In this talk we will discuss the state of AI/ML in Healthcare, the challenges and pitfalls facing the industry, effective ways of mitigating negative impacts, considerations for ethics and governance in health AI, improving health equity and the business case for responsible AI. As AI/ML usage becomes commonplace across the healthcare ecosystem, implementing a transparent, ethical, sustainable AI/ML solutions becomes critical for the entire spectrum of healthcare players; from patients to policy makers, from providers to payers, and everyone in between. Since healthcare is a field that impacts one and all, the pitfalls and concerns from flaws and biases in data, algorithms, and humans are become increasingly evident and is a cause for concern for all involved. Thus, there is an increasing urgency to address these issues at the very beginning of the making of any health AI product or system. When consumers demand privacy and transparency, when policy and regulatory frameworks are evolving and changing rapidly, how do we ensure that the health ML solutions are keeping up? As healthcare delivery becomes more and more global, with digital healthcare tools and apps, care and consideration for international regulations and the needs and demands of a global population is a must for any solution provider.
In this talk, I will discuss the tremendous value and impact ML creates in healthcare as well as the huge pitfalls of sub-optimal development and implementation of health AI, all with real world examples. I will then explore the necessary considerations for AI/ML in healthcare and health applications, the challenges facing the health AI community, and how we can mitigate the negative impacts of AI in healthcare from the very beginning. This will include best practices, governance and ethics considerations that will help in the creation of a robust, sustainable AI solution. I will conclude the talk by making the business case for ethical and responsible AI while improving health equity.