Xiaoxuan Ren

SAMS improves spike sorting accuracy and efficiency for hPSC-derived neurons, enabling better disease modeling and drug development.

This paper introduces a method to reconstruct neuronal network connectivity from silent neurons using stimulation and supervised learning, achieving high-fidelity connection weight inference and accurate spike train prediction.

The paper introduces a supervised learning model to infer presynaptic connectivity from temporally blurry spike trains, demonstrating its effectiveness in reconstructing network connectivity under varying temporal blur conditions.