highly standardized data on people living with various brain disorders – ranging from pediatric neurodevelopmental disorders through to neurodegenerative diseases of aging. We hoped to engage high-end data scientists like you in a discussion around potential collaboration opportunities with our large network of clinical experts/researchers.\

Modality:

  • MR
  • Gene Expression

Organ:

  • Brain

Disease:

  • Neurological Disease
  • Phantom
  • Psychological
Download Presentation PDF

QIPCM – Advanced Imaging Core Lab

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

QIPCM2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenter:

Julia Publicover

Abstract:

The QIPCM imaging core lab was conceived for high-quality, auditable and secure image management for clinical trials. More and more, users are looking to leverage the platform for large volume image analysis and processing, including radiomics and ML applications. This talk gave an overview of the QIPCM data-sharing platform including tools, services and quality management program that ensure high-quality image collection, analysis and regulatory compliance. I also discussed current use-cases in radiomics and ML and future aims.

Disease:

  • Cancer

Other:

  • Data source/database
Download Presentation PDF

Deep learning for automatic prostate segmentation in 3D ultrasound

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Orlando2021-02-19

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Nathan Orlando

Abstract:

Nathan discussed the development and validation of a generalizable and efficient deep learning method for automatic prostate segmentation in 3D ultrasound, compared performance to state of the art algorithms and explored the impact of dataset size and diversity on segmentation performance.

Modality:

  • US

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

MRI in Prostate Cancer: Opportunities for AI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

OICR-Prostate-ML-Haider

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Masoom Haider

Abstract:

MRI has established itself as an adjunctive test in detecting occult prostate cancer. New guidelines will recommend MRI for patients with elevated PSA and no prior biopsy resulting in a marked increase in MRI in the province. There are several use cases that can benefit from the application of machine learning to improve diagnostic performance and outcomes, which was reviewed.

Modality:

  • MRI
  • Gene Expression

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

Machine learning for prognostic modelling in head and neck cancer using multimodal data

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Kazmierski2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Michal Kazmierski

Abstract:

Michal discussed the results of a machine learning challenge for head and neck cancer (HNC) survival prediction with the aim of 1) developing an accurate prognostic model for HNC survival using clinical, demographic and routinely collected CT imaging data and 2) evaluating the true added value of CT radiomics compared to other prognostic factors.

Modality:

  • CT

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Radiomics and Machine Learning for Oropharyngeal Cancer

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Mattonen2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Sarah Mattonen

Abstract:

Sarah discussed preliminary work investigating radiomics and machine learning models to predict outcomes and toxicities in patients with oropharyngeal cancer treated with chemoradiotherapy. Sarah also discussed efforts to validate existing models, and described some challenges and opportunities for radiomics research in head and neck cancer.

Modality:

  • CT
  • MR

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Modality

  • All
  • - CT
  • - Gene Expression
  • - Genomics
  • - Histopathology
  • - MR
  • - Optical / Microscopy
  • - US

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

Past Event Page

Presenter:

Sangwook Kim, PhD Student
University of Toronto

Abstract:

Deep learning-based automated treatment planning cansignificantly enhance the efficiency and accuracy of radiotherapy. However, current planning approaches often depend on manually generated contours, limiting their efficiency. To address this, Sangwook and his colleagues implemented a multi-task learning framework that integrates automated contouring with voxel-based dose prediction, reducing the need for manual input and streamlining the planning process. Using two datasets – an in-house prostate cancer dataset and the publicly available OpenKBP head and neck cancer dataset – they developed a system that performs simultaneous segmentation and dose prediction.

Compared to conventional methods, the new framework improved the average absolute difference in dose volume histogram metrics by 2.90% for prostate cancer and 13.12% for head and neck cancer. Additionally, it enhanced dose prediction performance while maintaining high segmentation accuracy, with dice score coefficients of 0.824 for prostate and 0.716 for head and neck, compared to baseline scores of 0.818 and 0.674, respectively. These improvements can lead to more precise treatment plans and better patient outcomes.

The multi-task learning framework is not only generalizable to other anatomical sites and conditions but also holds promise for significantly reducing clinical workload and enhancing radiotherapy efficiency. By integrating automated contouring and dose prediction, this newapproach minimizes the need for sequential steps in the planning process, potentially allowing clinics to handle higher patient volumes with greater consistency and accuracy. This work illustrates the potential of AI to enable fully automated, efficient radiotherapy planning, supporting broader adoption of AI-driven tools in clinical practice.

Applications of Automated Treatment Planning in Adaptive Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Applications of Automated Treatment Planning in Adaptive Radiotherapy

Past Event Page

Presenter:

Aly Khalifa, PhD Student
University of Toronto

Abstract:

Technological advancements in radiotherapy have significantly enhanced treatment precision and patient safety, but they have also introduced a greater burden of manual, time-intensive tasks for clinicians to manage these innovations effectively. Machine Learning (ML) offers a promising solution to streamline these processes and optimize treatment outcomes.

The work of Aly and his colleagues explores applications of automated treatment planning in adaptive radiotherapy procedures. ML is used to predict an ideal radiation dose distribution based on the position of the tumour during treatment to reduce unnecessary radiation exposure to healthy tissues. In comparison to existing clinical methods, the ML automated technique requires no human intervention during the planning process. This removes the reliance on human skill to drive the treatment process.

The work of Aly and his colleagues demonstrates that ML improves the quality of treatment by reducing the radiation dose delivered to healthy tissues, compared to current clinical adaptation techniques. These findings suggest that automated ML-based approaches could substantially improve clinical workflows and patient outcomes in adaptive radiotherapy.

AI-Driven Multimodal Computational Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Faisal Mahmood

Abstract:

This talk will discuss the use of AI in pathology and methods of training digitized images through the utilization of smaller patches to make the data efficient. Models have been developed to identify the most important images patches. In addition, the development of a breast cancer lymph node metastasis model will be discussed, showing how different imaging approaches were used. Models predicting histology have been developed for cancer applications.

Modality:

  • CT
  • Optical/microscopy
  • Histopathology
  • Genomics

Organ

  • Breast
  • Cardiac
  • Prostate

Disease

  • Cancer

AI-Based Precision Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Mattias Rantalainen

Abstract:

This talk will discuss precision medicine and its use in the development of AI in histopathology for the development of outcome prediction and treatment response prediction models. The development of a risk stratification model for breast cancer histological assessment will be discussed and the development of a medical device.

Modality:

  • Optical/microscopy
  • Gene expression
  • Histopathology

Organ

  • Breast
  • Lung
  • Prostate
  • Skin
  • Colorectal

Disease

  • Cancer

Other

  • Commercialization

Machine learning to predict disease progression

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Predicting Disease Progression

Past Event Page

Presenter:

Sun-In Lee

Abstract:

This talk will describe the use of text and phenotypic data in medical records for predicting patients’ clinical course, and potential uses for medical management. The presentation will be followed by questions and answers about the talk, and discussion focused on lessons learned, opportunities to collaborate, and sharing of data.

Modality:

  • Gene expression

Organ

  • Brain
  • Hematopathology
  • Lung
  • Skin

Disease

  • Cancer
  • Kidney and Liver Disease
  • Neurological Disease
  • Stroke and Cardiovascular
  • COVID

Other

  • Therapeutics
  • Personalized medicine

Artificial intelligence augmented ultrasound detection of hip dysplasia

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2023-04-21Jaremko

Open Forum on AI and Precision Health Initiatives at the University of Alberta

Past Event Page

Presenter:

Jacob Jaremko

Abstract:

These talks will present new efforts underway at the University of Alberta in Edmonton AB to develop and translate AI technology into clinical practice. The College of Health Sciences at UAlberta is leading a new initiative in AI for precision health. These efforts are conducted in collaboration with the Alberta Machine Intelligence Institute (Amii), one of three national AI institutes supported through the Pan-Canadian AI Strategy. Amii is located in Edmonton and supports 40 professors conducting basic and applied AI research. Precision Health is the largest application area within Amii.

The UAlberta College of Health Sciences also collaborates extensively with Alberta Health Services (AHS), Canada’s first and largest province-wide, fully integrated health system. AHS provides health services to over 4.4 million people. For the past 20 years they have developed and maintained a broadly integrated data warehouse to collect and protect health data everywhere they provide services. This allows construction of large population-level datasets for precision health research.

These presentations will discuss some of the projects currently underway, and look ahead at new efforts to expand AI and Precision Health research at UAlberta and beyond.

Modality:

  • US

Organ:

  • Musculoskeletal

Disease

  • Musculoskeletal Disease

Other

  • Commercialization
Download Presentation PDF

Sensitivity of convolutional neural networks to common imaging parameters, perturbations and artifacts in MRI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

lyndon_malmic_2021

Open Forum on Alzheimer’s and Small Vessel Disease Imaging

Past Event Page

Presenter:

Lyndon Boone

Abstract:

A number of studies have shown that deep learning methods are capable of achieving near-human-level performance on neuroimaging segmentation tasks. With that said, most of the results quoted in the literature supporting this statement are in the context of test sets drawn from the same overarching dataset as the training data. Clinically-deployed models faced with out-of-distribution data (i.e. data that does not resemble the training set) may severely underperform relative to the standards set in the literature if they aren’t designed specifically with robustness to out-of-distribution data in mind. In this talk, Lyndon highlighted the sensitivity of modern CNN-based architectures to image corruptions, artifacts, and post-processing transforms commonly found in MRI. He then presented a methodology for benchmarking different architectures and models on the basis of robustness to out-of-distribution data, inspired by similar work in the computer vision literature.

Modality:

  • MR

Organ:

  • Brain
Download Presentation PDF

Deep Learning for Automated Segmentation of Left Ventricle Myocardium and Myocardial Scar From 3-D MR Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-06-18Zabihollahy

Open Forum on Machine Learning in Cardiac Imaging

Past Event Page

Presenter:

Fatemeh Zabihollahy

Abstract:

Deep learning has demonstrated promise for various cardiac imaging applications. However, the performance is usually degraded when the models are trained with small and under-annotated training datasets and tested on previously unseen domains, limiting the potential for broad clinical use. In this talk, Fumin presented his recent work on combining deep learning and machine learning models for cardiac MRI segmentation, where smaller datasets and fewer annotations are required for algorithm training. He also provided examples of integrating the segmentation tools for myocardial infarct heterogeneity quantification in contrast enhancement MRI in the context of MRI-guided cardiac arrhythmia treatment.

Modality:

  • MR

Organ:

  • Cardiac

Disease

  • Heart Disease
  • Stroke and Cardiovascular
Download Presentation PDF

Deep learning with uncertainty quantification in MRI-guided radiation therapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Hemsley

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Matt Hemsley

Abstract:

Deep learning methods are able to match or surpass the performance of ​​classical methods as well as human experts in a growing number of medical imaging related tasks. However, most deep learning methods are unable to quantify uncertainty, hindering clinical translation. In this talk, Matt presented an overview of techniques used to model uncertainty on network output, demonstrated the connection between uncertainty quantification and reproducibility, and presented examples where uncertainty estimates were obtained from networks that performed clinical tasks related to real-time MRI-guided adaptive radiation therapy.

Modality:

  • CT
  • MR

Organ:

  • Brain
  • Head and Neck

Disease

  • Cancer
Download Presentation PDF

Brain-CODE

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Brain-CODE2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenters:

Tom Mikkelsen, Kirk Nylen, Brendan Behan

Abstract:

The Ontario Brain Institute (OBI) provided an overview of our Brain-CODE platform, its data holdings, and the opportunity to access and analyze highly standardized data on people living with various brain disorders – ranging from pediatric neurodevelopmental disorders through to neurodegenerative diseases of aging. We hoped to engage high-end data scientists like you in a discussion around potential collaboration opportunities with our large network of clinical experts/researchers.\

Modality:

  • MR
  • Gene Expression

Organ:

  • Brain

Disease:

  • Neurological Disease
  • Phantom
  • Psychological
Download Presentation PDF

Deep learning for automatic prostate segmentation in 3D ultrasound

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Orlando2021-02-19

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Nathan Orlando

Abstract:

Nathan discussed the development and validation of a generalizable and efficient deep learning method for automatic prostate segmentation in 3D ultrasound, compared performance to state of the art algorithms and explored the impact of dataset size and diversity on segmentation performance.

Modality:

  • US

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

MRI in Prostate Cancer: Opportunities for AI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

OICR-Prostate-ML-Haider

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Masoom Haider

Abstract:

MRI has established itself as an adjunctive test in detecting occult prostate cancer. New guidelines will recommend MRI for patients with elevated PSA and no prior biopsy resulting in a marked increase in MRI in the province. There are several use cases that can benefit from the application of machine learning to improve diagnostic performance and outcomes, which was reviewed.

Modality:

  • MRI
  • Gene Expression

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

Machine learning for prognostic modelling in head and neck cancer using multimodal data

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Kazmierski2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Michal Kazmierski

Abstract:

Michal discussed the results of a machine learning challenge for head and neck cancer (HNC) survival prediction with the aim of 1) developing an accurate prognostic model for HNC survival using clinical, demographic and routinely collected CT imaging data and 2) evaluating the true added value of CT radiomics compared to other prognostic factors.

Modality:

  • CT

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Radiomics and Machine Learning for Oropharyngeal Cancer

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Mattonen2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Sarah Mattonen

Abstract:

Sarah discussed preliminary work investigating radiomics and machine learning models to predict outcomes and toxicities in patients with oropharyngeal cancer treated with chemoradiotherapy. Sarah also discussed efforts to validate existing models, and described some challenges and opportunities for radiomics research in head and neck cancer.

Modality:

  • CT
  • MR

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Organ

  • All
  • - Bladder
  • - Brain
  • - Breast
  • - Cardiac
  • - Colorectal
  • - Head-Neck
  • - Hematopathology
  • - Lung
  • - Musculoskeletal
  • - Prostate
  • - Skin

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

Past Event Page

Presenter:

Sangwook Kim, PhD Student
University of Toronto

Abstract:

Deep learning-based automated treatment planning cansignificantly enhance the efficiency and accuracy of radiotherapy. However, current planning approaches often depend on manually generated contours, limiting their efficiency. To address this, Sangwook and his colleagues implemented a multi-task learning framework that integrates automated contouring with voxel-based dose prediction, reducing the need for manual input and streamlining the planning process. Using two datasets – an in-house prostate cancer dataset and the publicly available OpenKBP head and neck cancer dataset – they developed a system that performs simultaneous segmentation and dose prediction.

Compared to conventional methods, the new framework improved the average absolute difference in dose volume histogram metrics by 2.90% for prostate cancer and 13.12% for head and neck cancer. Additionally, it enhanced dose prediction performance while maintaining high segmentation accuracy, with dice score coefficients of 0.824 for prostate and 0.716 for head and neck, compared to baseline scores of 0.818 and 0.674, respectively. These improvements can lead to more precise treatment plans and better patient outcomes.

The multi-task learning framework is not only generalizable to other anatomical sites and conditions but also holds promise for significantly reducing clinical workload and enhancing radiotherapy efficiency. By integrating automated contouring and dose prediction, this newapproach minimizes the need for sequential steps in the planning process, potentially allowing clinics to handle higher patient volumes with greater consistency and accuracy. This work illustrates the potential of AI to enable fully automated, efficient radiotherapy planning, supporting broader adoption of AI-driven tools in clinical practice.

Applications of Automated Treatment Planning in Adaptive Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Applications of Automated Treatment Planning in Adaptive Radiotherapy

Past Event Page

Presenter:

Aly Khalifa, PhD Student
University of Toronto

Abstract:

Technological advancements in radiotherapy have significantly enhanced treatment precision and patient safety, but they have also introduced a greater burden of manual, time-intensive tasks for clinicians to manage these innovations effectively. Machine Learning (ML) offers a promising solution to streamline these processes and optimize treatment outcomes.

The work of Aly and his colleagues explores applications of automated treatment planning in adaptive radiotherapy procedures. ML is used to predict an ideal radiation dose distribution based on the position of the tumour during treatment to reduce unnecessary radiation exposure to healthy tissues. In comparison to existing clinical methods, the ML automated technique requires no human intervention during the planning process. This removes the reliance on human skill to drive the treatment process.

The work of Aly and his colleagues demonstrates that ML improves the quality of treatment by reducing the radiation dose delivered to healthy tissues, compared to current clinical adaptation techniques. These findings suggest that automated ML-based approaches could substantially improve clinical workflows and patient outcomes in adaptive radiotherapy.

AI-Driven Multimodal Computational Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Faisal Mahmood

Abstract:

This talk will discuss the use of AI in pathology and methods of training digitized images through the utilization of smaller patches to make the data efficient. Models have been developed to identify the most important images patches. In addition, the development of a breast cancer lymph node metastasis model will be discussed, showing how different imaging approaches were used. Models predicting histology have been developed for cancer applications.

Modality:

  • CT
  • Optical/microscopy
  • Histopathology
  • Genomics

Organ

  • Breast
  • Cardiac
  • Prostate

Disease

  • Cancer

AI-Based Precision Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Mattias Rantalainen

Abstract:

This talk will discuss precision medicine and its use in the development of AI in histopathology for the development of outcome prediction and treatment response prediction models. The development of a risk stratification model for breast cancer histological assessment will be discussed and the development of a medical device.

Modality:

  • Optical/microscopy
  • Gene expression
  • Histopathology

Organ

  • Breast
  • Lung
  • Prostate
  • Skin
  • Colorectal

Disease

  • Cancer

Other

  • Commercialization

Machine learning to predict disease progression

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Predicting Disease Progression

Past Event Page

Presenter:

Sun-In Lee

Abstract:

This talk will describe the use of text and phenotypic data in medical records for predicting patients’ clinical course, and potential uses for medical management. The presentation will be followed by questions and answers about the talk, and discussion focused on lessons learned, opportunities to collaborate, and sharing of data.

Modality:

  • Gene expression

Organ

  • Brain
  • Hematopathology
  • Lung
  • Skin

Disease

  • Cancer
  • Kidney and Liver Disease
  • Neurological Disease
  • Stroke and Cardiovascular
  • COVID

Other

  • Therapeutics
  • Personalized medicine

Artificial intelligence augmented ultrasound detection of hip dysplasia

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2023-04-21Jaremko

Open Forum on AI and Precision Health Initiatives at the University of Alberta

Past Event Page

Presenter:

Jacob Jaremko

Abstract:

These talks will present new efforts underway at the University of Alberta in Edmonton AB to develop and translate AI technology into clinical practice. The College of Health Sciences at UAlberta is leading a new initiative in AI for precision health. These efforts are conducted in collaboration with the Alberta Machine Intelligence Institute (Amii), one of three national AI institutes supported through the Pan-Canadian AI Strategy. Amii is located in Edmonton and supports 40 professors conducting basic and applied AI research. Precision Health is the largest application area within Amii.

The UAlberta College of Health Sciences also collaborates extensively with Alberta Health Services (AHS), Canada’s first and largest province-wide, fully integrated health system. AHS provides health services to over 4.4 million people. For the past 20 years they have developed and maintained a broadly integrated data warehouse to collect and protect health data everywhere they provide services. This allows construction of large population-level datasets for precision health research.

These presentations will discuss some of the projects currently underway, and look ahead at new efforts to expand AI and Precision Health research at UAlberta and beyond.

Modality:

  • US

Organ:

  • Musculoskeletal

Disease

  • Musculoskeletal Disease

Other

  • Commercialization
Download Presentation PDF

Sensitivity of convolutional neural networks to common imaging parameters, perturbations and artifacts in MRI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

lyndon_malmic_2021

Open Forum on Alzheimer’s and Small Vessel Disease Imaging

Past Event Page

Presenter:

Lyndon Boone

Abstract:

A number of studies have shown that deep learning methods are capable of achieving near-human-level performance on neuroimaging segmentation tasks. With that said, most of the results quoted in the literature supporting this statement are in the context of test sets drawn from the same overarching dataset as the training data. Clinically-deployed models faced with out-of-distribution data (i.e. data that does not resemble the training set) may severely underperform relative to the standards set in the literature if they aren’t designed specifically with robustness to out-of-distribution data in mind. In this talk, Lyndon highlighted the sensitivity of modern CNN-based architectures to image corruptions, artifacts, and post-processing transforms commonly found in MRI. He then presented a methodology for benchmarking different architectures and models on the basis of robustness to out-of-distribution data, inspired by similar work in the computer vision literature.

Modality:

  • MR

Organ:

  • Brain
Download Presentation PDF

Deep Learning for Automated Segmentation of Left Ventricle Myocardium and Myocardial Scar From 3-D MR Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-06-18Zabihollahy

Open Forum on Machine Learning in Cardiac Imaging

Past Event Page

Presenter:

Fatemeh Zabihollahy

Abstract:

Deep learning has demonstrated promise for various cardiac imaging applications. However, the performance is usually degraded when the models are trained with small and under-annotated training datasets and tested on previously unseen domains, limiting the potential for broad clinical use. In this talk, Fumin presented his recent work on combining deep learning and machine learning models for cardiac MRI segmentation, where smaller datasets and fewer annotations are required for algorithm training. He also provided examples of integrating the segmentation tools for myocardial infarct heterogeneity quantification in contrast enhancement MRI in the context of MRI-guided cardiac arrhythmia treatment.

Modality:

  • MR

Organ:

  • Cardiac

Disease

  • Heart Disease
  • Stroke and Cardiovascular
Download Presentation PDF

Deep learning with uncertainty quantification in MRI-guided radiation therapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Hemsley

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Matt Hemsley

Abstract:

Deep learning methods are able to match or surpass the performance of ​​classical methods as well as human experts in a growing number of medical imaging related tasks. However, most deep learning methods are unable to quantify uncertainty, hindering clinical translation. In this talk, Matt presented an overview of techniques used to model uncertainty on network output, demonstrated the connection between uncertainty quantification and reproducibility, and presented examples where uncertainty estimates were obtained from networks that performed clinical tasks related to real-time MRI-guided adaptive radiation therapy.

Modality:

  • CT
  • MR

Organ:

  • Brain
  • Head and Neck

Disease

  • Cancer
Download Presentation PDF

Brain-CODE

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Brain-CODE2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenters:

Tom Mikkelsen, Kirk Nylen, Brendan Behan

Abstract:

The Ontario Brain Institute (OBI) provided an overview of our Brain-CODE platform, its data holdings, and the opportunity to access and analyze highly standardized data on people living with various brain disorders – ranging from pediatric neurodevelopmental disorders through to neurodegenerative diseases of aging. We hoped to engage high-end data scientists like you in a discussion around potential collaboration opportunities with our large network of clinical experts/researchers.\

Modality:

  • MR
  • Gene Expression

Organ:

  • Brain

Disease:

  • Neurological Disease
  • Phantom
  • Psychological
Download Presentation PDF

Deep learning for automatic prostate segmentation in 3D ultrasound

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Orlando2021-02-19

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Nathan Orlando

Abstract:

Nathan discussed the development and validation of a generalizable and efficient deep learning method for automatic prostate segmentation in 3D ultrasound, compared performance to state of the art algorithms and explored the impact of dataset size and diversity on segmentation performance.

Modality:

  • US

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

MRI in Prostate Cancer: Opportunities for AI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

OICR-Prostate-ML-Haider

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Masoom Haider

Abstract:

MRI has established itself as an adjunctive test in detecting occult prostate cancer. New guidelines will recommend MRI for patients with elevated PSA and no prior biopsy resulting in a marked increase in MRI in the province. There are several use cases that can benefit from the application of machine learning to improve diagnostic performance and outcomes, which was reviewed.

Modality:

  • MRI
  • Gene Expression

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

Machine learning for prognostic modelling in head and neck cancer using multimodal data

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Kazmierski2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Michal Kazmierski

Abstract:

Michal discussed the results of a machine learning challenge for head and neck cancer (HNC) survival prediction with the aim of 1) developing an accurate prognostic model for HNC survival using clinical, demographic and routinely collected CT imaging data and 2) evaluating the true added value of CT radiomics compared to other prognostic factors.

Modality:

  • CT

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Radiomics and Machine Learning for Oropharyngeal Cancer

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Mattonen2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Sarah Mattonen

Abstract:

Sarah discussed preliminary work investigating radiomics and machine learning models to predict outcomes and toxicities in patients with oropharyngeal cancer treated with chemoradiotherapy. Sarah also discussed efforts to validate existing models, and described some challenges and opportunities for radiomics research in head and neck cancer.

Modality:

  • CT
  • MR

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Disease

  • All
  • - COVID
  • - Cancer
  • - Heart Disease
  • - Kidney and Liver Disease
  • - Mental Health
  • - Musculoskeletal Disease
  • - Neurological Disease
  • - Phantom
  • - Psychological
  • - Stroke and Cardiovascular

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

Past Event Page

Presenter:

Sangwook Kim, PhD Student
University of Toronto

Abstract:

Deep learning-based automated treatment planning cansignificantly enhance the efficiency and accuracy of radiotherapy. However, current planning approaches often depend on manually generated contours, limiting their efficiency. To address this, Sangwook and his colleagues implemented a multi-task learning framework that integrates automated contouring with voxel-based dose prediction, reducing the need for manual input and streamlining the planning process. Using two datasets – an in-house prostate cancer dataset and the publicly available OpenKBP head and neck cancer dataset – they developed a system that performs simultaneous segmentation and dose prediction.

Compared to conventional methods, the new framework improved the average absolute difference in dose volume histogram metrics by 2.90% for prostate cancer and 13.12% for head and neck cancer. Additionally, it enhanced dose prediction performance while maintaining high segmentation accuracy, with dice score coefficients of 0.824 for prostate and 0.716 for head and neck, compared to baseline scores of 0.818 and 0.674, respectively. These improvements can lead to more precise treatment plans and better patient outcomes.

The multi-task learning framework is not only generalizable to other anatomical sites and conditions but also holds promise for significantly reducing clinical workload and enhancing radiotherapy efficiency. By integrating automated contouring and dose prediction, this newapproach minimizes the need for sequential steps in the planning process, potentially allowing clinics to handle higher patient volumes with greater consistency and accuracy. This work illustrates the potential of AI to enable fully automated, efficient radiotherapy planning, supporting broader adoption of AI-driven tools in clinical practice.

Applications of Automated Treatment Planning in Adaptive Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Applications of Automated Treatment Planning in Adaptive Radiotherapy

Past Event Page

Presenter:

Aly Khalifa, PhD Student
University of Toronto

Abstract:

Technological advancements in radiotherapy have significantly enhanced treatment precision and patient safety, but they have also introduced a greater burden of manual, time-intensive tasks for clinicians to manage these innovations effectively. Machine Learning (ML) offers a promising solution to streamline these processes and optimize treatment outcomes.

The work of Aly and his colleagues explores applications of automated treatment planning in adaptive radiotherapy procedures. ML is used to predict an ideal radiation dose distribution based on the position of the tumour during treatment to reduce unnecessary radiation exposure to healthy tissues. In comparison to existing clinical methods, the ML automated technique requires no human intervention during the planning process. This removes the reliance on human skill to drive the treatment process.

The work of Aly and his colleagues demonstrates that ML improves the quality of treatment by reducing the radiation dose delivered to healthy tissues, compared to current clinical adaptation techniques. These findings suggest that automated ML-based approaches could substantially improve clinical workflows and patient outcomes in adaptive radiotherapy.

Chatbots as Mental Healthcare Proxies: Possibilities and Limitations

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Chatbots as Mental Healthcare Proxies: Possibilities and Limitations

Past Event Page

Presenter:

Catherine Stinson, PhD
Assistant Professor, Queen’s University

Abstract:

There is a great need for more affordable, accessible mental health treatment options, especially labour-intensive talk therapy. Given the impressive abilities of a new generation of chatbots like ChatGPT to mimic human conversational skills, there is hope that they might prove useful as proxies for human psychotherapists. In particular, there is hope that for communities facing barriers to mental health care, chatbots might fill the gap. We look in detail at the current generation of chatbots to understand what they do well, and what their limitations are, with support from empirical work in natural language processing. Where these tools perform best is on formulaic language tasks, in domains well covered in training corpora, using standard English. Unfortunately among the under-served communities are migrant and minority groups who may not communicate in standard English, and are not well represented in training corpora. For some psychotherapeutic interactions, particularly formulaic ones, the capacities of chatbots may be a good match. However, for for interactions where an empathetic relationship is essential, the current generation of therapy chatbots are potentially dangerous. While there is some room for chatbots to act as proxies for human psychotherapists, we should not overestimate their abilities.

Disease:

  • Mental Health

Other:

  • Health Equity

AI-Driven Multimodal Computational Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Faisal Mahmood

Abstract:

This talk will discuss the use of AI in pathology and methods of training digitized images through the utilization of smaller patches to make the data efficient. Models have been developed to identify the most important images patches. In addition, the development of a breast cancer lymph node metastasis model will be discussed, showing how different imaging approaches were used. Models predicting histology have been developed for cancer applications.

Modality:

  • CT
  • Optical/microscopy
  • Histopathology
  • Genomics

Organ

  • Breast
  • Cardiac
  • Prostate

Disease

  • Cancer

AI-Based Precision Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Mattias Rantalainen

Abstract:

This talk will discuss precision medicine and its use in the development of AI in histopathology for the development of outcome prediction and treatment response prediction models. The development of a risk stratification model for breast cancer histological assessment will be discussed and the development of a medical device.

Modality:

  • Optical/microscopy
  • Gene expression
  • Histopathology

Organ

  • Breast
  • Lung
  • Prostate
  • Skin
  • Colorectal

Disease

  • Cancer

Other

  • Commercialization

Machine learning to predict disease progression

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Predicting Disease Progression

Past Event Page

Presenter:

Sun-In Lee

Abstract:

This talk will describe the use of text and phenotypic data in medical records for predicting patients’ clinical course, and potential uses for medical management. The presentation will be followed by questions and answers about the talk, and discussion focused on lessons learned, opportunities to collaborate, and sharing of data.

Modality:

  • Gene expression

Organ

  • Brain
  • Hematopathology
  • Lung
  • Skin

Disease

  • Cancer
  • Kidney and Liver Disease
  • Neurological Disease
  • Stroke and Cardiovascular
  • COVID

Other

  • Therapeutics
  • Personalized medicine

Segment Anything in Medical Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

202312-MedSAM

Open Forum on Enforcing Geometry in Machine Learning for Computational Neuroimaging

Past Event Page

Presenter:

Jun Ma

Abstract:

Medical imaging plays an indispensable role in clinical practice. Accurate and efficient medical image segmentation provides a means of delineating regions of interest and quantifying various clinical metrics. However, building customized segmentation models for each medical imaging task can be a daunting and time-consuming process, limiting the widespread adoption in clinical practice. In this talk, Jun will introduce MedSAM, a segmentation foundation model that enables universal segmentation across a wide range of medical imaging tasks and modalities. MedSAM achieved remarkable improvements in 30 segmentation tasks, surpassing the existing segmentation foundation model by a large margin. MedSAM also demonstrated zero-shot and few-shot capabilities to segment unseen tumor types and adapt to new imaging modalities with minimal effort. The results validate the versatility of MedSAM compared to existing customized segmentation models, emphasizing its potential to transform medical image segmentation and enhance clinical practice.

Disease

  • Cancer

Other

  • Resource Sharing
  • Reproducibility
Download Presentation PDF

Artificial intelligence augmented ultrasound detection of hip dysplasia

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2023-04-21Jaremko

Open Forum on AI and Precision Health Initiatives at the University of Alberta

Past Event Page

Presenter:

Jacob Jaremko

Abstract:

These talks will present new efforts underway at the University of Alberta in Edmonton AB to develop and translate AI technology into clinical practice. The College of Health Sciences at UAlberta is leading a new initiative in AI for precision health. These efforts are conducted in collaboration with the Alberta Machine Intelligence Institute (Amii), one of three national AI institutes supported through the Pan-Canadian AI Strategy. Amii is located in Edmonton and supports 40 professors conducting basic and applied AI research. Precision Health is the largest application area within Amii.

The UAlberta College of Health Sciences also collaborates extensively with Alberta Health Services (AHS), Canada’s first and largest province-wide, fully integrated health system. AHS provides health services to over 4.4 million people. For the past 20 years they have developed and maintained a broadly integrated data warehouse to collect and protect health data everywhere they provide services. This allows construction of large population-level datasets for precision health research.

These presentations will discuss some of the projects currently underway, and look ahead at new efforts to expand AI and Precision Health research at UAlberta and beyond.

Modality:

  • US

Organ:

  • Musculoskeletal

Disease

  • Musculoskeletal Disease

Other

  • Commercialization
Download Presentation PDF

Deep Learning for Automated Segmentation of Left Ventricle Myocardium and Myocardial Scar From 3-D MR Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-06-18Zabihollahy

Open Forum on Machine Learning in Cardiac Imaging

Past Event Page

Presenter:

Fatemeh Zabihollahy

Abstract:

Deep learning has demonstrated promise for various cardiac imaging applications. However, the performance is usually degraded when the models are trained with small and under-annotated training datasets and tested on previously unseen domains, limiting the potential for broad clinical use. In this talk, Fumin presented his recent work on combining deep learning and machine learning models for cardiac MRI segmentation, where smaller datasets and fewer annotations are required for algorithm training. He also provided examples of integrating the segmentation tools for myocardial infarct heterogeneity quantification in contrast enhancement MRI in the context of MRI-guided cardiac arrhythmia treatment.

Modality:

  • MR

Organ:

  • Cardiac

Disease

  • Heart Disease
  • Stroke and Cardiovascular
Download Presentation PDF

Deep learning with uncertainty quantification in MRI-guided radiation therapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Hemsley

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Matt Hemsley

Abstract:

Deep learning methods are able to match or surpass the performance of ​​classical methods as well as human experts in a growing number of medical imaging related tasks. However, most deep learning methods are unable to quantify uncertainty, hindering clinical translation. In this talk, Matt presented an overview of techniques used to model uncertainty on network output, demonstrated the connection between uncertainty quantification and reproducibility, and presented examples where uncertainty estimates were obtained from networks that performed clinical tasks related to real-time MRI-guided adaptive radiation therapy.

Modality:

  • CT
  • MR

Organ:

  • Brain
  • Head and Neck

Disease

  • Cancer
Download Presentation PDF

Brain-CODE

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Brain-CODE2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenters:

Tom Mikkelsen, Kirk Nylen, Brendan Behan

Abstract:

The Ontario Brain Institute (OBI) provided an overview of our Brain-CODE platform, its data holdings, and the opportunity to access and analyze highly standardized data on people living with various brain disorders – ranging from pediatric neurodevelopmental disorders through to neurodegenerative diseases of aging. We hoped to engage high-end data scientists like you in a discussion around potential collaboration opportunities with our large network of clinical experts/researchers.\

Modality:

  • MR
  • Gene Expression

Organ:

  • Brain

Disease:

  • Neurological Disease
  • Phantom
  • Psychological
Download Presentation PDF

QIPCM – Advanced Imaging Core Lab

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

QIPCM2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenter:

Julia Publicover

Abstract:

The QIPCM imaging core lab was conceived for high-quality, auditable and secure image management for clinical trials. More and more, users are looking to leverage the platform for large volume image analysis and processing, including radiomics and ML applications. This talk gave an overview of the QIPCM data-sharing platform including tools, services and quality management program that ensure high-quality image collection, analysis and regulatory compliance. I also discussed current use-cases in radiomics and ML and future aims.

Disease:

  • Cancer

Other:

  • Data source/database
Download Presentation PDF

Deep learning for automatic prostate segmentation in 3D ultrasound

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Orlando2021-02-19

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Nathan Orlando

Abstract:

Nathan discussed the development and validation of a generalizable and efficient deep learning method for automatic prostate segmentation in 3D ultrasound, compared performance to state of the art algorithms and explored the impact of dataset size and diversity on segmentation performance.

Modality:

  • US

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

MRI in Prostate Cancer: Opportunities for AI

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

OICR-Prostate-ML-Haider

Open Forum on Machine Learning in Prostate Cancer

Past Event Page

Presenter:

Masoom Haider

Abstract:

MRI has established itself as an adjunctive test in detecting occult prostate cancer. New guidelines will recommend MRI for patients with elevated PSA and no prior biopsy resulting in a marked increase in MRI in the province. There are several use cases that can benefit from the application of machine learning to improve diagnostic performance and outcomes, which was reviewed.

Modality:

  • MRI
  • Gene Expression

Organ:

  • Prostate

Disease:

  • Cancer
Download Presentation PDF

Machine learning for prognostic modelling in head and neck cancer using multimodal data

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Kazmierski2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Michal Kazmierski

Abstract:

Michal discussed the results of a machine learning challenge for head and neck cancer (HNC) survival prediction with the aim of 1) developing an accurate prognostic model for HNC survival using clinical, demographic and routinely collected CT imaging data and 2) evaluating the true added value of CT radiomics compared to other prognostic factors.

Modality:

  • CT

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Radiomics and Machine Learning for Oropharyngeal Cancer

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Mattonen2021-01-21

Open Forum on Machine Learning in Head and Neck Cancer

Past Event Page

Presenter:

Sarah Mattonen

Abstract:

Sarah discussed preliminary work investigating radiomics and machine learning models to predict outcomes and toxicities in patients with oropharyngeal cancer treated with chemoradiotherapy. Sarah also discussed efforts to validate existing models, and described some challenges and opportunities for radiomics research in head and neck cancer.

Modality:

  • CT
  • MR

Organ:

  • Head and Neck

Disease:

  • Cancer
Download Presentation PDF

Other

  • All
  • - Commercialization
  • - Data Source / Database
  • - Health Equity
  • - Personalized Medicine
  • - Reproducibility
  • - Resource Sharing
  • - Therapeutics

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

Past Event Page

Presenter:

Sangwook Kim, PhD Student
University of Toronto

Abstract:

Deep learning-based automated treatment planning cansignificantly enhance the efficiency and accuracy of radiotherapy. However, current planning approaches often depend on manually generated contours, limiting their efficiency. To address this, Sangwook and his colleagues implemented a multi-task learning framework that integrates automated contouring with voxel-based dose prediction, reducing the need for manual input and streamlining the planning process. Using two datasets – an in-house prostate cancer dataset and the publicly available OpenKBP head and neck cancer dataset – they developed a system that performs simultaneous segmentation and dose prediction.

Compared to conventional methods, the new framework improved the average absolute difference in dose volume histogram metrics by 2.90% for prostate cancer and 13.12% for head and neck cancer. Additionally, it enhanced dose prediction performance while maintaining high segmentation accuracy, with dice score coefficients of 0.824 for prostate and 0.716 for head and neck, compared to baseline scores of 0.818 and 0.674, respectively. These improvements can lead to more precise treatment plans and better patient outcomes.

The multi-task learning framework is not only generalizable to other anatomical sites and conditions but also holds promise for significantly reducing clinical workload and enhancing radiotherapy efficiency. By integrating automated contouring and dose prediction, this newapproach minimizes the need for sequential steps in the planning process, potentially allowing clinics to handle higher patient volumes with greater consistency and accuracy. This work illustrates the potential of AI to enable fully automated, efficient radiotherapy planning, supporting broader adoption of AI-driven tools in clinical practice.

Applications of Automated Treatment Planning in Adaptive Radiotherapy

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Applications of Automated Treatment Planning in Adaptive Radiotherapy

Past Event Page

Presenter:

Aly Khalifa, PhD Student
University of Toronto

Abstract:

Technological advancements in radiotherapy have significantly enhanced treatment precision and patient safety, but they have also introduced a greater burden of manual, time-intensive tasks for clinicians to manage these innovations effectively. Machine Learning (ML) offers a promising solution to streamline these processes and optimize treatment outcomes.

The work of Aly and his colleagues explores applications of automated treatment planning in adaptive radiotherapy procedures. ML is used to predict an ideal radiation dose distribution based on the position of the tumour during treatment to reduce unnecessary radiation exposure to healthy tissues. In comparison to existing clinical methods, the ML automated technique requires no human intervention during the planning process. This removes the reliance on human skill to drive the treatment process.

The work of Aly and his colleagues demonstrates that ML improves the quality of treatment by reducing the radiation dose delivered to healthy tissues, compared to current clinical adaptation techniques. These findings suggest that automated ML-based approaches could substantially improve clinical workflows and patient outcomes in adaptive radiotherapy.

Chatbots as Mental Healthcare Proxies: Possibilities and Limitations

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Chatbots as Mental Healthcare Proxies: Possibilities and Limitations

Past Event Page

Presenter:

Catherine Stinson, PhD
Assistant Professor, Queen’s University

Abstract:

There is a great need for more affordable, accessible mental health treatment options, especially labour-intensive talk therapy. Given the impressive abilities of a new generation of chatbots like ChatGPT to mimic human conversational skills, there is hope that they might prove useful as proxies for human psychotherapists. In particular, there is hope that for communities facing barriers to mental health care, chatbots might fill the gap. We look in detail at the current generation of chatbots to understand what they do well, and what their limitations are, with support from empirical work in natural language processing. Where these tools perform best is on formulaic language tasks, in domains well covered in training corpora, using standard English. Unfortunately among the under-served communities are migrant and minority groups who may not communicate in standard English, and are not well represented in training corpora. For some psychotherapeutic interactions, particularly formulaic ones, the capacities of chatbots may be a good match. However, for for interactions where an empathetic relationship is essential, the current generation of therapy chatbots are potentially dangerous. While there is some room for chatbots to act as proxies for human psychotherapists, we should not overestimate their abilities.

Disease:

  • Mental Health

Other:

  • Health Equity

Healthcare Horizons: AI to Navigate the Evolving Landscape

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Healthcare Horizons: AI to Navigate the Evolving Landscape

Past Event Page

Presenter:

Elham Dolatabadi, PhD
Assistant Professor, York University

Abstract:

As healthcare evolves, data expands, and learning algorithms become increasingly sophisticated, the integration of artificial intelligence (AI) emerges as a transformative force, reshaping the healthcare landscape. In this talk, Elham will explore how AI goes beyond being a mere tool, driving revolutionary changes across the field. From continuous monitoring for specific single tasks to complex multimodal decision-making and conversational AI, the potential applications are vast. However, this journey is not without its challenges. Key hurdles include robust evaluation processes, mitigating biases, and cultivating the right mindset and knowledge to ensure that AI innovations are both effective and equitable. These challenges, and the strategies to overcome them, will be the focus of this talk.

Other:

  • Health Equity

AI-Based Precision Pathology

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Computational Pathology and Cancer Prediction

Past Event Page

Presenter:

Mattias Rantalainen

Abstract:

This talk will discuss precision medicine and its use in the development of AI in histopathology for the development of outcome prediction and treatment response prediction models. The development of a risk stratification model for breast cancer histological assessment will be discussed and the development of a medical device.

Modality:

  • Optical/microscopy
  • Gene expression
  • Histopathology

Organ

  • Breast
  • Lung
  • Prostate
  • Skin
  • Colorectal

Disease

  • Cancer

Other

  • Commercialization

Machine learning to predict disease progression

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Open Forum on Predicting Disease Progression

Past Event Page

Presenter:

Sun-In Lee

Abstract:

This talk will describe the use of text and phenotypic data in medical records for predicting patients’ clinical course, and potential uses for medical management. The presentation will be followed by questions and answers about the talk, and discussion focused on lessons learned, opportunities to collaborate, and sharing of data.

Modality:

  • Gene expression

Organ

  • Brain
  • Hematopathology
  • Lung
  • Skin

Disease

  • Cancer
  • Kidney and Liver Disease
  • Neurological Disease
  • Stroke and Cardiovascular
  • COVID

Other

  • Therapeutics
  • Personalized medicine

Segment Anything in Medical Images

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

202312-MedSAM

Open Forum on Enforcing Geometry in Machine Learning for Computational Neuroimaging

Past Event Page

Presenter:

Jun Ma

Abstract:

Medical imaging plays an indispensable role in clinical practice. Accurate and efficient medical image segmentation provides a means of delineating regions of interest and quantifying various clinical metrics. However, building customized segmentation models for each medical imaging task can be a daunting and time-consuming process, limiting the widespread adoption in clinical practice. In this talk, Jun will introduce MedSAM, a segmentation foundation model that enables universal segmentation across a wide range of medical imaging tasks and modalities. MedSAM achieved remarkable improvements in 30 segmentation tasks, surpassing the existing segmentation foundation model by a large margin. MedSAM also demonstrated zero-shot and few-shot capabilities to segment unseen tumor types and adapt to new imaging modalities with minimal effort. The results validate the versatility of MedSAM compared to existing customized segmentation models, emphasizing its potential to transform medical image segmentation and enhance clinical practice.

Disease

  • Cancer

Other

  • Resource Sharing
  • Reproducibility
Download Presentation PDF

Artificial intelligence augmented ultrasound detection of hip dysplasia

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2023-04-21Jaremko

Open Forum on AI and Precision Health Initiatives at the University of Alberta

Past Event Page

Presenter:

Jacob Jaremko

Abstract:

These talks will present new efforts underway at the University of Alberta in Edmonton AB to develop and translate AI technology into clinical practice. The College of Health Sciences at UAlberta is leading a new initiative in AI for precision health. These efforts are conducted in collaboration with the Alberta Machine Intelligence Institute (Amii), one of three national AI institutes supported through the Pan-Canadian AI Strategy. Amii is located in Edmonton and supports 40 professors conducting basic and applied AI research. Precision Health is the largest application area within Amii.

The UAlberta College of Health Sciences also collaborates extensively with Alberta Health Services (AHS), Canada’s first and largest province-wide, fully integrated health system. AHS provides health services to over 4.4 million people. For the past 20 years they have developed and maintained a broadly integrated data warehouse to collect and protect health data everywhere they provide services. This allows construction of large population-level datasets for precision health research.

These presentations will discuss some of the projects currently underway, and look ahead at new efforts to expand AI and Precision Health research at UAlberta and beyond.

Modality:

  • US

Organ:

  • Musculoskeletal

Disease

  • Musculoskeletal Disease

Other

  • Commercialization
Download Presentation PDF

Design, conduct, and reporting of radiomic analyses: Let’s not reinvent the wheel

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Moskowitz

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Chaya Moskowitz

Abstract:

The number of published papers reporting on radiomic analyses has been growing exponentially. Very few of these paper produce results that are translated into clinical practice at least in part because of methodological flaws in the work. Although complex methods for producing radiomic signatures are increasingly available and user-friendly and progress has been made in radiomic biomarker taxonomy and standardization, fundamental elements of study design, rigorous statistical analysis, and quality of reporting methods and results are frequently overlooked. In this talk, Chaya highlighted common pitfalls encountered in radiomic studies that could be avoided by knowledge of existing methods and adherence to existing reporting standards.

Other:

  • Reproducibility
Download Presentation PDF

Improving Peproducibility in Machine Learning Research

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

2021-05-21Haibe-Kains

Open Forum on Reproducibility and Good Practices in Machine Learning Research

Past Event Page

Presenter:

Benjamin Haibe-Kains

Abstract:

While biostatistics and machine learning are essential to analyze biomedical data, researchers are facing multiple challenges around research reproducibility and transparency. It is essential for independent researchers to be able to scrutinize and reproduce the results of a study using its materials, and build upon them in future studies.

Computational reproducibility is achievable when the data can easily be shared and the required computational resources are relatively common. However, the complexity of current algorithms and their implementation, the need for specific computer hardware and the use of sensitive biomedical data represent major obstacles in healthy-related research. In this talk, Benjamin described the various aspects of a typical biomedical study that are necessary for reproducibility and the platforms that exist for sharing these materials with the scientific community.

Other:

  • Resource Sharing
  • Reproducibility
Download Presentation PDF

Introduction to the Ontario Health Data Platform

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

OHDP2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenter:

Amber Simpson

Abstract:

The Ontario Health Data Platform (OHDP) is a secure, federated high-performance computing environment built for linkage and analysis of large health data sets. The platform was built to aid Ontario in their fight against COVID-19. This talk described the platform, available data, and the potential for future data such as genomic and imaging data repositories.

Other:

  • Data source/database
Download Presentation PDF

Brain-CODE

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

Brain-CODE2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenters:

Tom Mikkelsen, Kirk Nylen, Brendan Behan

Abstract:

The Ontario Brain Institute (OBI) provided an overview of our Brain-CODE platform, its data holdings, and the opportunity to access and analyze highly standardized data on people living with various brain disorders – ranging from pediatric neurodevelopmental disorders through to neurodegenerative diseases of aging. We hoped to engage high-end data scientists like you in a discussion around potential collaboration opportunities with our large network of clinical experts/researchers.\

Modality:

  • MR
  • Gene Expression

Organ:

  • Brain

Disease:

  • Neurological Disease
  • Phantom
  • Psychological
Download Presentation PDF

QIPCM – Advanced Imaging Core Lab

 150 150 MaLMIC - Machine Learning in Medical Imaging Consortium

QIPCM2021-03-19

Open Forum on Machine Learning in Health Data Integration Platforms

Past Event Page

Presenter:

Julia Publicover

Abstract:

The QIPCM imaging core lab was conceived for high-quality, auditable and secure image management for clinical trials. More and more, users are looking to leverage the platform for large volume image analysis and processing, including radiomics and ML applications. This talk gave an overview of the QIPCM data-sharing platform including tools, services and quality management program that ensure high-quality image collection, analysis and regulatory compliance. I also discussed current use-cases in radiomics and ML and future aims.

Disease:

  • Cancer

Other:

  • Data source/database
Download Presentation PDF
Privacy Preferences

When you visit our website, it may store information through your browser from specific services, usually in the form of cookies. Here you can change your Privacy preferences. It is worth noting that blocking some types of cookies may impact your experience on our website and the services we are able to offer.

Click to enable/disable Google Analytics tracking code.
Click to enable/disable Google Fonts.
Click to enable/disable Google Maps.
Click to enable/disable video embeds.
Our website uses cookies, mainly from 3rd party services. Define your Privacy Preferences and/or agree to our use of cookies.