Brain Imaging: From Macro to Micro

Lecture

Speaker(s)

Maryana Alegro

Postdoctoral Scholar, Grinberg Lab, University of California, San Francisco

Maryana Alegro is a computer scientist and received her MS and PhD in electrical engineering from the University of São Paulo Polytechnic School. Her major experience is in medical image processing, especially MRI and histological images and, more recently, polarized light imaging. She is currently a post-doc at the UCSF Grinberg Lab where she is responsible for writing image-processing tools that can assist researchers in studying dementia. She also has strong experience as an enterprise software developer, having worked on the development of several medical systems for the Brazilian government.

Friday, September 30, 2016

BIDS Data Science Lecture Series | September 30, 2016 | 1:10-2:30 p.m. | 190 Doe Library, UC Berkeley

Image registration and machine learning are brought together to help researches better understand Alzheimer's disease (AD) using ex-vivo tissue images. We use registration algorithms to build high-resolution whole-brain histology volumes that allow researchers to visualize and measure macroscopic features, such as volume, even in deep brain structures not visible in standard brain MRI. On the other hand, we use machine learning to automate cell detection and counting on microscopy images, thus improving researchers' productivity when assessing microscopic tissue features. These tools are put together to study two elusive brain structures, the Locus Coeruleos (LC) and Dorsal Raphe nucleus (DRN), which are highly affected by AD even during early stages of the disease but are rarely seen in MRI and other brain-imaging modalities due to their reduced size. In this talk, I’m going to present our current results in high-resolution registration and the challenges posed by this task, together with our results on segmentation of LC and DRN microscopy images using machine learning techniques. 

Share this video