A study led by Qiong Wu, PhD, assistant professor of biostatistics and health data science at Pitt’s School of Public Health, is using artificial intelligence (AI) to tailor treatment and care in hospital settings.
Wu, who started the work while a postdoctoral research fellow at Penn Medicine, and other Penn colleagues analyzed the electronic health records of long-COVID patients using a machine learning technique called latent transfer learning. Through this process, they were able to identify four sub-categories of patient conditions: mental health, atopic/allergic, chronic non-complex, and chronic complex.
“Without identifying these distinct subpopulations, clinicians and hospitals would likely provide a one-size-fits-all approach to follow-up care and treatment,” said Wu. “While this unified approach might work for some patients, it may be insufficient for high-risk subgroups that require more specialized care. For example, our study found that patients with complex chronic conditions experience the most significant increases in inpatient and emergency visits.”
“Existing studies pool data from multiple hospitals but fail to consider differences in patient populations, which limits the applicability of their findings in local decision-making,” added Yong Chen, PhD, Penn professor of biostatistics and senior author of the study, which was published in the journal Cell Patterns. “Our work moves toward providing actionable insights that can be tailored to individual institutions and can further the goal of offering more adaptive, personalized care.”
If the machine learning system had been in place in early 2020, Wu contends that it may have provided some key insights to mitigate some of the impacts of the COVID-19 pandemic, including the ability to focus resources and care for populations most in need.
“This would have allowed each hospital to better anticipate needs for ICU beds, ventilators or specialized staff—helping to balance resources between COVID-19 care and other essential services,” she said. “Furthermore, in the early stages of the pandemic, collaborative learning across hospitals would have been particularly valuable, addressing data scarcity issues while tailoring insights to each hospital’s unique needs.”
Beyond crises such as the COVID-19 pandemic and its aftermath, the AI system developed by Wu, Chen, and their team could help hospitals manage much more common conditions.
“Chronic conditions like diabetes, heart disease, and asthma often exhibit significant variation across hospitals because of differences in available resources, patient demographics, and regional health burdens,” said Wu. “For example, an urban hospital may have access to advanced cardiac care, impacting treatment strategies for heart disease, while a rural facility might follow a more generalized approach due to limited resources. Additionally, geographic factors—such as higher asthma rates in areas with elevated pollution—shape hospital practices and patient needs.”
Even hospitals not able to actively employ machine learning would benefit, however, through shared information.
“By utilizing the shared findings from network hospitals, it would allow them to gain valuable insights,” concluded Wu.