Project Scope |
The use cases of artificial intelligence and machine learning (AI/ML) in digital health technologies to improve healthcare through software. Understand the challenges, and identify the gaps. Connect different stakeholders, share knowledge, and advance in developing AI/ML in DHTs. |
| Project Statement |
|---|
DHTs are revolutionising the healthcare industry, with AI/ML playing a key role in the development of new solutions. With more applications of AI/ML in practice, from optimising workflows to improving diagnostic capabilities, the collaborations to learn from the use cases and the partnership to overcome challenges are urgently needed. |
| Project Impact |
|---|
The integrated effort to study real-world applications will ensure the emerging technologies are used effectively and in compliance with relevant guidelines and regulations. |
| Project Leads | |
|---|---|
Ying Su, Pfizer | |
Radha Railkar, Merck | radha_railkar@merck.com |
| Kerry Robson, PHUSE Project Assistant |
CURRENT STATUS Q1 2024 |
|---|
|
Objectives & Deliverables | Timelines |
|---|---|
| Identify the industry knowledge-sharing community of practice, prioritise future project topics | Q2/3 2023 |
| PHUSE/FDA CSS presentation/poster | Q3 2023 |
| Start gathering use cases | Q4 2023 |
Quarterly Community Forums Invited expert talks | Q1 2024 |
| Q2 2025 |
AI/ML Sub-teams |
|---|
| The project volunteers are organised in sub-teams to learn a specific topic through planning/facilitating a forum with experts, and collecting use cases. Please indicate your participation by filling out the form to join the sub-team of your interest |
Sub-Team | Forum Topic | Lead |
GA | Generative AI in healthcare | Jeffrey Lavenberg |
AP | Application of AI/ML in precision medicine (includes RWE) | Shraddha Thakkar |
RL | Regulatory landscape of AI/ML in DHTs (current landscape, knowledge gaps, best practices for regulatory submissions, challenges of regulating AI) | Richard Baumgartner |
MD | AI/ML models (logistic regression, support vector machines, decision tree, convolutional neural networks, etc.) | Hanming Tu |
UC | Challenges of use of AI/ML in DHTs (ethical concerns, privacy issues/cybersecurity, misuse of data, complexity of data management including data interoperability, etc.) | Jessica Hu |
SD | Software-driven medical devices | Anders Vidstrup |