Baruch Students Publish AI-Informed Diabetes Study in Healthcare Journal

Baruch Students Publish AI-Informed Diabetes Study in Healthcare Journal

September 18, 2026

Amid all the doom and gloom about AI, one of the many upsides is its potential to improve healthcare. For example, AI can help medical researchers and healthcare providers analyze vast amounts of data quickly and use the information to provide real-world benefits.

That's what a group of Executive MBA in Healthcare Administration (EMBA HCA) students did recently in AI-Assisted Healthcare Analytics and Quality, a course taught by Associate Professor Tolga Aydinliyim, PhD (Narendra Paul Loomba Department of Management). The students-who all have several years' experience working in local healthcare systems, including Mount Sinai and NYU Langone-were tasked with devising a project that would highlight predictive analytics and decision modeling in a healthcare setting they chose themselves. They came up with a data-driven method for following up with diabetes patients to prevent hospital readmission within 30 days of discharge.

Hospitals have long grappled with the expensive problem of readmitting diabetes patients. Managing diabetes at home can be complicated, and if patients misunderstand discharge instructions or forget to take their insulin, their blood sugar may become unstable again, sending them back to the hospital. Traditionally, hospitals have handled this by sending phone or text reminders to all patients upon discharge, but this method is time-consuming and inefficient.

Dr. Aydinliyim's students built a model to predict which patients were most at risk of readmission within 30 days of discharge, based on key factors: extended hospital stays, comorbidities (additional diagnoses), multiple diabetes medications, recent hospital stays, and missing A1C (blood sugar) tests. The result was a tailored approach, in which the highest-risk patients got the most follow-up care. Targeting resources to the patients most in need of them can help hospitals improve patient outcomes without dramatically increasing costs.

So how did this project end up published in Health Management, Policy and Innovation, an open-access journal for healthcare practitioners? The journal is published by the Business School Alliance for Health Management (BAHM), a consortium to which Baruch College and the Zicklin School belong. When Aydinliyim learned that BAHM was sponsoring a student essay competition, he encouraged his students to write up the diabetes project and submit it.

"They didn't win the competition, but the submission was good enough that they were invited to contribute to the journal," Aydinliyim says. "At that point I got more involved as a co-author because I'm more familiar with academic writing. I also brought in one of my doctoral students to help replicate the students' class project using a large, publicly available multi-hospital data set of 30,000 diabetes patients."

The final result was "Operational Effectiveness of AI-Assisted Targeted Intervention System Design in Healthcare: Reducing 30-Day Readmission Rates for Diabetes Patients," published July 1 in Health Management, Policy and Innovation by EMBA HCA students Amanda Wong, Tanzina Khan, Melissa Florus, Milagros Ramos, and Miguel Rodriguez; Zicklin School doctoral candidate Yi Zhuang, and Aydinliyim.

"It's a huge accomplishment for master's candidates to be published in a journal, even a practitioner-focused one like this," Aydinliyim concludes. "I told the students they should consider pitching this project to the healthcare systems where they're based, as a simple, cost-effective intervention that can benefit both patients and hospitals."

Discover more information about Zicklin's Executive MBA in Healthcare Administration program.

* This copy was written by the Zicklin School. Visit the Zicklin School of Business to read about its undergraduate and graduate academic programs.