Dr. Krishna Chauhan

Selected projects

Unpublished educational project

Real-World Evidence: AI/ML Horizons

As the capstone of an online clinical research training program, I wrote a narrative review on how real-world data and machine learning are used in evidence generation.

Fig. 1 A generic causal diagram of the kind the review discusses, with real-world data sources feeding a model of confounders, treatment and outcome. It shows no data from the review.

What it covers

  • Real-world data sources
  • The FDA Real-World Evidence Program and the EU AI Act
  • Causal inference methods described in the literature, including targeted maximum likelihood estimation (TMLE) and causal forests
  • Natural language processing for cohort identification
  • AI-enhanced pharmacovigilance
  • A literature case study of a published EHR-based machine learning program for suicide risk prediction at Mayo Clinic (2018 to 2024)

The methods named above are topics the review discusses. They are not software I have used.

The program’s letter of 12 August 2025 confirms that my project met its required standards.

What it is not

  • Not original clinical research or analysis of patient data
  • Not a systematic review with a registered protocol
  • Not peer reviewed or published
  • Not an institutional project of AIIMS or Mayo Clinic