Skip to content
View Mittal-kumar's full-sized avatar
0 followers · 1 following

Organizations

@stanford-developers

Block or report Mittal-kumar

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mittal-kumar/README.md
Typing SVG



         

Education

Institution Credential Year
🌲 Stanford University Data Science in Medicine 2025
🦅 SUNY Stony Brook M.S. Biomedical Informatics 2024–25
🏛️ Sapienza University, Rome B.S. Bioinformatics 2020–23
🇪🇸 San Jorge University, Spain Erasmus Exchange 2023

⚡ I operate at the intersection of genomics and GPU clusters, turning raw EHR signals into early clinical warnings.
My work spans FHIR pipelines, HIPAA-grade governance, and deep learning systems validated on real patient outcomes across 3 continents.

📡   CLINICAL IMPACT METRICS

┌─────────────────────────────────────────────────────────────────────────────┐
│  OUTCOME DASHBOARD  ·  Real Metrics  ·  Real Patients  ·  Real Stakes       │
├──────────────────────────────┬────────────────────┬─────────────────────────┤
│  PROJECT                     │  METRIC            │  CLINICAL VALUE         │
├──────────────────────────────┼────────────────────┼─────────────────────────┤
│  ECG Fall Risk Detection     │  70.99% Accuracy   │  Elderly inpatient      │
│  (XAI / SHAP Interpretable)  │  ████████████░░░░  │  safety & intervention  │
├──────────────────────────────┼────────────────────┼─────────────────────────┤
│  GNN 30-Day Readmission      │  AUROC +11%        │  Multi-modal EHR        │
│  (Stony Brook Medicine)      │  ████████████████  │  predictive precision   │
├──────────────────────────────┼────────────────────┼─────────────────────────┤
│  U-Net MRI Segmentation      │  Dice Score +14%   │  Surgical planning      │
│  (Published · Sapienza 2023) │  ████████████████  │  boundary precision     │
├──────────────────────────────┼────────────────────┼─────────────────────────┤
│  Published Research          │  1 Paper · 2023    │  Peer-reviewed CV Lab   │
│  (Sapienza / Prof. Pannone)  │  ████              │  Computer Vision + MRI  │
└──────────────────────────────┴────────────────────┴─────────────────────────┘

⚙️   TECHNOLOGY MATRIX



[ 🏥 HEALTHCARE SYSTEMS ]

Epic EHR HL7 / FHIR HIPAA ICD-10 CDS

[ 🧠 AI / MACHINE LEARNING ]

GNN U-Net XAI EHR ML ECG

[ 🧬 BIOINFORMATICS ]

Genomics GATK BLAST Multi-Omics TCGA

[ 📊 VISUALIZATION & ANALYTICS ]

Tableau Power BI Plotly R Shiny Seaborn

🔬   RESEARCH SYSTEMS

🧠   [2025]   Interpretable Deep Learning — ECG-Based Fall Risk Detection 
SYSTEM    : XAI Clinical Framework · Fall Risk Stratification in Elderly Inpatients
APPROACH  : ECG time-series → PyTorch model → SHAP explainability layer
OUTCOME   : 70.99% accuracy — every prediction comes with a clinician-readable rationale
STACK     : Python · PyTorch · ECG Signal Processing · SHAP · Clinical Validation
IMPACT    : Clinicians can interrogate predictions — not just receive a black-box score
🕸️   [2024–2025]   Graph Neural Networks — 30-Day Hospital Readmission · Stony Brook Medicine 
SYSTEM    : Heterogeneous Graph Neural Network on Multi-Modal EHR Data
INPUT     : Patient history · Lab results · Diagnoses · Procedures (as graph nodes)
OUTCOME   : AUROC +11% over clinical baseline scoring systems
PROGRAM   : Biomedical Informatics · SUNY Stony Brook · Built within Stony Brook Medicine
STACK     : Python · PyTorch Geometric · Epic EHR · Graph Construction Pipeline
🫁   [2023 · Published]   U-Net MRI Segmentation — Sapienza Computer Vision Lab 
SYSTEM    : Enhanced U-Net for Clinical MRI Tumor Boundary Segmentation
DATA      : 1 TB of multi-modal imaging data · Processed via Snakemake on HPC Cluster
OUTCOME   : Dice Score +14% · Used for precision surgical planning
PUBLISHED : Under Prof. Daniele Pannone · Sapienza University of Rome · October 2023
STACK     : Python · PyTorch · Snakemake · HPC Cluster · Medical Imaging Pipeline

📊   GITHUB INTELLIGENCE

  





🏅   CERTIFICATIONS & CREDENTIALS

     



      

"I work where genomics meets GPU clusters — making healthcare as intelligent as it deserves to be."
                                                                        — Mittal Kumar

   

Pinned Loading

  1. stanford-developers/mittal-kumar stanford-developers/mittal-kumar Public

    biomedicl informatics