Ai Exptnetwork Logo White
Moving to a world of predictive methodologies to protect and advance public health.

The AI Experts Focus Sessions cover frequently-requested topics to enhance knowledge and insights into AI.

Whether you are an AI Expert or not, you can still view the recorded focus session portion from the meetings.

August 4, 2023 

Featuring Lane Desborough, CEO, Nudge BG

Algorithms: apex predators of the software eating the world!

As algorithms (decision support, feedback control) eat their way into medical products, what can and should we learn from other industries which have been disrupted by algorithms?  

Cyberphysical systems: software is a harmless mental abstraction until it is instantiated in the physical world.

Data:  good experimental designs yield good experiments, which yield good data, which yield good models, which yield good predictions, which yield good algorithms, which yield good outcomes. 

Human factors: algorithms change the tasks and roles of humans; people go from being “in the loop” to being “on the loop” or “out of the loop”.  

Using an automated insulin delivery system case study, we’ll discuss why and how to tame these apex predators, ensuring safe and effective algorithm development and deployment.

June 30, 2023

Featuring Mike Salem, Associate Data Science, Gilead Sciences

As AI becomes more mainstream, companies and regulatory agencies are racing to understand AI best practices. Historically many of these models have been “black boxes,” but advances in areas like XAI have aimed to make many of these models’ outputs and inputs more understandable.

Watch Video

Download Presentation

April 21, 2023

Featuring Professor Jon Chun

Large Language Models, Tools, Automation and Future Applications in HealthTech

Large Language Models, Tools, Automation and Future Applications in HealthTech

The rapid rise of Large Language Models (LLMs) such as ChatGPT and GPT-4 in recent months has sparked an unprecedented surge in new research, models, frameworks, and industry applications. This talk will provide a concise overview of the historical development of LLMs and examine current models, frameworks, and applications. We will conclude with a forward-looking discussion of future applications, including LLM tool integration (e.g., LangChain), multimodal processing (e.g., JARVIS, HuggingGPT), automation frameworks (e.g., AutoGPT), and potential integration with Distributed Autonomous Organizations (DAOs)/Blockchain. Special emphasis will be placed on the unique challenges and opportunities associated with incorporating LLMs and related technologies within the healthcare and HealthTech startups.

Watch Video

Download Presentation

March 31, 2023 

Featuring Erez Kaminski, Owner & CEO, Ketryx

Best Practices When building AI-based Medical Devices at Scale

An inherent part of building data-driven software is the constant need for change, but existing lifecycle management tools and processes for medical systems create significant resistance to change management. These less flexible operations stand in contrast to the tech industry, where change management is a sought-after, routine process.

This talk will cover the best practices for building cloud-based, AI-driven medical software and the challenges of building at scale: the lack of software tooling, high attrition rates, and the difficulties of executing in a regulated, ever-changing environment. Along with best practices, real-world case studies of building validated software will be used to illustrate how to move fast, and break nothing.

Watch Video

Download Presentation

February 24, 2023

Featuring Akhilesh Mishra, Sr. Medical Devices Industry Manager, MathWorks

Explainable and Interpretable AI for medical devices certification

Watch Video

Download Presentation

December 3, 2021

Featuring Alexej Gossmann, Staff Fellow at the Division of Imaging, Diagnostics, and Software Reliability, FDA-CDRH

Issues in Reference Standard Determination for Performance Evaluation Studies of AI/ML-based Medical Devices

Evaluation of an AI/ML-based medical device is typically performed by comparing the system output to a reference standard. A poorly chosen reference standard that is a poor proxy to the targeted real-world clinical task may result in the measurement and reporting of misleading performance results for the AI/ML algorithm, which do not adequately represent the real-world performance or utility of this algorithm. This is particularly concerning when it applies only to specific socio-economically disadvantaged groups or racial minorities.

A more appropriately chosen reference standard can still be subject to several distinct types of bias that can lead to systematic flaws in algorithm training and testing. This includes, for instance, verification bias, measurement bias, and imperfect reference bias, which need to be assessed and accounted for when evaluating the performance of medical AI/ML algorithms. In addition, the reference standard can be partially missing, which can be accounted for with appropriate statistical analysis methods in some settings. This presentation provides a discussion of these and other reference standard issues in the assessment of AI/ML-based medical devices.

Watch Video

Download Presentation

October 8, 2021

Featuring Lars Bielak, a researcher at the University Medical Center in Freiburg, Germany

Incorporating AI Techniques for Medical Imaging

Tumor segmentation, or segmentation using CNNs in general, is a well-developed field in AI research. Unfortunately, the data acquisition and the data post-processing are often two completely separate topics, leaving computer scientists to refine their models based only on large, publicly available, and fixed datasets. In this session, Lars Bielak presents a strategy for prospective input data optimization for CNN head and neck tumor segmentation based on seven unique MRI channels.

Watch Video

Download Presentation

August 6, 2021

Featuring Toni Manzano, Co-Founder and CSO, Aizon

Pharma Manufacturing – When Complexity Cannot be Linearly Managed or Solved via a Singular AI Model

Toni Manzano presents two use cases, one of AI application in biopharmaceutical operations and one of biotech research in which Xavier AI was involved, to explain the hard reality behind amazing AI outputs.

June 25, 2021

Featuring Kirthi Devleker, Mathworks, and Akhilesh Mishra, Mathworks

Developing AI-Enabled Smart Medical Devices Using Model-Based Design

In this talk, Kirthi Devleker and Akhilesh Mishra provide an overview of how AI algorithms can be developed and deployed in Medical Devices using Model Based Design approach – an approach adopted by large medical device organizations globally to develop next generation medical devices.

Watch Video

April 16, 2021

Featuring Erez Kaminski, Researcher/Graduate Student, MIT

GxP AI Workflow: An NLP Case Study

In this second AI Learning Session covering the topics most frequently requested by our AI Experts Network, Erez Kaminski provides an overview of a systematic way to develop GxP artificial intelligence applications focused on code reusability and streamlined development operations. The talk includes a case study on the development of a Natural Language Processing application for adverse event detection, which is used to reduce processing costs and intake error rates.

February 5, 2021

Featuring Emily Luvison, Cybersecurity Compliance Lead for Digital Health Technologies, Genentech

Cybersecurity Compliance and AI

In this inaugural AI Learning Session, Emily Luvison (Genentech) will be providing an overview of cybersecurity considerations as they apply to AI.

Watch Video

The AFDO/RAPS Healthcare Products Collaborative is a joint venture established in 2022 between the Association of Food and Drug Officials and the Regulatory Affairs Professionals Society. Learn more.

Newsletter

Stay in the know about upcoming events and breaking news from the collaborative.

Subscribe

ContACT

Have questions about our events or how to get involved with one of our committees or workgroups?

Get in touch

© 2024 AFDO/RAPS Healthcare Products Collaborative