Utilizing an integrated multi-omic approach (with genomic, transcriptomic, proteomic, and epigenomic data, as well as clinical and digital health data), we completed analyses for each dataset and used significant results as features in our machine learning (ML) models to predict ALS disease progression. Data is from Answer ALS, the largest public ALS data portal that is partnered with Microsoft, which currently has 1200+ participants (patients/controls) with 2.6 T data points. Learn how integrating these significant multi-omic results forms stronger personalized models to predict disease progression more accurately to better understand disease pathology and can be applied in other therapeutic areas.

The prognosis of lung carcinoma has changed since the discovery of molecular targets and their specific drugs. Somatic EGFR mutations have been reported in lung carcinoma, and these mutant proteins act as substrates for targeted therapies. However, in a resource-constrained country like India, panel-based next-generation sequencing cannot be made available to the population at large. Additional challenges include adequacy of tissue in case of lung core biopsies, locating suitable tumour tissues as a result of innate intratumoral heterogeneity, the shift in EGFR mutation status after subsequent chemotherapy, and reduced DNA quality indicates the necessity of an AI-based end-to-end pipeline capable of automatically detecting and learning more effective lung nodule features from CT images and predicting the probability of the EGFR-mutant. This will help the oncologists and patients in resource-limited settings to achieve near-optimal care and appropriate therapy.  

Artificial intelligence permeates lives, promising to reduce human error while improving speed and accuracy. Specifically, in healthcare where there are vast mountains of data plus advances in computational power, AI is the ideal candidate to support advancements. However, the U.S. FDA which provides guidance to ensure the safety and efficacy of products and services have provided little guidance regarding AI outside of the “ten principles” as part of GMLP. These ten principles while beneficial may be inadequate in providing a robust framework for the development of AI-enabled technology. After attending this presentation, you will walk away with the following: 1) Fundamentals of AI, exploration of AI-enabled technology including clinical decision-making tools, as well as patient care; 2) Exploration of the FDA’s 10 principles of GMLPs; and, 3) Recommendation for a revised framework to support AI development.