Portfolio

Research Projects

Neuro Insight

Brain tumor grade (HGG/LGG) classification using ConvNext-Base (98.6% Accuracy)

Project Code

Muhammad Usman Khan1, Dr. Eid Rehman2

Brain tumor grade (HGG/LGG) classification using ConvNeXt-Base, achieving 98.6% accuracy on BraTS 2019.

Computer Vision Medical Imaging
Neuro Insight

Towards Explainable Facial Emotion Recognition

Project Code

Muhammad Usman Khan1

A trustworthy facial emotion recognition pipeline combining ResNet-50, Grad-CAM explainability, and demographic fairness auditing.

Computer Vision Trustworth AI Explainable AI
Neuro Insight

Trust-Aware Federated Learning for Network Intrusion Detection

Project Code

Muhammad Usman Khan1

A privacy-preserving intrusion detection system built with Federated Learning, a Dual-Head Autoencoder, and a lightweight A/R/C trust gating mechanism including zero-day and poisoning attack scenarios.

Federated Learning AI in Cyber security Intrusion detection
Neuro Insight

Brain Tumor Segmentation using Attention-UNet

Project Code

Muhammad Usman Khan1, Dr. Eid Rehman2

Precise multi-class brain tumor segmentation using Attention-UNet on the BraTS 2019 dataset, fully integrated into the Neuro Insight clinical application

Computer Vision Medical Imaging