som-shahlab

ehrshot-benchmark

A benchmark for few-shot evaluation of foundation models for electronic health records (EHRs)

femr

FEMR (Framework for Electronic Medical Records) provides tooling for large-scale, self-supervised learning using electronic health records

mimic_tutorial

meds_reader

meds_reader_paper_code

INSPECT_public

INSPECT dataset/benchmark paper, accepted by NeurIPS 2023

Clinfo.AI

This is Clinfo.AI Demo Instruction

aplusml

APLUS ML = A Python Library for Usefulness Simulations of Machine Learning models

slurmtool

SLURM helper

gpt4usmle

Tools for automated evaluation and generation of USMLE-style questions

shahlab-website

Shah Lab Jekyll website

RATE-experiments

Code for reproducing the analyses that appear in "Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects"

inspect-website

medalign-website

Landing page and website for the MedAlign project

clinical_trial_patient_matching

Zero-shot clinical trial matching with LLMs

data-align

Clinical text summarization by adapting large language models

ehrshot-website

Public website for EHRSHOT

motor_tutorial

Tutorial repository for the official MOTOR release

motor_code_release

temp_ds_shift_robustness

clinical_trial_patient_matching-website

Public facing website for zero-shot CT patient matching paper

pancreatic_cancer_tutorial

medalign

MedAlign is a clinician-generated dataset for instruction following with electronic medical records.

pad

A repository for building and testing a simple PAD model

survlearners

Metalearners for Survival Data

manifest

Prompt programming with FMs.

TrialProbe

CLMBR-contrastive-learning

constdb-python

ConstDB is a very simple read-only embedded key-value database

ehr_ml

Code for doing machine learning with various EHRs