The ability of AI to integrate and interrogate multi-modal, large datasets and inputs within seconds has ushered in a new era in medicine that is changing every area of practice. Oncology lends itself well to AI applications because it is a genomically driven disease with complex management pathways. This presentation will discuss platforms that optimize treatment selection, drug discovery, and patient empowerment.

AI, in all of its forms, boils down to looking at and learning from data, better-its potential impact to medicine and healthcare is well known-yet its impact remains tethered by our ability to leverage these tools without introducing more bias in the algorithms generated. This talk will discuss translational science efforts and workforce training, funded by the NIH, to focus on Good Algorithmic Practice and its propagation.

Despite advances in tumor measurement technologies and computational approaches such as AI for biomarker discovery, there remains a critical gap in adapting treatments to the dynamic biology of tumors as they mutate and develop resistance to therapies. This presentation will discuss a new national program, “ADAPT: ADvanced Analysis for Precision cancer Therapy” that will support the identification of multi-gene biomarkers from multi-modal data to inform therapy choices as cancer changes within a new evolutionary clinical trial design that follows metastatic cancer patients longitudinally throughout their disease progression.

The FDA regulates artificial intelligence in precision medicine by ensuring that AI-based medical devices and software meet rigorous safety, efficacy, and quality standards before they are approved for clinical use. This presentation will cover some FDA frameworks for continuous oversight and post-market surveillance to address the unique challenges presented by adaptive learning algorithms.

With the explosion of advanced digital innovations now employed across all aspects of healthcare, the journey of a data point is no longer linear. But what does its new journey look like? What is the feedback loop from patients and clinicians to early development scientists and vice versa? How has this changed collaboration within companies and with their external partners? Are these long-sought efficiencies creating new challenges? How can companies ensure strategic and thoughtful investments in these data-first approaches? And most importantly, what do any of these ultimately do for patients?

Artificial Intelligence (AI) presents opportunities to accelerate personalized medicine drug and diagnostic discovery, foster adoption of personalized medicine approaches, and improve the patient experience with personalized medicine. Implementation of AI can also help address challenges to the clinical adoption of personalized medicine. This session will inform attendees about principles developed by leaders in the personalized medicine community to guide the development of public policies that facilitate implementation and uptake of AI.

We apply the latest deep learning technologies to improve interpretation of human genetic variation in large biobank-scale cohorts, comprising hundreds of thousands of individuals with medical record data, and demonstrate novel insights for genetic risk prediction and drug target discovery.