Muhammad Khalid Khan Niazi

This study uses deep learning to analyze spatial immune patterns in tuberculosis-infected mice, identifying distinct granuloma features associated with asymptomatic, acute, and chronic disease states.

Weakly supervised deep learning localizes focal tissue regions enriched for therapeutically relevant biology in routine histology, offering a scalable strategy to study therapeutic resistance across large patient populations.

This paper applies geographically weighted regression to spatial proteomics data to better model protein dispersion in tumors, showing improved accuracy over global regression methods.

The paper integrates replication timing and morphology into individualized gene networks to improve subtype prediction and network robustness in pancreatic cancer.

The paper develops a computational pipeline to analyze TP53 missense mutations using quantum chemistry, integrating whole-exome sequencing data with chemoinformatics to study electronic properties of protein mutants.

This paper characterizes neural network learning dynamics in digital pathology by measuring activation structure, weight evolution, and entropy changes during training.

This paper evaluates spectral models in histopathology tasks and finds they underperform CNN baselines but offer interpretability benefits.

This paper introduces a deep learning classifier using PointNet on RNA-Seq data to classify tumors with over 99% accuracy.