In 2011, Marc Andreessen penned a widely acclaimed essay in The Wall Street Journal entitled “Why Software is Eating the World”. Today, driven by the advancements in machine learning models (LLMs) and artificial general intelligence (AGI), a new disruptive force is taking shape, with profound implications for the technology community and humanity as a whole. Consequently, it is now time to expand on Andreessen’s astute observation that software is consuming the world, as AI is rapidly devouring the very software that was one at the forefront of this consumption.
The transformative power of AI within the healthcare industry in particular is truly remarkable. Its impact spans various aspects, ranging from revolutionizing clinical trials and streamlining medical billing to enhancing patient adherence and addressing the pervasive issue of physician burnout. The profound impact that AI can have on these areas is both awe-inspiring and promising for the future of the healthcare industry.
For nearly half a decade, Shashank Agarwal’s work in healthcare data science has fundamentally shaped the healthcare ecosystem. His groundbreaking contributions in machine learning, digital transformation, payor formulary analytics, and the establishment of integrated data warehouses have garnered widespread recognition and have been readily embraced by numerous Fortune 500 companies on a global scale.
Shashank’s journey, which began in India, led him to pursue a master’s degree at the prestigious Johns Hopkins university, where he was bestowed with the esteemed CLE Departmental fellowship. Today, he stands as an international expert in Healthcare AI, propelling innovation and driving advancements within the field. Through the fusion of AI and healthcare, Shashank continues to push the boundaries of possibility, revolutionizing the industry, creating new revenue streams for his stakeholders through innovative projects and significantly enhancing the lives of countless patients. His exceptional expertise and unwavering passion serve as an inspiration to all those who aspire to make a meaningful impact within the realm of analytics.
In a recent interview, we had the privilege of exploring Shashank’s insights regarding the future of AI and its role in the rapidly evolving healthcare industry. His highly regarded work on patient adherence in which he developed AI models to build customized patient profiles has garnered acclaim worldwide, and he graciously shared his expertise and invaluable advice for aspiring analytics professionals
Can you explain the need for adoption and impact of Artificial intelligence (AI) in healthcare?
The adoption and impact of AI in healthcare are of paramount importance at present. AI offers us the remarkable ability to analyze and derive insights from an unprecedented volume of data, which has never before been possible in human history. This breakthrough empowers us to achieve pre-emptive disease detection, revolutionize drug discovery processes, enhance clinical trials, identify medical insurance claim fraud, boost medication adherence, alleviate physician burnout, and prepare for future pandemics. The possible use cases for AI in healthcare are truly endless.
In terms of impact, AI has the potential to streamline healthcare delivery processes, reduce patient out-of-pocket expenses, and provide personalized treatment options. These advancements ultimately lead to improved health outcomes and a more efficient healthcare system. By harnessing the power of AI, we can essentially tap into the vast wealth of data and leverage its insights to transform the industry as we know it, leading to improved patient care and well-being.
In your opinion, what are the major challenges in adopting AI?
To successfully build and implement responsible AI systems, organizations must give special attention to the ethical implications of this technology. There are two key challenges that stand out in this regard. First, it is crucial to address the potential risks of bias and hallucinations in AI models. Since these models are trained on massive datasets, there is a possibility that they can inadvertently incorporate biased information, thereby leading to inaccurate responses or discrimination against certain demographic groups. It is therefore imperative to actively mitigate these risks and ensure that AI systems are fair, unbiased, and reliable in their outputs. Second, efforts should be made to establish clear guidelines for the use of AI in healthcare. These guidelines should define the scope of AI’s role in specific medical settings and delineate the level of human involvement required in the decision-making process. Striking an appropriate balance between AI-driven automation and human expertise is essential to ensuring patient safety and maintaining the trust of healthcare professionals.
What is medication adherence, and can you please tell us about a bit more about your work in the field?
In simple terms, medication adherence refers to the extent to which a person follows the prescribed recommendations from a healthcare provider. Although it notably plays a crucial role in the effectiveness of their treatment, many fail to achieve desired health outcomes due to non-compliance with their prescribed medication schedules. In my work to transform patient medication adherence while employed by pharmaceutical manufacturers, I focused on building statistical algorithms using an integrated patient-level dataset. These algorithms aimed to predict each patient’s propensity to discontinue medication. Based on these predictions and patient profiles, I developed customized plans to prevent discontinuation.
Will there come a time when machines and Artificial Intelligence replace physicians?
Well, as of 2023, the idea of “AI Doctor” remains purely a work of science fiction. AI operates by processing data, recognizing patterns, and conducting statistical analyses. Unlike human reasoning, which is often based on intuition, experience, creativity, and deep understanding of context, AI systems conversely rely on algorithms to analyze data and draw conclusions based on predetermined rules. Even in the foreseeable future, it is highly improbable for AI to possess the same level of understanding, provide equivalent care, or develop prognoses comparable to those of experienced physicians. However, I do envision a future where AI serves as a powerful assistant, assuming more operational tasks and thereby allowing physicians to focus on the bigger picture.
What's the best advice you would like to give to budding analytics professionals?
My primary advice to new analytics professionals is to deepen their domain knowledge in a specific field, master technical skills, collaborate with more experienced professionals, embrace continuous learning, and develop effective communication skills. By understanding the intricacies of their chosen domain, remaining current with the latest technical tools, working closely with established professionals, and enhancing their verbal communication, budding analytics professionals can make a meaningful impact on the rapidly evolving field of AI.
