Session Chair Profile

Developed a new and reliable technique for diagnosing Alzheimer’s disease and measuring the efficacy of experimental treatments

Ph.D., Masters or Bachelor Degree, Staff Scientist, Affiliate Faculty, Data Scientist, LBNL, UCSF, UCB

Dani Ushizima is a Computer Scientist focused on Computer Vision and Machine Learning algorithms to characterize materials toward self-driving labs. Her research has impacted projects that depend on experimental data coming from instruments reliant on x-ray, electron, confocal, and other light-matter interactions. In 2015, Ushizima received the U.S. DOE Early Career Research award to work on pattern recognition for scientific images. In 2021, she was honored by 3M as one of the top "25 Latina in Science" for biomedical research. Also in 2021, Ushizima and colleagues received the Berkeley Lab Halback award for creating machine-learning-based techniques to solve a problem that has plagued third-generation light sources: fluctuations in beam size. Currently, Ushizima leads research projects in the Center for Advanced Mathematics for Energy Related Applications (CAMERA). Jointly with UCSF Grinbergs lab, she has developed a new technique for diagnosing Alzheimer’s disease and supporting measurement of the efficacy of experimental treatments.

Alzheimer's immunohistochemistry and deep learning
We will discuss our end-to-end deep learning-based workflow to process human brain, combining imaging antibodies with CT, MRI, and PET. We will show how Tau marked with multiple antibodies holds the promise to help validation of the relationship between molecular signals detected by PET imaging AND the corresponding neural microstructures they originate from.