BME PhD Defense Announcement for Sang Mok Park (Y. Kim, advisor)

Everyone is invited to attend the public presentation beginning at 3:00pm.

Title: Tissue Optics-Informed Hyperspectral Learning for Mobile Health

Date: September 11, 2023

Time: 3:00pm

Location: MJIS 2001 and Zoom: https://purdue-edu.zoom.us/j/4572570338

Examining committee members:

Dr. Young L. Kim, Chair

Weldon School of Biomedical Engineering

Dr. Craig J. Goergen

Weldon School of Biomedical Engineering

Dr. Jacqueline Linnes

Weldon School of Biomedical Engineering

Dr. George T. Chiu

School of Mechanical Engineering

Abstract:

Blood hemoglobin (Hgb) testing is a widely used clinical laboratory test for various patient care needs. However, conventional blood Hgb measurements involve invasive blood sampling, causing iatrogenic blood loss. Although noninvasive blood Hgb prediction methods are under development, their performance often does not align with clinical laboratory blood Hgb tests. Optical spectroscopy can provide reliable blood Hgb tests, but its diagnostic applications are limited by bulky optical components, high costs, and slow data acquisition. Mobile health (mHealth) colorimetric diagnostics have a potential for point-of-care blood Hgb testing. However, achieving diagnostic color accuracy is a complex matter, affected by device models, light conditions, and image file formats. To address these limitations, we propose biophysics-based algorithms that combine hyperspectral learning and spectroscopic gamut-informed learning for reliable mHealth blood Hgb quantification. The palpebral conjunctiva serves as an ideal peripheral tissue site, owing to its easy accessibility, relatively uniform microvasculature, and absence of skin pigmentation. First, hyperspectral learning reconstructs a high-resolution spectrum of the palpebral conjunctiva from red-green-blue values of a digital camera, eliminating the need for complex and costly optical instrumentation. Second, spectroscopic analyses of peripheral tissue establish a unique color gamut and design diagnostic reference colors highly sensitive to blood Hgb. Informed by tissue optics and machine vision, the Hgb gamut-based learning algorithm offers device/light/format-agnostic color recovery of the palpebral conjunctiva. This mHealth blood Hgb assessment exhibits comparable accuracy to capillary blood sampling tests over a wide range of blood Hgb values. Importantly, a single-shot photograph allows noninvasive, continuous, and real-time Hgb readings in resource-limited and at-home settings. Furthermore, our biophysics-based digital health approaches can lay the foundation for personalized medicine and facilitate the tempo of clinical translation.