*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Juan Carlos (JC) Vizcarra
BME PhD Proposal Presentation
Date:2022-01-12
Time: 10:30 AM-12:30 PM
Location / Meeting Link: https://bluejeans.com/412543815/0399
Committee Members:
David A. Gutman, MD/PhD (Advisor) May D. Wang, PhD Jia Shu, PhD Eva L. Dyer, PhD Thomas Pearce, MD/PhD
Title: A computational platform for standardizing neuropathology cohorts and deploying deep learning tools to improve postmortem evaluation
Abstract: Postmortem evaluation of brain tissue is the definitive method for diagnosing neurodegenerative diseases. These semi-quantitative evaluations suffer from poor agreement across raters and institutions, affecting research and clinical studies. The objectives of my proposed project include the creation of an annotated neuropathology (NP) dataset of whole-slide-images, detection of neurofibrillary tangles (NFTs) using deep learning (DL), and the development of an interactive platform for multi-institutional data harmonization and computational analysis. Participants with various backgrounds will annotate a novel NP dataset for inter-rater agreement of NP evaluation and NFT annotations (Aim 1). Object detection, semantic segmentation, and ensemble modeling will be used to develop an NFT detection tool. This will be used to compare computationally-derived scores against human NP evaluation (Aim 2). Incorporation of this work into a novel platform for data harmonization and DL analysis deployment will be used to investigate NFT morphological associations with proteomic signatures (Aim 3). The platform proposed here will provide a standard for the harmonization of NP cohorts from various institutions, providing access to an ever growing database for research studies. Furthermore, the infrastructure for deploying various types of DL analysis in NP datasets will be provided, including efficient scaling to large cohorts.
