What if data could help predict patient prognosis, streamline hospital operations, and optimize medical talent? A book fresh from the shelf called “The Analytical Edge of Healthcare” shows that this is already happening and shows how to expand it.
With an emphasis on Open Learning at MIT, written by Dimitris Bertsimas, two former students at Bertsimas, Agni Orfanoudaki Phd '21, Associate Professor of Operations Management at Said Business School at Oxford University, and Holly Wiberg PhD '22, providing public policy and operational research at Carnegie Mellon University, the first part of the book presents an integrated case study covering a variety of clinical specialties and problem types using descriptive, predictive and normative analysis, and the second part of the book establishes a technical foundation spanning machine learning and optimization.
Part of a broader series, The Analytics Edge in Healthcare shows how to leverage data and models to make better decisions within the healthcare sector, but its predecessor, the Edge of Analytics, is divided into science that uses data to build models, improve decision-making, and add value to institutions and individuals.
Bertsimas, a global operations professor at the MIT Sloan Management School of Management, and an associate dean and Boeing Leader in Business Analytics, is the innovator behind MIT Open Learning's MITX course, 15.071 (The Analytics Edge). Bertsimas takes a break from his research, and his work at MIT Open Learning discusses how the field of analysis is changing the health care system, sharing the incredible ways in which analysis is already being used in hospitals.
Q: How is the field of analysis changing the way hospitals provide care and manage their businesses?
A: As an academic, I have always been educated, written publications and aimed to make use of what we actually do. Therefore, we established Holistic Hospital Optimization (H20) with the aim of optimizing hospital operations with machine learning to improve patient care. We developed a variety of tools at MIT and implemented them in hospitals around the world. For example, it manages patient length of stay and degradation index (a computerized tool that predicts the risk of clinical degradation in patients). It manages nurse optimization and how hospitals properly allocate personnel. It also optimizes surgical blocking. This is the beginning of a change in which analytics and AI methods are currently in great popularity. My hope is that this work and this book will accelerate the effectiveness of using these tools.
Additionally, I taught nine lecture courses twice at Agni and Holly at the Hartford Hospital System. Here we realized that these analytical methods (usually not taught in medical schools) can be demonstrated for healthcare professionals, including doctors, nurses and administrators. To make an impact, you need to have the right methods, implement and apply them, but you also need to educate people about how they are used. This links well with my role in open learning. Here, our aim is to educate learners globally. In fact, Open Learning has launched Universal AI this fall, a dynamic online learning experience that provides comprehensive knowledge about artificial intelligence, preparing a global audience for employment in the rapidly evolving job market.
Q: What is the amazing way in which analytics is used in healthcare that most people don't expect?
A: Using the analysis, patients stayed at Hartford Hospital was reduced from 5.67 days to 5 days. There are algorithms that predict patient release potential. Therefore, doctors prioritize the patients with the highest probability and prepare them for discharge. This means that the hospital can treat far more patients, and patients stay in the hospital.
Furthermore, when nurse turnover rates increased during the Covid-19 pandemic, hospitals have developed an analytical system that takes into account stocks and fairness, which reduces overtime costs, giving nurses a priority slot, and significantly reducing overall sales. These are two examples. There are many others whose analytical perspectives on healthcare and medicine have made a significant difference.
Q: Looking ahead, how do you see artificial intelligence shaping the future of healthcare?
A: In a very important way – use machine learning to make better predictions, but generative AI can explain them. I'm already watching the movement in that direction. It is truly the evolution of AI that made this possible, and it is exciting. It is also important to the world because it has the ability to improve care and save lives.
For example, through a program at the Hartford Hospital System, we discovered that patients were exacerbated and predicted that they would become even worse through our analysis. After our prediction, the doctor examined the patient more closely and found that the patient had an early case of sepsis. If no sepsis was detected previously, the patient may have died. This made a real difference in saving lives.
Q: What happens if you have to explain “the analytical edge of healthcare” in one or two words?
A: This book is a gradual transition in healthcare as it can impact the healthcare sector in ways not previously done. This book really outlines my work in healthcare and its applications over the past decade.
