User Tools

Site Tools


This is an old revision of the document!

Stanford Medicine Program for Artificial Intelligence in Healthcare

Consider the case of Vera, a 60 year old woman of Asian descent with a history of hypertension and asthma, entering the doctor's office with symptoms consistent with a diagnosis of pneumonia. Her primary care physician must diagnose and treat the acute illness, but also manage risk for chronic diseases such as heart attack, stroke, renal failure, and osteoporosis. Ideally, treatment decisions for Vera are guided by risk stratification to decide if to treat, and evidence based selection of how to treat. However, without accurate risk-stratification to decide who to treat, and when, Vera's care remains reactive and suboptimal. Imagine how Vera’s experience would change if we could predict risks of specific events and take proactive action.

Our efforts

We are working on a set of efforts, which are collectively referred to as the Stanford Medicine Program for Artificial Intelligence in Healthcare, with the mission of bringing AI technologies to the clinic, safely, cost-effectively and ethically. See brochure. The four key components are:

  1. Implementation: We partner with analytics and IT teams at Stanford Hospital to deploy predictive models in care delivery workflows. See our first effort in improving palliative care, and recent coverage in Statnews.
  2. Rethinking Utility: Typically, we evaluate models, quantify potential net-benefit, and realize some of it with real world operational constraints. Clinical utility of making a prediction, and taking actions depends on factors beyond model accuracy, such as lead time offered by the prediction, the existence of a mitigating action, the cost and ease of intervening, the logistics of the intervention, and incentives of the providers. Check out our viewpoints on making machine learning models clinical useful, estimating the hidden deployment cost of predictive models, and bridging the implementation gap of machine learning in healthcare.
  3. Safety, ethics, and health system effects: We map the multiple groups involved in executing a prediction-action pair and study their varying perspectives, positions, stakes, and commitments to pre-empt ethical challenges. We believe that the use of AI can lead to good decisions if we keep human intelligence in the loop. Read our perspective on addressing ethical challenges.
  4. Training and Partnerships: We partner with multiple groups to figure out “what would we do differently” if we had a prediction from a model, and to investigate the pros and cons of using AI to guide care. For example, we examined the accuracy vs coverage trade off in patient facing diagnosis models, and partnered with Google on efforts to enable scalable and accurate deep learning with electronic health records.

If done right, the adoption of efficacious prediction-action pairs can massively improve the ability of a health system to find patients at risk and act early. As part of this effort, we search for clinical situations where the application of risk-stratification and proactive action can provide cost-effective, health system level benefits.

Russ Altman and Nigam Shah taking an in-depth look at the growing influence of “data-driven medicine.”

Keeping the Human in the Loop for Equitable and Fair Use of ML in Healthcare, at AIMiE 2018

Building a Machine Learning Healthcare System, at XLDB, April 30 2018

aihc.1634167073.txt.gz · Last modified: 2021/10/13 16:17 by nigam