UC Berkeley CPH 100 | Foundations of Computational Precision Health

Teaching Staff

Xujing (Crystal) Zhang,
Reader

Welcome!

In this upper-division undergraduate course, students will build expertise in developing machine-learning tools to address challenges in health care. The course emphasizes both “how to formulate computational problems”, and “how to develop solutions”. This requires strong foundations in health care and machine learning methods. On the health side, we’ll get clinical guest lectures exploring diverse challenges, with perspectives from the front lines (primary and specialty) care, the diagnosis workflow, treatment selection, and post-treatment follow-ups. On the machine learning side, the course will cover machine learning and deep learning foundations, state-of-the-art neural networks, and then advanced research topics. The course will emphasize rigorous evaluation, algorithmic bias, deployment, and auditing.

Class Schedule

Note: tentative schedule is subject to change.

Lecture Number	Date	Topic and Readings	Lecturer
1	Aug 28	Introduction to CPH [slides] Artificial intelligence in healthcare (Yu et al, 2018) Do no harm: a roadmap for responsible machine learning for health care (Wiens et al, 2019)	Prof. Irene Chen
2	Aug 28	Case Study: AI and Radiology [slides] CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning (Rajpurkar et al, 2017) Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: a cross-sectional study (Zech et al, 2018)	Prof. Irene Chen
3	Sep 4	What makes healthcare unique? Machine learning in medicine (Rajkomar et al, 2019) A review of challenges and opportunities in machine learning for health (Ghassemi et al, 2020)	Prof. Irene Chen
4	Sep 4	Real World Applications of Generative AI in Liver Diseases and Transplantation Development of a liver disease-specific large language model chat interface using retrieval-augmented generation (Ge et al, 2023) Hepatology e-consult responses generated by artificial intelligence demonstrate accuracy but require human oversight (Huang et al, 2026)	Dr. Jin Ge (UCSF)
5	Sep 11	Problem Set 1 Quiz	---
6	Sep 11	ML Foundations: Risk Stratification and Supervised Learning Intelligible models for healthcare: predicting pneumonia risk and hospital 30-day readmission (Caruana et al, 2015) Dissecting racial bias in an algorithm used to manage the health of populations (Obermeyer et al, 2019)	Prof. Irene Chen
7	Sep 18	ML Foundations: LLMs (diagnosis, Q&A, treatment) Large language models encode clinical knowledge (Singhal et al, 2023) Do We Still Need Clinical Language Models? (Lehman et al, 2023)	Prof. Irene Chen
8	Sep 18	Clinical Decisions with LLMs	Dr. Travis Zack (UCSF, OpenEvidence)
9	Sep 25	ML Foundations: Vision Models ImageNet classification with deep convolutional neural networks (Krizhevsky et al, 2012) AI for radiographic COVID-19 detection selects shortcuts over signal (DeGrave et al, 2021)	Prof. Irene Chen
10	Sep 25	TBD	Zhongyuan Liang (UC Berkeley and UCSF)
11	Oct 2	ML Foundations: Causal Inference Estimating individual treatment effect: generalization bounds and algorithms (Shalit et al, 2016) Causal Inference: What If (Hernán and Robins, 2020)	Prof. Irene Chen
12	Oct 2	Patient-facing LLM Usage HealthBench: Evaluating Large Language Models Towards Improved Human Health (Arora et al, 2025) ChatGPT Health performance in a structured test of triage recommendations (Ramaswamy et al, 2026)	Prof. Monica Agrawal (Duke)
13	Oct 9	Problem Set 2 Quiz	---
14	Oct 9	Evaluation and Benchmarking The evaluation illusion of large language models in medicine (Agrawal et al, 2025) Are We Learning Yet? A Meta Review of Evaluation Failures Across Machine Learning (Liao et al, 2021)	Prof. Irene Chen
15	Oct 16	Distribution Shift The Clinician and Dataset Shift in Artificial Intelligence (Finlayson et al, 2021) WILDS: A Benchmark of in-the-Wild Distribution Shifts (Koh et al, 2021)	Prof. Irene Chen
16	Oct 16	AI for Pediatrics Algorithmic fairness in artificial intelligence for medicine and healthcare (Chen et al, 2023) Underrepresentation of children in public medical imaging datasets (Hua et al, 2026)	Stanley Hua (UC Berkeley and UCSF)
17	Oct 23	Multimodal Multimodal biomedical AI (Acosta et al, 2022) Expert-level detection of pathologies from unannotated chest X-ray images via self-supervised learning (Tiu et al, 2022)	Prof. Irene Chen
18	Oct 23	Deep Dive into the Complexity of the Healthcare System	Prof. Ida Sim (UCSF)
19	Oct 30	Problem Set 3 Quiz	---
20	Oct 30	Interpretability Towards A Rigorous Science of Interpretable Machine Learning (Doshi-Velez and Kim, 2017) “Why Should I Trust You?”: Explaining the Predictions of Any Classifier (Ribeiro et al, 2016)	Prof. Irene Chen
21	Nov 6	Data Quality Biases in electronic health record data due to processes within the healthcare system: retrospective observational study (Agniel et al, 2018) Access to care improves EHR reliability and clinical risk prediction model performance (Zink et al, 2025)	Prof. Irene Chen
22	Nov 6	AI for Personalized Cancer Care Multi-Institutional Validation of a Mammography-Based Breast Cancer Risk Model (Yala et al, 2021) Pillar-0: A New Frontier for Radiology Foundation Models (Agrawal et al, 2025)	Prof. Adam Yala (UC Berkeley and UCSF, Voio)
23	Nov 13	Regulation and Monitoring California Assembly Bill 2575 External Validation of a Widely Implemented Proprietary Sepsis Prediction Model in Hospitalized Patients (Wong et al, 2021)	Prof. Irene Chen
24	Nov 13	Causal Inference for Health Decision-Making	Prof. Maya Petersen (UC Berkeley)
25	Nov 20	Ethics and Bias Ethical Machine Learning in Healthcare (Chen et al, 2021) Disparities in dermatology AI performance on a diverse, curated clinical image set (Daneshjou et al, 2022)	Prof. Irene Chen
26	Nov 20	Physiology-Informed AI	Prof. Sandya Subramanian (UC Berkeley and UCSF)
27	Dec 4	Poster Session	---
28	Dec 4	Poster Session	---

Deadlines

Week	Deadline	Date	Time
2	Problem Set 1 Released	Mon, Aug 31
3	Problem Set 1 Due	Fri, Sep 11	10:00 AM PT
4	Problem Set 2 Released	Mon, Sep 14
7	Problem Set 2 Due	Fri, Oct 9	10:00 AM PT
8	Problem Set 3 Released	Mon, Oct 12
10	Problem Set 3 Due	Fri, Oct 30	10:00 AM PT
11	Final Project Progress Report Due	Fri, Nov 6	10:00 AM PT
15	Final Project Due	Thur, Dec 3	10:00 AM PT

Overview

Course Info

Location: Gateway B1022, UC Berkeley
Timing: Fridays, 10:00 AM – 12:00 PM Pacific Time
Contact: Students should check bCourses and the course GitHub for announcements, assignments, and updates. For course-related questions, please post on Ed Discussion. For private matters, you may email the instructor directly at iychen@berkeley.edu.

Office Hours

Instructor: Thursdays 9:00 AM - 10:00 AM, Gateway 1110A
Additional office hours will be announced on bCourses and in class.

Prerequisites

DATA C100 and DATA C140 (or equivalent background in machine learning and data science).
Students should be comfortable with linear algebra, probability, Python, and introductory ML concepts.

Academic Accommodations

From UC Berkeley’s Academic Accommodations Hub: UC Berkeley is committed to creating a learning environment that meets the needs of its diverse student body. Students who anticipate or experience barriers to learning are encouraged to contact the Disabled Students’ Program (DSP) to determine authorized academic accommodations. The DSP is located at 260 César E. Chávez Student Center, phone (510) 642-0518, URL https://dsp.berkeley.edu.
If you already have a DSP accommodation letter, please share it with the instructors as early as possible so that we can work with you and DSP to ensure appropriate arrangements are in place.

Well-Being, Stress Management, & Mental Health

If you are experiencing personal, academic, or emotional difficulties, you are encouraged to reach out to Counseling and Psychological Services (CAPS) through University Health Services at the Tang Center. CAPS provides confidential counseling, crisis intervention, and mental-health resources for Berkeley students. Appointments can be made by calling (510) 642-9494 or visiting the CAPS website.

Final Project

Details

Write-up Deadline: Dec 3, 10:00 AM PT
Group Size: 2-3 people per group
Scope: Projects can be related to any topic covered in class or within a generous interpretation of machine learning and healthcare. Ambition of project match the problem sets. As some examples, you may choose to analyze a novel dataset, implement a method taught in class, evaluate or examine current AI models used in healthcare, or perform a literature review of why AI models are or are not working for a medical specialty.
Grading Criteria: Projects will be graded on clarity, originality, significance, and quality.
Poster Presentation:The last class of the semester (Dec 4) will be a poster presentation where students will showcase their work.

Final Project Progress Report

Deadline: Nov 6, 10:00 AM PT
Format: Please submit one slide containing five sections, including research problem, impact, methods, preliminary results, and next steps.

CPH 100 / Data C146: Foundations of Computational Precision Health

UC Berkeley / Fall 2026