Scope

Transparent model reporting is research about one of the most critical topics about Machine Learning and Health: transparency.

When a researcher, laboratory, or company discovers a new technology or method to evaluate one topic, there is a standard behavior not to publish data or the codes, preserving as a secret object to the owner.

AI Audit doesn’t have the objective to publish somebody’s codes or data but creates trustworthiness between the users and providers. A study on the practice of transparent model reporting.

If you have a use case, feel free to submit us and receive an AI Audit analysis with total safety and expand the trustworthiness of your company.

Send an e-mail to [tmr.daisam.fgai4h@aiaudit.org].

Aims

Create a discussion about transparency and trustworthiness between the users and developers
Promotes more transparency between companies and users.
Paper and regulatory documents to manufacturers, academics, health users notified regulatory bodies.
Scientific publications, seminars & webinars.

Outputs

ITU/WHO

Data and artificial intelligence assessment methods (DAISAM) Audit Reporting Template

Verks, Boris, and Oala, Luis

In Proceedings of the ITU/WHO Focus Group on Artificial Intelligence for Health (FG-AI4H) - Meeting J 2020

Abs HTML PDF

Standardized templates to report results for the assessment processes developed by WG-DAISAM. In this version the, template comprises three elements: Data Specification Sheet, ML Model Specification Sheet and ML Model Summary Findings.
NeurIPS Spotlight
Top 10%

ML4H Auditing: From Paper to Practice

Oala, Luis, Fehr, Jana, Gilli, Luca, Balachandran, Pradeep, Leite, Alixandro Werneck, Calderon-Ramirez, Saul, Li, Danny Xie, Nobis, Gabriel, Alvarado, Erick Alejandro Munoz, Jaramillo-Gutierrez, Giovanna, Matek, Christian, Shroff, Arun, Kherif, Ferath, Sanguinetti, Bruno, and Wiegand, Thomas

In Proceedings of the Machine Learning for Health NeurIPS Workshop 2020

Abs HTML PDF Slides Video

Healthcare systems are currently adapting to digital technologies, producing large quantities of novel data. Based on these data, machine-learning algorithms have been developed to support practitioners in labor-intensive workflows such as diagnosis, prognosis, triage or treatment of disease. However, their translation into medical practice is often hampered by a lack of careful evaluation in different settings. Efforts have started worldwide to establish guidelines for evaluating machine learning for health (ML4H) tools, highlighting the necessity to evaluate models for bias, interpretability, robustness, and possible failure modes. However, testing and adopting these guidelines in practice remains an open challenge. In this work, we target the paper-to-practice gap by applying an ML4H audit framework proposed by the ITU/WHO Focus Group on Artificial Intelligence for Health (FG-AI4H) to three use cases: diagnostic prediction of diabetic retinopathy, diagnostic prediction of Alzheimer’s disease, and cytomorphologic classification for leukemia diagnostics. The assessment comprises dimensions such as bias, interpretability, and robustness. Our results highlight the importance of fine-grained and caseadapted quality assessment, provide support for incorporating proposed quality assessment considerations of ML4H during the entire development life cycle, and suggest improvements for future ML4H reference evaluation frameworks.

Collaboration resources

You are welcome to inquire about the work stream and opporunities for collaboration directly with the work stream team.

General contact Jana Fehr, jana@aiaudit.org

Meetings

Regular meetings for this work stream take place at the below coordinates.

Meetings Every Thursday, 4:00 PM, CET
Zoom room Click here to join meeting

Communication

You can subsbscribe to the work stream mailing list to receive updates and join the asynchronous group chat.

Group chat https://discord.gg/53zWbgxt
Mailing list

Tools

We use different tools in our remote work. They include shared documents, github projects for code as well as task tracking and a collaborative whiteboard for ideation. You can request access via the below links.

Shared drive
Github project
Collaborative whiteboard

You can find more information about the way we usually carry out our work remotely in teams here.

Milestones

Important reference material

This is a list of related work and resources relevant for this work stream. It comprises resources the work stream contributors consider good practice.

ITU/WHO

Data and artificial intelligence assessment methods (DAISAM) Audit Reporting Template

Verks, Boris, and Oala, Luis

In Proceedings of the ITU/WHO Focus Group on Artificial Intelligence for Health (FG-AI4H) - Meeting J 2020

Abs HTML PDF

Standardized templates to report results for the assessment processes developed by WG-DAISAM. In this version the, template comprises three elements: Data Specification Sheet, ML Model Specification Sheet and ML Model Summary Findings.
NeurIPS Spotlight
Top 10%

ML4H Auditing: From Paper to Practice

Oala, Luis, Fehr, Jana, Gilli, Luca, Balachandran, Pradeep, Leite, Alixandro Werneck, Calderon-Ramirez, Saul, Li, Danny Xie, Nobis, Gabriel, Alvarado, Erick Alejandro Munoz, Jaramillo-Gutierrez, Giovanna, Matek, Christian, Shroff, Arun, Kherif, Ferath, Sanguinetti, Bruno, and Wiegand, Thomas

In Proceedings of the Machine Learning for Health NeurIPS Workshop 2020

Abs HTML PDF Slides Video

Healthcare systems are currently adapting to digital technologies, producing large quantities of novel data. Based on these data, machine-learning algorithms have been developed to support practitioners in labor-intensive workflows such as diagnosis, prognosis, triage or treatment of disease. However, their translation into medical practice is often hampered by a lack of careful evaluation in different settings. Efforts have started worldwide to establish guidelines for evaluating machine learning for health (ML4H) tools, highlighting the necessity to evaluate models for bias, interpretability, robustness, and possible failure modes. However, testing and adopting these guidelines in practice remains an open challenge. In this work, we target the paper-to-practice gap by applying an ML4H audit framework proposed by the ITU/WHO Focus Group on Artificial Intelligence for Health (FG-AI4H) to three use cases: diagnostic prediction of diabetic retinopathy, diagnostic prediction of Alzheimer’s disease, and cytomorphologic classification for leukemia diagnostics. The assessment comprises dimensions such as bias, interpretability, and robustness. Our results highlight the importance of fine-grained and caseadapted quality assessment, provide support for incorporating proposed quality assessment considerations of ML4H during the entire development life cycle, and suggest improvements for future ML4H reference evaluation frameworks.

Transparent Model Reporting