Often deemed “elusive” due to under-documentation and lack of clear diagnostic markers, conditions such as endometriosis, uterine fibroids, and menstrual disorders inflict significant patient burden and contribute to rising healthcare costs. Our work in the Ensari Lab confronts these prevalent yet poorly-understood women’s health challenges through research-based mobile health tools and AI methodologies. We strive to illuminate these “hidden” diseases, accelerate diagnosis, and personalize care, directly addressing health disparities that have too often left vulnerable populations behind.
Our interdisciplinary team pursues three strategic pillars to fulfill this mission:
-
Patient-Centered Data Integration: Combining patient-generated data from mobile apps and wearables with electronic health records to define more accurate disease profiles and support timely, tailored interventions.
-
AI-Driven Personalized Pain Management: Developing innovative machine learning models to empower non-pharmacological, self-managed approaches for pain relief that improve quality of life and reduce reliance on opioids.
-
Expanding Access Through Digital Health: Delivering scalable mHealth solutions that reach underserved communities, bridging critical gaps in healthcare access and education.
Currently active research studies focus on:
1) ML-instilled generalized additive modeling for harmonizing and summarizing temporal multi-modal data
2) Investigation of large language models for predictive modeling of infrequently-documented diagnoses and conditions for improving diagnostic performance under real-life clinical settings.
3) Integrating genomic markers and genetic data into mHealth and clinical data for improving model performance.