Shiwei Li

The paper introduces a bioimaging data processing pipeline that improves data transformation efficiency and accelerates downstream tasks by optimizing I/O throughput and reducing memory overhead.

This paper introduces GeminiMol, a molecular representation model that incorporates conformational space profiles to improve performance on molecular property prediction and zero-shot tasks like virtual screening.

This paper introduces ST-LFV, a method that improves weak signal identification in neuronal images, achieving higher recall and precision than existing tools for large-scale neurite tracing.