Junming Zhu

This paper develops a hybrid deep learning model to decode human emotion states from intracranial neural activity, achieving high performance and real-time implementation.

This study uses spatial transcriptomics to identify gene expression changes in the mouse brain following nucleus accumbens deep brain stimulation, revealing alterations in synaptic function and neuronal connectivity.

The paper uses infrared neural stimulation and fMRI to map mesoscale functional networks in the primate amygdala, revealing distinct connectivity patterns of its nuclei.

The paper investigates dynamic lateralization in contralateral-projecting corticospinal neurons during motor learning in mice, finding heterogeneous activity changes and flexible network reconfiguration.

This study investigates the role of high gamma activity between the precuneus and hippocampus in mnemonic metacognition using intracranial recordings from epilepsy patients.

The paper uses Utah array recordings to show that the human motor cortex encodes handwriting through a sequence of primitive neural states, improving trajectory reconstruction accuracy by 84%.

This study investigates the role of cortical and subcortical brain structures in speech decoding using sEEG signals and a deep learning model, finding that both contribute to different aspects of speech prediction.