Automatic Labeling of Cortical Sulci Using Spherical Convolutional Neural Networks in a Developmental Cohort (DSI-SRP)
This DSI-SRP fellowship funded Lingyan Hao to work in the laboratory of Professor Ilwoo Lyu in the Department of Electrical Engineering and Computer Science during the summer of 2019. Lingyan graduated in 2020 with a major in Mathematics.
The project funded by this fellowship initially aimed to understand the specific characterization of Alzheimer’s disease using biomarkers of the human brain based on existing studies. The research shifted to pediatric brain development as Hao and Lyu felt that it involves more challenging problems from a data science perspective, such as how to handle complex data structure with big structural changes in early ages. The project was in collaboration with an UC-Berkeley neuroscience team. Lyu feels that Lingyan has achieved great success in her project by publishing her work in one conference (preliminary results) and a journal paper (extension) as an outcome of this program. The paper, titled “Automatic Labeling of Cortical Sulci Using Spherical Convolutional Neural Networks in a Developmental Cohort,” was published in the conference IEEE International Symposium on Biomedical Imaging (2020). The follow-up work Labeling Lateral Prefrontal Sulci using Spherical Data Augmentation and Context-aware Training was published in the journal NeuroImage. Lingyan is now studying at Stanford.
In addition to receiving support through a DSI-SRP fellowship, this project was supported and facilitated by the DSI Data Science Team through their regular summer workshops and demo sessions.