Abstract: Cataracts and corneal opacities are eye disorders that affect the vision and are two of the most common causes of blindness world-wide, ranking as first and fourth, respectively. Early detection of these disorders can facilitate the treatment and minimize the need for complex surgeries or transplants. However, the detection of cataracts and corneal opacities can be delayed due to the interruptions of eye care visits such as those caused by the Covid-19 pandemic, limited access to healthcare and eye specialists for people living in remote and resource-limited areas, rising healthcare costs, as well as delayed diagnosis or referral of patients. In this work, I am developing a novel screening and diagnostic tool that will allow the objective, remote, cost-effective and automated detection of cataracts and corneal opacities. This tool will be based on machine learning and computer vision algorithms for automated detection of cataracts and corneal opacities. Additionally, it will include automatic classification of these disorders based on the degree of severity. Furthermore, I will design a digital health platform to enable the use of these screening and diagnostic models on the smartphones or other digital devices.

Title: “Clinically Meaningful Integration of Wearables Data for Disease Progression Monitoring in Patients Newly Diagnosed with Multiple Sclerosis”

Speaker: Mithra Vankipuram, PhD, Innovate for Health Fellow

Abstract: For patients who are newly diagnosed with multiple sclerosis (MS), this experience is filled with uncertainty. Questions patients face include: How bad is my version of MS? Will I just have flares (relapsing MS), or will I keep getting worse until I’m a burden (progressive)? How much worse will I get and how fast? In addition to uncertainty when newly diagnosed, there is uncertainty in the care experience for clinicians as well. Patients visit a specialist between once in two years to twice a year, depending on the severity of their diagnosis. While clinicians get a snapshot of what is going on with the patient at the time of the visit, there is little visibility into the patient’s life other than patient reports. The use of wearables data provides an opportunity for newly diagnosed patients to understand how the small “steps” they take every day with respect to nutrition, exercise, stress management (e.g., meditation), medication adherence, or sleep affect their outcome. In addition, this data may be used to power a data product that informs the clinician if the patient’s disease state has worsened and how that compares to a clinical metrics captured on per patient or cohort level. At UCSF, the largest unmet need is data visualization research to understand how best to display wearables data at the point of care that is meaningful to clinicians and patients in their interaction. Project Curie will take an iterative product development approach, starting with building out a minimally viable product for integrating wearables-based activity data into clinical practice. This includes prototypes of patient and clinician experience powered by synthetic data generated from statistical models for daily step count data derived from real MS patients.