Researchers from Penn Engineering and the University of Pennsylvania Perelman School of Medicine (PSOM) are advancing artificial intelligence (AI) systems to enhance personalized medical care. Their efforts have been bolstered by a $7 million, four-year grant from the Advanced Research Projects Agency for Health (ARPA-H). The initiative will initially target breast cancer, heart attacks, and sepsis.
The project is spearheaded by Rajeev Alur, Zisman Family Professor of Computer and Information Science at Penn Engineering and Director of the AI-enabled Systems: Safe, Explainable, and Trustworthy (ASSET) Center. The team aims to create AI tools that not only provide accurate clinical predictions but also offer explanations that build trust among clinicians. Alur highlights this dual focus, stating, “We are excited to tackle the challenge of creating AI models that deliver both precision and interpretability.”
While AI technologies such as machine learning and large language models have shown promise in tasks like image classification and medical inference, the researchers note that these systems often lack the transparency and reliability required for clinical application. One obstacle is that AI systems depend on vast datasets for accurate predictions, but access to medical data is restricted due to privacy regulations. Additionally, many current AI systems function as “black boxes,” obscuring the processes behind their decisions and outputs.
To overcome these challenges, the research team will pursue approaches that integrate data-driven AI models with logical and symbolic reasoning. This effort seeks to make AI systems both transparent and precise enough for real-world medical use.
The team will work closely with clinicians to address practical considerations, such as data availability, integration with existing workflows, and patient-specific variability in medical markers. By doing so, the researchers hope to develop systems that help predict patient responses to treatments for breast cancer, heart attacks, and sepsis, ultimately improving health outcomes.
Alur emphasizes the collaborative nature of this project, which combines expertise from Penn Engineering and PSOM. “Our research thrives on the synergy between Penn Engineering’s strengths in machine learning and PSOM’s clinical expertise,” he explains. Initiatives such as seed grants for AI research and symposiums on trustworthy AI underscore the value of this partnership in driving innovation.
Vijay Kumar, Nemirovsky Family Dean of Penn Engineering, echoes this sentiment, calling the project a transformative collaboration. “Clinical challenges offer impactful opportunities for AI, and Professor Alur has assembled a stellar team to establish the foundations for trustworthy AI in medicine,” Kumar states.
This groundbreaking initiative represents a significant step toward harnessing AI to advance patient care through transparent and reliable systems.
Are you interested in how AI is changing healthcare? Subscribe to our newsletter, “PulsePoint,” for updates, insights, and trends on AI innovations in healthcare.
