Dr. Ilias Tagkopoulos, DEUS EX MACHINA founder, discusses the potential of AI in Healthcare at the Aristotle Medical Forum 2022
Jul 20, 2022 – Our founder Dr. Ilias Tagkopoulos and our scientific advisor Dr. Michael Chourdakis participated in the Aristotle Medical Forum 2022 to discuss innovative initiatives about biomedical technologies.
Dr. Ilias Tagkopoulos presented the challenges of applying AI in medicine and healthcare and shared insights from past scientific work his team has accomplished.
Perfect Storm [1:42]
Today, the AI market in healthcare is estimated to be 11B $ and expected to grow to 190B $ by 2030 although only 20% of physicians attest to the changes that AI brought to medicine. What’s more, pioneer companies exaggerated the capabilities of deep learning and AI which resulted in them failing to live up to the expectations of the community.
AI Application in Medicine [4:55]
In recent years, a variety of healthcare applications were created like voice medical coaches and paramedic dx of heart attack/stroke, to name just a few.
Artificial intelligence is not just a predictor; it encompasses an overarching system that can help in critical decision-making and provide data-informed, timely recommendations, Dr. Tagkopoulos says.
Relevant Areas of Machine Learning and AI [8:08]
AI can be the backbone of healthcare by:
- Creating predictors: Regression and Classification
- Finding patterns in unknown data: Clustering and Dimensionality Reduction
- Learning from other domains: Transfer Learning
- Learning optimally as you go: Active Learning
- Learning about the environment and best actions: Reinforcement Learning
Challenges of AI in Medicine [11:35]
Data, data, data. Unfortunately, there are significant difficulties in finding public data, merging them, and extracting real value from them with high accuracy.
Our Work: Examples of Health-Related Projects [19:03]
The University of California*, AIFS, and the companies that Dr. Ilias Tagkopoulos founded, PIPA LLC and DEUS EX MACHINA IKE, have been using AI for more than 10 years. One of the most important innovations is the creation of an AI-powered diagnostic tool for cats that predicts chronic kidney disease ( CKD) up to 2 years before it occurs.
Using AI to Discover Novel Bioactives [21:19]
Dr. Tagkopoulos’s team built a science discovery platform, LEAP, that integrates structured and unstructured data to enable scientists to discover novel bioactive compounds, find supporting evidence for their experiments, and accelerate target identification.
The vision is to build a world where AI-assisted healthcare will be more affordable, safe, precise, and personalized for everyone.