Vamsi Sudersanam

Hi, I'm Vamsi. I'm a visiting scholar at the Medical College of Wisconsin, where I build data pipelines and machine learning models that support clinical decision-making.

My latest project predicts, early on, when non-invasive ventilation will fail in patients with acute hypoxemic respiratory failure. I'm first author on the manuscript, now in preparation for submission. You can read about the work on my research page, and if it interests you, reach out. I'd love to talk about it.

I'm passionate about bringing new AI into clinical care in ways that are responsible, safe and make it more accurate.

I graduated in Computer Science from MSOE and am applying to PhD programs in biomedical informatics for Fall 2027.

Vamsi Sudersanam

Live synthetic patient monitor

RiskModerate

37%

SpO2Watch

90%

FiO2Stable

42%

A synthetic patient generated in code, updated every second. Risk is a toy function of the vitals for illustration.
Vamsi presenting
Presenting my Capstone project (RL Studio) at the Senior Design Symposium

Skills

{
  "computational_medicine_and_ml": [
    "Time-Series Modeling",
    "Deep Survival Frameworks",
    "Gradient Boosted Trees",
    "Explainable AI (SHAP)",
    "Signal Processing",
    "Reinforcement Learning"
  ],
  "survival_analysis": {
    "frameworks": [
      "Non-Linear Survival Models",
      "AFT & Cox Extensions",
      "DeepHit Estimators",
      "Random Survival Forests"
    ],
    "impact": "NIV Failure Stratification"
  },
  "clinical_informatics": [
    "EHR Data Mining",
    "MIMIC-IV Database",
    "eICU Database",
    "SICdb Database",
    "Cohort Selection Analytics"
  ],
  "research_languages": [
    "Python (PyTorch)",
    "SQL (Cohort Extraction)",
    "R (Bioconductor)",
    "C/C++"
  ],
  "research_engineering": [
    "Cloud Compute (AWS)",
    "Docker Containerization",
    "FastAPI Architectures",
    "Git/GitHub Workflows"
  ],
  "certifications": [
    "AWS Certified AI Prac.",
    "AWS Cloud Prac."
  ]
}
AWS certifications: verify on Credly