G. Zaki

The paper introduces a deep learning model, MAXIM, to impute missing protein markers in multiplex images, improving efficiency in staining and imaging processes.

The paper identifies MAIT cells as immunosuppressive in glioblastoma, correlating their presence with poor patient survival and immune suppressive gene upregulation.

This paper develops a deep learning pipeline to evaluate nuclear segmentation models trained on small, custom annotated datasets, showing they can match or exceed performance of pre-trained models.