Computational Precision Health

Featuring: Dr. Jonathan Kolstad System incentives when considering AI advances in health

“It’s about the human components of practicing everyday.”

CPH faculty & Center for Healthcare Marketplace Innovation’s Jon Kolstad discusses why we need to think about system incentives when considering #AI advances in #health.
#BerkeleyHaas #AI4all #economics

Great both to welcome @Serinachang5.bsky.social to CPH faculty, and to see her honored with a Google Research award to advance a new generation of AI-powered population health models!

Featuring Harvineet Singh "Why did the AUC drop?" Explaining Performance Changes of ML Models across Hospital Sites

#ML models often drop in performance from one #hospital to another. How do we understand reasons for the drop & use them to improve #models? Dr. Harvineet Singh
tells us about a framework that provides an actionable decomposition of performance. buff.ly/Gl8sNW4 #AIforall

Featuring Sylvia Cheng Doubly Robust Policy Learning for Causal Stochastic Interventions Thru Auto-debiased Neural Networks

Sylvia Chen proposes doubly robust policy learning, where targeted maximum likelihood estimation with automatically debiased neural nets to learn the optimal shifts for multidimensional #treatment #variables under a stochastic treatment regime. #causalinference www.youtube.com/watch?v=7DhF...

For #patients with #heart disease, #AI is transforming the #diagnosis process. @amalaa.bsky.social discusses the #data driven techniques that are highlighting the most efficient and affordable solutions for individuals.
#ozempic #AIforall #research #hearthealthy

Big congrats to CPH grad students, Radhika Bhalerao for receiving the Julius R. and Patricia A. Krevans Fellowship and Adú Matory, who has been awarded the UCSF Graduate Research Mentorship Fellowship! We’re so proud to celebrate this achievement! #CPH #UCSF #UCB

Featuring Michael Nercessian, Computational Precision Health PhD student Mirai-1.5: Assessing Breast Cancer Risk From Any Mammogram

#Breast #cancer screening reduces mortality, but significant controversy exists over optimal screening guidelines. CPH #PhD student, Michael Nercessian, build on Dr. Adam Yala & colleagues' Mirai, a state of the art risk #prediction model. #AI #ML buff.ly/rejSXij

@egnij.bsky.social , @irenetrampoline.bsky.social and colleagues used an #LLM to extract #data on structural determinants of health—and identified its impact on #patients who got #liver transplants. Dr. Ge describes their work. @ucsfhealth.bsky.social
@amcollegegastro.bsky.social

Congratulations to CPH's @amalaa@bsky.social, winner of a Google Research Scholar award for groundbreaking work on use of AI and high-throughput proteomics to predict treatment impact for populations often excluded from clinical trials. @ucsf-bchsi.bsky.social

Diagnosis of Acute Coronary Syndrome using ECG waveforms: A ML framework and benchmark dataset Featuring Alexander Schubert, Computational Precision Health PhD student

Can an open #dataset of #ECGs linked to #cardiac catheterization outcomes help improve #diagnosis of acute coronary syndromes, by allowing a broad set of #researchers to develop new #ML approaches? CPH #Phd student Alex Schubert tells us more! #AIforall buff.ly/q80cj3g

The Designated Emphasis program is looking for students to join the Fall 2025 cohort. Interested students are encourage to apply here: bit.ly/45R68mF
#CPH #DE #Fall25 #admissions #apply

Featuring: Dr. Irene Chen The data addition dilemma: when adding more data helps or hinders model performance

CPH's @irenetrampoline.bsky.social explains the #data addition dilemma, and how adding more doesn't always make a #model better. See this and other quick takes on cutting-edge research on the CPH #YouTube channel:

Prospective students are welcomed to join the information session on Thursday, June 26 at 12:00PST. Come and ask the questions that will help you prepare for the Fall 2026 admissions cycle. Register: bit.ly/43zkZ3M #admissions #cph #fall26 #AI4Halth #infosession

Featuring Alice Tang, PhD Unlocking the Mysteries of Dementia: How Generative AI Helps Decode Patient Symptoms

It's hard to tell different types of #dementia apart, so we need better ways to describe heterogeneity in #symptoms. Using #AI from #clinical notes, we can find distinct patterns that can help diagnosis & treatment. Dr. Alice Tang tells us more! ‪@ucsf-bcshi.bsky.social‬ buff.ly/Gs5nuP2

@draguilera.bsky.social from @ucberkeleyofficial.bsky.social School of Social Welfare brings attention to #health #equity in the digital & computational ecosystem when it comes to physical activity and other related #health conditions. #spanish #personalized #AI4health
youtu.be/sMB9wG2ZcGc

Peter Washington, PhD Computation in Consumer Digital Health: Challenges and Opportunities. CPH Seminar on March 18, 2025.

ICYMI--Prof. Peter Washington talks computation and consumer digital health in the CPH seminar series. Full video on our YouTube channel here. buff.ly/UJpWxcU @UCSFDocIT.bsky,social #digitalhealth #AI4Good #autism #SUD

Featuring Zhongyuan Liang, Computational Precision Health PhD student Revealing Treatment Non‑Adherence Bias in Clinical Machine Learning Using Large Language Models

#ML systems trained on #treatment records often assume patient adherence. Using a #LLM to extract adherence from #clinical notes, ignoring non-adherence can reverse treatment effect estimates and harm model performance. ‪@irenetrampoline.bsky.social‬
‪@amalaa.bsky.social‬

Featuring: Dr. Sandya Subramanian Monitoring unconscious pain: nociception and personalized anesthesia management

How do we measure nociception, and how does this open new ways for #personalized
#anaesthesia management? CPH's Assistant Professor, Dr. Sandya Subramanian tells us how she used non-invasive, wearable sensors and computational methods during #surgery. #pain #neuroscience #AIforall.

.@RuoguFang, PhD, on Digital Twins and Mulitmodal 3D Foundation Models for Brain Health. Watch this and other CPH seminars on our YouTube channel! Full video of Dr. Fang's talk here: buff.ly/KrybBvA #digitaltwin @labfang

Featuring: Dr. Julian Hong SHIELD-RT: reducing acute care for cancer patients through machine learning

A clinical trial showed that #machinelearning accurately identified #cancer patients at highest risk during radiotherapy, helping prevent rehospitalization. Analysis in @ai.nejm.org showed it was also cost effective! @julianhong.bsky.social tells us more. #AI4all

Meet Hyelee Kim, one of our amazing CPH students in the Designated Emphasis program! She’s defending her dissertation on May 27 at 11AM PT, diving into data tools and neighborhood insights to tackle chronic disease disparities. Register here: tiny.ucsf.edu/o5KirL

Join us for a CPH PhD info session on 5/15 @ 4pm{register: berkeley.zoom.us/meeting/regi... ) or 5/22 @ 9 AM (register: berkeley.zoom.us/meeting/regi... ) Come learn about CPH! #AI4Health #AI #ML #Healthcare #PhD #CPH

Featuring: Marina Sirota, PhD Leveraging AI and knowledge graphs to predict Alzheimer’s disease onset and risk.

Dr. Marina Sirota on how she & colleagues used the EHR, genetics data & @ucsf-bcshi.bsky.social SPOKE knowledge graph to improve predictive power for #Alzheimer's among women, and detect links with #osteoporosis. Head over to our Youtube Channel for more! www.youtube.com/watch?v=eVoA...

CPH PhD student Li-Ching Chen will be presenting her research poster at our very own CPH Conference during the morning poster session from 11:00–11:40 AM! We're so proud of our students and are excited to see them making waves!#AI4Health #AI #ML #Healthcare #PhD #CPH

Interested in applying to the Computational Precision Health PhD program for Fall 2026? Schedule an info session to learn about the program. Email: biancav@berkeley.edu to schedule a session. #AI4Health #AI #ML #healthcare #PhD #CPH #FutureLeaders

🚨 CPH Designated Emphasis deadline: April 30! Apply now ---> tinyurl.com/cphdesignate...
📌 Must apply at least 1 year before your Qualifying Exam! (see slides 2+3)
#AI4Health #AI #ML #healthcare #PhD #CPH #FutureLeaders #DesignatedEmphasis

Yala lab's @kumarkagrawal.bsky.social introduces Atlas — a new AI model, inspired by need for better cancer detection, that uses multi-scale attention to analyze large images. At 4K resolution, it’s 7× faster than Vision Transformers and 30% more accurate than MambaVision. buff.ly/6wYexRm

🚀 Meet Irene Chen! Irene is an Assistant Professor at UC Berkeley and UCSF in Computational Precision Health working at the intersection of machine learning and healthcare, using data to improve patient outcomes in meaningful ways. #AI #MachineLearning #Healthcare

🎉Introducing the Computational Precision Health Graduate Student Association! Meet PhD students Radhika, Michael, Xinyang from our 2nd cohort, and Hongzhou from our 1st cohort. We can't wait to see what you accomplish! #AI4Health #AI #ML #Healthcare #PhD #CPH

🚀 Kicking off CPH's new series Off the Streets! In our first interview, we catch up with our second cohort PhD students to ask: "In an alternate life, what would your PhD be in?" 🤔
Follow us on Instagram @ucjointcph for more updates and to catch the rest of the series! #AI4Health #healthcare #PhD