Shah Lab

ace

Anonymous (anonymous@undisclosed.example.com) — 2020-06-14T13:23:04+00:00

The Advanced Cohort Engine (ACE) Welcome to ACE, the search engine that powers the Informatics Consultation Service. See a brief video of ACE in action below. Note: the video refers to ACE by its older name based on the Advanced Temporal Language Aided Search (ATLAS). ----------

ai-care

Anonymous (anonymous@undisclosed.example.com) — 2024-09-14T17:07:04+00:00

AI-Care: AI for Cost-effective, Accurate, Reliable, and Ethical Solutions AI-Care a training program in using AI for Cost-effective, Accurate, Reliable, and Ethical Solutions at Stanford Healthcare where students enrolled in a graduate PhD program at Stanford University can participate in conducting discovery, feasibility assessments, building proof of concept applications, monitoring deployed solutions, or conducting prospective evaluations in the

ai-for-physicians

Anonymous (anonymous@undisclosed.example.com) — 2023-12-14T13:58:54+00:00

Our campus has several learning opportunities in Clinical Informatics and Data Science for interested physicians. Below are some of the training programs as well as seminar series that may be of interest. Training programs * Clinical Informatics Fellowship, is a American Board of Medical Specialties approved Clinical Informatics board-eligible subspecialty fellowship sponsored by the American Board of Preventive Medicine. This program focuses on the understanding, integration and application…

ai-voice

Anonymous (anonymous@undisclosed.example.com) — 2024-06-11T15:04:20+00:00

AI-Values Oriented Implementation and Context Evaluation - This is a survey instrument to elicit end-user values from stakeholders impacted by AI in healthcare.

aihc-programs

Anonymous (anonymous@undisclosed.example.com) — 2025-12-12T10:30:38+00:00

There are multiple efforts on campus that focus on AI, Data Science and Digital innovation in Healthcare. Below is a listing of the ones I know. Efforts in SOM * Center for Artificial Intelligence in Medicine & Imaging (AIMI), which aims to responsibly innovate and implement advanced AI methods and applications to enhance health for all.

biomedin215-2011

Anonymous (anonymous@undisclosed.example.com) — 2012-08-10T10:57:53+00:00

BIOMEDIN 215 DATA DRIVEN MEDICINE With the spread of electronic health records, increasingly large data repositories of clinical and other patient derived data are being built. These databases are large and difficult for any one specialist to analyze. To find the hidden associations within such data, we review methods for large-scale

biomedin215

Anonymous (anonymous@undisclosed.example.com) — 2025-09-02T13:53:04+00:00

In 2025 BIOMEDIN 215 course is renamed to BMDS 215 New page is at

blogs-and-media

Anonymous (anonymous@undisclosed.example.com) — 2024-05-21T10:14:01+00:00

Blogs at the Human Centered AI Institute * Does every model need to be explainable? * Do healthcare models need to be generalizable? * What do we do when healthcare algorithmic fixes fail? * How do we ensure the Fairness of Algorithms that Predict Patient Disease Risk? * Policy Brief Promoting Algorithmic Fairness in Clinical Risk Prediction * How can we make sure healthcare models are useful? * Are medical AI tools delivering on what they promise? * What should healthcare exe…

bmds215

Anonymous (anonymous@undisclosed.example.com) — 2025-08-29T13:38:45+00:00

BMDS 215: Data Science for Medicine (Aut 2025) The Autumn 2025 class will be offered in person. – 2025 Teaching team: Nigam Shah, Suhana Bedi, Joshua Lazaro, Maya Drusinsky, Ivan Lopez This course is designed to prepare you to pose and answer meaningful clinical questions using routinely collected healthcare data. The practical skills you will learn in this class will be applicable to any task involving data manipulation and analysis. The course will use real, de-identified, large size patien…

chatehr

Anonymous (anonymous@undisclosed.example.com) — 2026-04-09T12:58:15+00:00

ChatEHR is a set of capabilities developed at Stanford Medicine, for using a set of Large Language Models (LLMs) with a single patient's longitudinal medical chart in context. There are two ways to use the ChatEHR capability: * via a fixed prompt automation, which applies an existing set of criteria to a patient's record and returns a response based on that patient's record. Examples include decisions about transfer eligibility, referrals to certain services, and creating summaries of a hospi…

clmbr

Anonymous (anonymous@undisclosed.example.com) — 2023-12-13T10:47:27+00:00

CLMBR (clinical language modeling based representations) This is a 141 million parameter autoregressive foundation model pretrained on 2.57 million deidentified EHRs from Stanford Medicine. This is the model from ( Wornow et al. 2023), and is based on the CLMBR architecture originally described in (

courses-to-consider

Anonymous (anonymous@undisclosed.example.com) — 2020-12-11T17:04:15+00:00

* BIOMEDIN 432: Analysis of Costs, Risks, and Benefits of Health Care (HRP 392) * BIOMEDIN 233: Intermediate Biostatistics: Analysis of Discrete Data (HRP 261, STATS 261) * HRP 262: Intermediate Biostatistics: Regression, Prediction, Survival Analysis (STATS 262)

covid19

Anonymous (anonymous@undisclosed.example.com) — 2020-12-10T11:25:13+00:00

There are three efforts we are working on. The first effort is to continuously profile the patients screened for SARS-CoV-2 in our health system. Using anonymized data, we watch trends in presenting symptoms of patients, test positivity rates, the age distribution of positive cases and hospitalization rates as well as monitor length of stay. This work as yielded a few insights already such as:

dsatshc

Anonymous (anonymous@undisclosed.example.com) — 2025-06-29T16:49:27+00:00

Data Science team at SHC Created in March 2022, we are an interdisciplinary team in Technology and Digital Solutions focused on ensuring Stanford Health Care is a leader in responsible artificial intelligence — from development and implementation to maintenance and optimization. We support Stanford Health Care’s long-term vision to harness artificial intelligence to support and enhance every aspect of health care delivery, AI research and medical education.

dsatshc_intranet

Anonymous (anonymous@undisclosed.example.com) — 2023-08-14T16:10:19+00:00

ehrshot

Anonymous (anonymous@undisclosed.example.com) — 2023-10-30T21:09:37+00:00

EHRSHOT: An EHR Benchmark for Few-Shot Evaluation of Foundation Models While the general machine learning (ML) community has benefited from public datasets, tasks, and models, the progress of ML in healthcare has been hampered by a lack of such shared assets. The success of foundation models creates new challenges for healthcare ML by requiring access to shared pretrained models to validate performance benefits. We help address these challenges through three contributions. First, we publish a n…

ehrshot_license

Anonymous (anonymous@undisclosed.example.com) — 2023-12-10T13:12:48+00:00

The EHRSHOT Credentialed Health Data License “” 1. The LICENSEE will not attempt to identify any individual or institution referenced in EHRSHOT restricted data. 2. The LICENSEE will exercise all reasonable and prudent care to avoid disclosure of the identity of any individual or institution referenced in EHRSHOT restricted data in any publication or other communication.

examples_of_prior_work

Anonymous (anonymous@undisclosed.example.com) — 2024-07-14T15:46:02+00:00

Answering clinical questions * In A ‘Green Button’ For Using Aggregate Patient Data At The Point Of Care we envision a “green button” function within EHRs for clinicians to use aggregate patient data for real time decision making at the point of care. Check out our recently launched Informatics Consult Service that puts this idea in action. * Profiling risk factors for chronic uveitis in juvenile idiopathic arthritis: We report a new association between allergic conditions and chronic u…

femr

Anonymous (anonymous@undisclosed.example.com) — 2023-12-13T10:44:44+00:00

Framework for Electronic Medical Records (FEMR) FEMR is a Python package for manipulating longitudinal EHR data for machine learning, with a focus on supporting the creation of foundation models and verifying their presumed benefits in healthcare. Such a framework is needed given the

foundationmodels

Anonymous (anonymous@undisclosed.example.com) — 2024-03-15T15:04:57+00:00

We believe that this new class of models -- called foundation models -- may lead to more affordable, easily adaptable health AI. In a blog post at HAI we discuss the opportunities foundation models offer in terms of a better paradigm of doing “AI in healthcare.” First, we outline what foundation models are and their relevance to healthcare. Then we highlight what we believe are key opportunities provided by the next generation of medical foundation models, specifically:

furm

Anonymous (anonymous@undisclosed.example.com) — 2024-09-14T17:24:58+00:00

Standing on FURM ground - A framework for evaluating Fair, Useful, and Reliable AI Models in healthcare systems The impact of using artificial intelligence (AI) to guide patient care or operational processes is an interplay of the AI model’s output, the decision-making protocol based on that output, and the capacity of the stakeholders involved to take the necessary subsequent action. To estimate this impact, the Data Science team at Stanford Health Care has developed a mechanism to identify fa…

greenbutton

Anonymous (anonymous@undisclosed.example.com) — 2026-02-15T08:55:23+00:00

Informatics Consultation Service at Stanford The Informatics Consultation service was an IRB approved project to study the use of routinely collected data on millions of individuals to provide on-demand evidence in those situations where good evidence is lacking. We informed clinical care decisions by summarizing “what happened to patients like mine” in the form a report with a descriptive summary of similar patients in Stanford’s clinical data warehouse, treatment choices made, and observed ou…

greenbutton_idea

Anonymous (anonymous@undisclosed.example.com) — 2019-09-26T15:51:49+00:00

The green button idea ---------- Usage model ---------- Steps in building a Green Button Gallego et al, Bringing cohort studies to the bedside: framework for a 'green button' to support clinical decision-making

guide-ai

Anonymous (anonymous@undisclosed.example.com) — 2025-07-28T15:51:12+00:00

Welcome to the placeholder page for the Stanford GUIDE-AI lab, which stands for Guidance for the Use, Implementation, Development, and Evaluation of AI. The work is done under the aegis of the RAISE-Health initiative to freely share progress made at Stanford Medicine with others.

heal-ai

Anonymous (anonymous@undisclosed.example.com) — 2024-06-11T14:59:21+00:00

This is place holder for a Stanford Impact Labs project to build a process for ethical assessment of healthcare AI. Co-funded by PCORI, and HEAL stands for Healthcare Ethical Assessment Lab for Artificial Intelligence (HEAL-AI). The effort is a collaboration with SHC, and the data science team in TDS.

healthadminbench

Anonymous (anonymous@undisclosed.example.com) — 2026-03-26T09:15:16+00:00

HealthAdminBench: Evaluating Computer-Use Agents on Realistic Healthcare Administration Tasks Healthcare administration accounts for $1 trillion of annual spend, representing an ideal application for LLM-based computer-use agents. While clinical applications of LLMs have received significant attention, no benchmark currently exists for evaluating LLMs on end-to-end administrative workflows. To address this gap, we introduce HEALTHADMINBENCH, which consists of four realistic

healthaiweek

Anonymous (anonymous@undisclosed.example.com) — 2025-08-14T10:24:52+00:00

A place holder for a future page of the highly popular health AI week. It will house the summary content of all the parallel events that occur in 2026. The 2025 events were: * RAISE Health Symposium - June 2, 2025 (Monday) * AIMI Symposium - June 3, 2025 (Tuesday)

healthgpt

Anonymous (anonymous@undisclosed.example.com) — 2023-04-30T18:03:15+00:00

Large language models in Healthcare The past year has seen significant advancements in artificial intelligence (AI) for various modalities, such as text, image, and video. Foundation models, which are AI models trained on large, unlabeled datasets and highly adaptable to new applications, are driving these innovations. These new class of models offer opportunities for a better paradigm of doing

inspect

Anonymous (anonymous@undisclosed.example.com) — 2023-11-08T17:27:09+00:00

INSPECT: A Multimodal Dataset for Pulmonary Embolism Diagnosis and Prognosis Synthesizing information from various data sources plays a crucial role in the practice of modern medicine. Current applications of artificial intelligence in medicine often focus on single-modality data due to a lack of publicly available, multimodal medical datasets. To address this limitation, we introduce INSPECT, which contains de-identified longitudinal records from a large cohort of pulmonary embolism (PE) patie…

jobs

Anonymous (anonymous@undisclosed.example.com) — 2024-07-14T15:26:24+00:00

We are currently not recruiting

lab_members

Anonymous (anonymous@undisclosed.example.com) — 2026-01-03T11:08:37+00:00

Lab Members * Nigam Shah (PI) * Alison Callahan (Research Scientist) * Akshay Swaminathan (DBDS, 4th year MD / PhD student) * Suhana Bedi (DBDS, 2nd year PhD student) * Alyssa Unell (CS, 2nd year PhD student) * Bobak Seddighzadeh (CI Fellow) * Brenna Li (Postdoc) * Lionel Jeremiah (Administrative Associate)

lumia

Anonymous (anonymous@undisclosed.example.com) — 2023-12-08T18:04:19+00:00

LUMIA: Language Understanding for Medical Insights and Action

medalign

Anonymous (anonymous@undisclosed.example.com) — 2023-08-31T12:37:46+00:00

MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records The ability of large language models (LLMs) to follow natural language instructions with human-level fluency suggests many opportunities in healthcare to reduce administrative burden and improve quality of care. However, evaluating LLMs on realistic text generation tasks for healthcare remains challenging. Existing question answering datasets for electronic health record (EHR) data fail to capture …

motor

Anonymous (anonymous@undisclosed.example.com) — 2024-03-05T17:37:07+00:00

MOTOR (Many Outcome Time Oriented Representations) MOTOR is a self-supervised, time-to-event (TTE) 143M parameter foundation model which is pretrained on timestamped sequences of events in 55 million electronic health records (EHR) comprising 9 billion clinical events.

nigam_shah

Anonymous (anonymous@undisclosed.example.com) — 2026-04-13T09:08:29+00:00

Dr. Nigam H. Shah, MBBS, PhD. Titles: Professor of Medicine, and of Biomedical Data Science Chief Data Scientist, Stanford Healthcare Associate Dean, School of Medicine Associate Director, Stanford Center for Biomedical Informatics Research Affiliations:

other_talks

Anonymous (anonymous@undisclosed.example.com) — 2020-08-20T12:38:18+00:00

---------- Invited talk at NIPS 2015, Workshop on Machine Learning For Healthcare ---------- Performing an Informatics Consult, at the 2016 Big Data meeting at Stanford. ---------- Building a Machine Learning Healthcare System, Invited talk at the Parc Forum, Summer 2016.

rail

Anonymous (anonymous@undisclosed.example.com) — 2024-05-12T10:55:17+00:00

Responsible AI in Healthcare Our team is focused on bringing AI into clinical use, safely, ethically and cost effectively. Our work is organized in two broad work-streams. Creation and adoption of foundation models in medicine Given the high interest in using large language models (LLMs) in medicine, the

service_details

Anonymous (anonymous@undisclosed.example.com) — 2022-05-11T11:02:56+00:00

* In 2014, we described the vision of a ‘Green Button’ for using aggregate patient data at the bedside in Health Affairs. * In 2015, we outlined the steps in realizing the green button vision, and for bringing cohort studies to the bedside. * In 2016, we built a search engine for patient timelines for rapid phenotyping to build cohorts of

shcds

Anonymous (anonymous@undisclosed.example.com) — 2024-09-15T20:24:16+00:00

sidebar

Anonymous (anonymous@undisclosed.example.com) — 2011-01-12T17:09:29+00:00

start

Anonymous (anonymous@undisclosed.example.com) — 2025-09-06T10:46:53+00:00

We are a group of doctors, engineers, informatics professionals and students focused on enabling better care using existing health data. We develop novel methods to learn from patient-level health data, answer clinical questions that enable better medical decisions at the point of care, and have an active effort to research safe, ethical, and cost-effective strategies for using predictive models to guide mitigating care actions. Our research group is part of the Department of Medicine at Stanfor…