AI & Data Scientist

AI Engineer and Bioinformatics Researcher with a strong background in Machine Learning, Computer Vision, and multi-omics data analysis.

My work focuses on applying deep learning and statistical modeling to real-world healthcare and genomics problems, including disease prediction, medical image analysis, single-cell transcriptomics, and causal inference.

I’m particularly interested in building AI systems that bridge computational research and clinical impact.

Mathematics & Statistics

• Linear Algebra • Probability & Statistics • Statistical Modeling • Hypothesis Testing • Causal Inference

Programming & Engineering

• Languages: Python • R • SQL • Bash • Git • Docker • Linux

AI & Deep Learning

• PyTorch • TensorFlow • CNNs (ResNet, VGG, EfficientNet) • Object Detection (YOLOv8) • Image Segmentation (U-Net) • Medical Image Analysis • Transfer Learning • Autoencoders • GANs • Transformers • LLM-based Agents

Machine Learning & Data Science

• Scikit-learn • XGBoost • LightGBM • Pandas • NumPy • Ensemble Methods • Feature Engineering • Model Interpretability (SHAP) • Hyperparameter Tuning

Cloud & MLOps

• Platforms: AWS • Huawei Cloud
• Deployment: Docker • Model deployment (ngrok, Flask & Fast APIs)

Bioinformatics & Omics

• Workflow & Environment Management: Conda • Nextflow • Snakemake
• Bioinformatics Tools: FastQC • fastp • Trimmomatic • Kallisto • Bowtie2 • GATK • SAMtools • BCFtools
• Omics Analysis: RNA-seq • scRNA-seq • WGS • WES • Metabolomics • Multi-omics Integration

Integrated scRNA-seq, DNA methylation, and GWAS data to uncover inflammatory drivers of periodontal-diabetic comorbidity using Mendelian Randomization.

• Tech: Seurat • Harmony • minfi • limma • TwoSampleMR • STRINGdb
• Skills: Single-cell analysis, epigenomics, causal inference, network biology

Developed an end-to-end medical imaging pipeline for early lung nodule detection and malignancy analysis from CT scans.

• Tech: PyTorch • YOLOv8 • U-Net • EfficientNet • OpenCV • LangChain • Gradio
• Skills: Object detection, medical image segmentation, classification, multimodal AI, report generation

Built deep learning models for retinal disease classification using fundus images, achieving 99% classification accuracy across 5 disease categories.

• Tech: TensorFlow • ResNet50 • VGG19 • Flask • Flutter
• Skills: Computer vision, transfer learning, medical imaging, deployment

Designed a Vision-Language AI assistant enabling clinicians to query medical images in natural language with context-aware responses.

• Tech: HuggingFace Transformers • BLIP • LangChain • OpenCV • Gradio
• Skills: Vision-language models, multimodal AI, LLM agents, medical AI

Developed ML models on clinical and genetic biomarkers for CAD risk prediction, achieving AUC up to 0.956.

• Tech: Scikit-learn • Random Forest • SVM • SciPy • StatsModels
• Skills: Predictive modeling, biostatistics, feature engineering, explainable AI

• 📧 Email: khaledghanem597@gmail.com
• 💼 LinkedIn: Khaled Ghanem
• 🌐 Upwork: Khaled M.

I'm open to collaborations in:

• Computational biology & bioinformatics research
• Multi-omics data analysis and integration
• Medical AI and disease prediction models
• Open-source bioinformatics tool development

Feel free to reach out if you'd like to discuss data science, genomics, or AI applications in healthcare!