medical ai phd

The AIMI Center

The Stanford Center for Artificial Intelligence in Medicine and Imaging (AIMI) was established in 2018 to responsibly innovate and implement advanced AI methods and applications to enhance health for all.

Director's Welcome

Back in 2017, I tweeted “radiologists who use AI will replace radiologists who don’t.”  The tweet has taken on a life of its own, perhaps because it has a double meaning.

AIMI Impact Report

This report highlights the progress of our research, education, and policy programs during our first five years (2018-2023)

AI+HEALTH 2024

Join us online December 10-11, for Stanford's largest annual convening of AI+HEALTH experts and learners

AIMI Dataset Index

AIMI has launched a community-driven resource of health AI datasets for machine learning in healthcare as part of our vision of catalyze sharing well curated, de-identified clinical data sets

AIMI NextGen Tech Talk: Nishith Khandwala

Watch Episode 5 of AIMI's NextGen Tech Talk series here . This webinar series is tailored towards high school students who are eager to explore the field of AI and health. 

AIMI Datasets for Research & Commercial Use

The AIMI Center is helping to catalyze outstanding open science by publicly releasing 20+ AI-ready clinical data sets (many with code and AI models) for research and commercial use.

Upcoming Events

Ibiis-aimi seminar: hugo aerts, phd, save the date: ai + health 2024.

• Departments & Institutes
• Research Cores
• Laboratories
• Clinical Trials
• Other Research Programs
• Office of Research Administration
• Technology & Innovations
• Clinical & Translational Research Center
• Electronic Research Notebook
• News & Breakthroughs
• Graduate Medical Education
• Graduate School of Biomedical Sciences
• Continuing Medical Education
• Professional Training Programs
• Women's Guild Simulation Center
• Center for the Arts and Humanities in Medicine
• Medical Library
• Campus Life
• Office of the Dean
• Academic Calendar
• Back to PhD Program in Health Artificial Intelligence

Training & Curriculum

• Application Information
• Faculty & Administration

The curriculum of the PhD in Health AI emphasizes an active learning approach that will be used to teach six required courses, including AI, ethical AI, machine learning, natural language processing, clinical applications of AI and biomedical informatics. Students will gain healthcare experience through clinical rotations, clinical collaborations and access to clinical data from the electronic health record.

Cedars-Sinai 's PhD in Health AI is pending WSCUC accreditation.

Program Overview

This program will provide doctoral students with the knowledge and practical experience to develop, evaluate and apply cutting-edge AI algorithms and methods for improving patient care. We utilize a hands-on, active learning approach to teaching that reinforces AI concepts through short projects completed during class. Students will be exposed to hospital rotations to better understand how AI might be used in healthcare. Students will work with clinical collaborators and will have access to data from the electronic health record. Graduates of the program will be positioned to improve healthcare and patient outcomes through the rigorous development and deployment of AI algorithms and software.

View the Program Schedule (PDF)

This course provides a comprehensive exploration of the intersection between artificial intelligence and biomedical sciences, aimed at equipping AI (Artificial Intelligence) and computer science professionals with the requisite clinical knowledge to develop and apply AI algorithms in healthcare. Students will delve into the principles of clinical medicine, examine case studies of AI applications in clinical settings, and engage in the development of AI solutions to address medical challenges. Key topics include feature engineering, data preprocessing, dimensionality reduction, explainable AI, and setting up appropriate evaluation methods for domain-specific problems. The course will also address the ethical, regulatory, and practical considerations of implementing AI in healthcare, including dealing with bias and fairness, preparing students to contribute to the advancement of AI-driven clinical and translational research. 

Imaging AI seeks to advance innovative diagnostic and prognostic algorithms in Radiology and Pathology, equipping you with the competencies to develop and validate AI / deep learning workflows for biomedical image analysis and translate theoretical knowledge into clinical solutions. Through hands-on learning, you will master AI-driven image analysis for disease biomarker identification, diagnostic and prognostic modeling, and progression tracking and monitoring. The course will also emphasize appropriate statistical validation (e.g., multilevel regression modeling) and evaluation of the AI models. Special topics include graph-based methods, spatial multimodal analysis, and user interface design.  

Designing and inventing new biomedical devices and wearables in any area of healthcare requires a comprehensive clinical and physics-based understanding of the human body integrated with the art of engineering design. This course focuses on developing devices and wearables for the neuromuscular system. We will start with a brief introduction to the human anatomy, the neuromuscular system, and the behavior of different types of signals, such as electrical, acoustic, and optical waves, that can be used to understand human tissue condition and behavior, along with examples of the current state of the art. We will then delve deep into 2 to 3 clinical problems medical providers face in musculoskeletal medicine, where biomedical devices could improve screening, diagnosis, or assessment and, therefore, improve clinical care.  We will focus on pathophysiology, the clinical workflow, and constraints inherent in human subject studies, and the engineering limitations, before exploring potential pathways to develop a biomedical device or wearable. The second part of the class will focus on developing a working prototype of a wearable device. You will gain hands-on experience with prototyping tools and devices such as high-end 3D printers, Computer-Aided Design (3D design), programming microcontrollers and sensors, and transducers. 

AI algorithms for personalized medicine require multi-modal data to capture the interactions between our genes and the environment in order to understand disease conditions. This course will cover algorithms and methods used to analyze complex biomedical data, including DNA sequences, genetics, epigenetics, proteomics, single-cell genomics, and molecular image data. A mentored term project will provide you with hands-on experience for carrying out independent research, highlighting the importance of interdisciplinary collaborations and the value of incorporating diverse perspectives in research.

Computational Biomedicine, a rapidly growing discipline at the intersection of biology, medicine, statistics, and computer science, offers exciting opportunities for real-world impact in healthcare. In the dynamic landscape of biomedical research, where data plays an increasingly crucial role, understanding scientific inquiries and developing quantitative skills for data analysis and interpretation are essential. This course, serving as an introduction to Computational Biomedicine, will focus on modeling health and disease systems. We will cover computational modeling principles, apply modeling techniques, analyze model performance and limitations, and explore innovative computational frameworks, algorithms, and architectures. These tools are not just theoretical concepts, but practical solutions to address unmet needs and open problems in biomedical research and clinical practice. The course will use project-based and hands-on learning experiences to enhance students’ understanding and application of the subject matter, preparing them for the exciting challenges of the field. 

This course explores the ethical challenges and considerations involved in developing and deploying artificial intelligence (AI) systems in healthcare and public health contexts, including responsible use, patient consent, bias of AI algorithms, and fairness in models. You will critically examine predictive models and AI applications used for making important health decisions, addressing factors that lead to trustworthy AI. Through a reverse classroom approach, students will engage in active learning activities to analyze the potential for bias, risk, and social inequity in AI systems. The course will emphasize project-based learning, allowing students to learn and apply ethical AI principles and practices to real-world healthcare scenarios. 

This course provides comprehensive coverage in machine learning, covering both theoretical foundations and practical applications. Students will learn concepts, algorithms, and techniques used in machine learning. Emphasis will be placed on real-world applications, particularly in biological and clinical sciences. Students will gain hands-on experience through practical exercises and projects and learn the theory and practice of machine learning from a variety of perspectives. Topics include supervised learning (classification, regression); unsupervised learning (clustering, dimensionality reduction); reinforcement learning; and computational learning theory. 

The significant advance of natural language processing (NLP) approaches in the last few years, with the advent of chatbots that seem to hold conversations and even express ‘chain-of-thought’ reasoning behind their answers, sets the bar high for what these systems can accomplish within the healthcare setting, facilitating patient-physician interaction and improving diagnostic accuracy. This course will take a hands-on approach to explore the boundaries of NLP and Artificial Intelligence, enabling deep understanding of cutting-edge technologies that could help address the hardest problems currently faced by clinicians and patients.

Clinical Rotations

All students are required to fulfill a minimum of 20 hours of clinical rotations across one or more specialties. During these rotations, students will shadow doctors during patient encounters and observe interactions, utilizing electronic health records and decision-support tools.

Research Rotations

All students will complete three rotations during the first year in candidate dissertation research labs. This process will culminate in identifying a willing research mentor to supervise a dissertation research project.

Dissertation Research

Students are expected to conduct a dissertation research project that generates new knowledge at the intersection of AI and healthcare. The project will facilitate collaboration between AI experts and clinicians, culminating in several peer-reviewed publications.

Have Questions or Need Help?

If you have questions or wish to learn more about the PhD program in Health AI, call us or send a message.

Artificial Intelligence–Generated Draft Replies to Patient Inbox Messages Read More

Bridging the Gap Between Community, Academia and Industry Learn More

Sowing the Future of Primary Care with the AI/ML Bootcamp at 2023 NAPCRG Annual Meeting Learn More

Unleashing the power of AI in Primary Care Read More

Code to Bedside: May 2024 Newsletter Read More

Our work covers a broad range of aims centered on the development and integration of artificial intelligence (AI) technologies that solve important, practical problems for patients, providers and health systems. 

We work with clinical, operational, and technical teams to advance the development of clinically relevant models, leveraging quality improvement, implementation science, design thinking, and traditional research methods.

Director's Message

Read a message from HEA 3 RT's Director Dr. Steven Lin and learn about the mission and vision of our program 

Work With Us

Learn about ways to engage with HEA 3 RT, whether you're a prospective partner from industry, academia, or the non-profit world

Explore our projects studying the implementation of AI technologies in healthcare to support patients, providers, and health systems

Recent News & Publications

Pioneering Patient Care with AI

JAMA Internal Medicine, March 2024.

HEA3RT’s collaboration on the implementation of EPIC's clinical deterioration index (CDI) at Stanford Health Care marks the first-ever demonstration of improved patient outcomes using the CDI tool. Click here to explore how these findings are setting new standards in patient care, ensuring safer hospital environments.

Towards Universal AI Code of Conduct

National Academy of Medicine Perspectives, April 2024.

Artificial Intelligence in healthcare holds remarkable potential but there remains a need for unified standards for its safe and ethical applications. Check out HEA3RT’s latest  commentary paper  with the National Academy of Medicine (NAM) and join the conversation shaping the future of national policies on health AI.