Efficiencies and happiness are around the corner! We all find joy in a day that is productive, high quality, and meaningful – yet, many times we feel bound by the ‘clutter’ of the day. Busy tracking things down, documenting, filling out forms, drafting communications, and so on. Artificial intelligence offers a solution to our day to day administrative burdens – streamlines, simplifies, and accelerates our work. Application of generative AI tools in health is really a game changer!

Biomedical informatics is harnessing new opportunities using artificial intelligence to address the real problems of diagnostic accuracy, efficiency of information retrieval, documentation burden, and coding. Yesterday, in primary care clinic, I witnessed first-hand with wonder how ambient scribe revolutionized my day. In trying to explain this I would compare the experience to the difference in working on typewriters from the mid-20th century, the kind I learned on and modern computer keyboards paired with word processing. Typewriters were saddled by usability challenges --heavy strokes, difficult to fix, hard to line up the page, so many issues. Today, that is how ‘EMR records’ feel. There is a lot of time moving around the cursor, backing up, forward, thinking what to say and so on. Ambient AI is a revolution to that, seamlessly recording all the stuff in the background and then plugging into the note for a final review.

Last Monday, I self-declared ‘That’s it! I’m never going back!’ One day and I can see the change. I drove home and was not tired. I had not realized how much remembering of what ‘Mary’ said happened, when, which foot, how long, how it felt and re-citing in a keyboard like an asynchronous transcriptionist had plagued my mind. I had so much less mental fatigue! Additionally, all the notes were ‘magically’ created and dropped into the record. Unbelievable, I gasped with amazement. All notes done by 6 pm, most messages finished, had lunch, talked to my nurse, and even called a patient~! That was a never event, time opened up and medicine felt like the old days but ‘even better’ as there was incredible information to help improve the depth of care.

This is exactly the kind of transformation that BMI should aim to have on the world, one that transforms experience, improves quality, and provides new paths for the advancement of the work being done today and health’s future discovery.

Amy M. Sitapati, MD

Interim Chief, Division of Biomedical Informatics, UCSD

Interim Chair, Department of Biomedical Informatics, UCSDH

Lawrence S. Friedman Professor of Population Health, Endowed Chair at UCSD

Pronouns: she/her/hers

The UC was recently awarded $15.5M in research grants to tackle big questions in health, AI, agriculture and climate justice. The award will fund multi-center study projects across all UC’s. One of them aims to scale up a proven method for detecting sepsis. The project, called California’s Groundbreaking and Responsible Evaluation of AI Tech for Sepsis Prevention (CALI-GREAT), will assess whether it is technically feasible to implement an AI model called COnformal Multidimensional Prediction of SEpsis Risk (COMPOSER) across three healthcare centers in the University of California system. The COMPOSER was originally developed at UCSD, with DBMI faculty, Shamim Nemati, PhD and Rob El-Kareh, MD among those who co-authored the initial study. As well as former DBMI PhD student, Aaron Boussina, PhD. The model is a life-saving algorithm that scans real-time data from electronic health records, assessing over 150 variables to detect early signs of infection so doctors can begin treatment. By rolling out this algorithm to more emergency departments across California, researchers aim to save 8,000 lives annually. Dr. Nemati will serve as the Principal Investigator for the CALI-GREAT at UCSD Health for the duration of the two-year study. We invite you to read more about it here.

In a response letter to the White House’s Artificial Intelligence Action Plan, DBMI faculty, Karandeep Singh, MD, MMSc, and Christopher Longhurst, MD, along with UC San Diego   Health, and colleagues at UCSF Health advocated for clearer guidelines that help health systems make informed decisions. The letter highlights three key areas to improve health care AI adoption Including: transparency in AI procurement to allow for informed decisions in place of relying on marketing claims, streamlined AI software regulation that will help reduce confusion and ensure safer adoption across the industry, and lifting restrictions on AI-powered translation tools to improve access to care for patients with limited English proficiency. These recommendations aim to create a more predictable, effective AI ecosystem in health care. We invite you to read more here.

DBMI joint-appointed faculty, Pengtao Xie, PhD  was awarded a $1.7 million NIH Maximizing Investigators' Research Award (MIRA) for Early Stage Investigators. The award is a single-PI award supporting early-career investigators over a five-year period. The grant will fund Dr. Xie’s project, Developing Multi-Modal Foundation Models for Sepsis Early Detection. Sepsis is a life-threatening condition characterized by widespread inflammation, organ failure, and high mortality rates. To address this challenge, the project will develop accurate, efficient, and interpretable multi-modal foundation models (FMs) for the early detection of sepsis. The models will be pre-trained on a comprehensive dataset of 29 million patient records, enabling their broad applicability. By making these FMs accessible to clinicians, Dr. Xie’s team aims to facilitate the rapid development of customized in-house tools for early sepsis detection, ultimately reducing mortality and mitigating organ damage.

In a recent publication, DBMI faculty, Jejo Koola, MD, and Rodney Gabriel, MD, Tsung-Ting Kuo, PhD, and Lucila Ohno-Machado, M.D., Ph.D, M.B.A., co-authored a study analyzing a decentralized cross-institutional learning method that would be capable of incorporating horizontally- and vertically-partitioned data while still protecting patients’ privacy and conforming with policies/regulations. The study introduced Distributed Cross-Learning for Equitable Federated models (D-CLEF), which incorporates horizontally- or vertically-partitioned data without disseminating patient-level records, to protect patients’ privacy. The team compared D-CLEF with centralized/siloed/federated learning in horizontal or vertical scenarios. Using data of more than 15,000 patients with COVID-19 from five University of California (UC) Health medical centers, surgical data from UC San Diego, and heart disease data from Edinburgh, UK, D-CLEF performed close to the centralized solution, outperforming the siloed ones, and equivalent to the federated learning counterparts, but with increased synchronization time. It concluded that D-CLEF presents a promising accelerator for healthcare systems to collaborate without submitting their patient data outside their own systems. We invite you to read the full publication here. 

DBMI joint appointed faculty, Ming Tai-Seale, PhD, has been appointed as the site principal investigator for the National Bureau of Economic Research (NBER) Roybal Center for Behavioral Change in Health project: Behavioral Interventions Combining AI/ML and Personalized Medicine to Reduce Cardiovascular Risk. The project aims to fill this gap between digital innovations of AI/ML and evidenced-based research of patient-tailored healthcare tools by evaluating the efficacy and effectiveness of a newly developed AI-based digital health intervention to reduce cardiovascular risk; and by reoptimizing the tool based on personalized medicine principles to account for heterogeneity in treatment effects in order to develop strategies of easy implementation and maximal adoption. The automated intervention platform consists of a remote monitoring system that ingests lifestyle and blood pressure data and builds a personalized machine learning (ML) model to generate tailored lifestyle recommendations most relevant to each patient’s blood pressure. The project's intervention will be conducted at UCSD as an embedded new program within existing patient care workflows and will be offered to its wide and diverse population of patients across the dimensions of gender, age, race/ethnicity, and geographic economic affluence. It will be carried out in a series of analyses that will promote the intervention from its current Stage 1 through Stages 3, 4, and 5 of the NIH Stage Model for Behavioral Intervention Development. For more information, we invite you to read more here.  

In a recent publication, DBMI predoc trainee, Wilfredo Gonzalez-Rivera along with faculty mentors Matteo D’Antonio, PhD and Melissa Gymrek, PhD, co-authored a study that found  Genome-wide association studies (GWAS) on creatinine and cystatin C levels in healthy individuals reveal both nonoverlapping genetic loci impacting their metabolism as well as overlapping genetic loci associated with kidney function; whereas GWAS on elevated levels of these biomarkers uncover novel loci primarily associated with kidney function in  chronic kidney disease  patients. We invite you to read the full publication here.

DBMI predoc trainee, Wilfredo Gonzalez-Rivera shared some of his Spring highlights. So far this Spring he:

1. Has been reappointed as an ARCS fellow for the 2024-2025 cohort.
2. Gave two presentations for the Pathways to STEM program titled: "How to survive and thrive during your graduate school experience" and "Learn how to vet programs, weigh your decisions, and select the right program for you."

We want to take time to acknowledge Wilfredo for all his achievements and notable strides toward his career in bioinformatics. We thank you for sharing!

We want to take time to celebrate the following students for successfully defending their dissertations! 

Caitlin Guccione
“Disentangling the cancer microbiome: overcoming host contamination to model population dynamics and advance diagnostics”

Did you know that you can give recognition to another UCSD Health employee using the MyPerformance system? Go into MyPerformance (Blink > Personal > UCLearning > MyPerformance > Continuous Feedback > Provide Recognition) to give them a shout-out. The staff member will receive notice of the positive recognition, and their supervisor will be cc’ed. If you received fantastic service from someone in IT, HR, housekeeping, or a colleague, don’t hesitate to use the system to thank them.

Did you know that DBMI has a virtual suggestion box where you can provide anonymous feedback? You can access that here. You can also access the virtual suggestion box on the DBMI website – scroll to the bottom of the homepage and click on the feedback link.

Dziegielewski C, Yuan Y, Ma C, Boland BS, Chang JT, Syal G, Vuyyuru SK, Peyrin-Biroulet L, Jairath V, Singh S. IL-23p19 Antagonists vs. Ustekinumab for Treatment of Crohn's Disease: A Meta-analysis of Randomized Controlled Trials. Am J Gastroenterol. 2025 Mar 12. doi: 10.14309/ajg.0000000000003406. Online ahead of print. PMID: 40071763

Abushamma S, Yadete T, Nero N, Falloon K, Parker CE, Abreu MT, Ahuja V, Armuzzi A, Bemelman W, Bruining DH, Deepak P, Dignass A, Dotan I, Feagan BG, Fulmer C, Halfvarson J, Hart A, Holubar SD, Leong RW, Ma C, Magro F, McCurdy JD, Narula N, Panés J, Raine T, Regueiro M, Rogler G, Singh S, Sparrow MP, Spinelli A, Van Koughnett JA, Vuyyuru SK, Solitano V, Yuan Y, Jairath V, Rieder F. Definitions, Diagnosis and Management of Postoperative Recurrence in Crohn's Disease Patients with Permanent Ileostomy - A Systematic Review and Meta-Analysis. J Crohns Colitis. 2025 Mar 10:jjaf041. doi: 10.1093/ecco-jcc/jjaf041. Online ahead of print. PMID: 40066499

Moussa K, Virk JS, Paciotti B, Durbin-Johnson BP, Shantha JG, Tsui E, Sun CQ, Baxter SL, Gore C, Yiu GY. Adherence to Hydroxychloroquine Dosing Guidelines at the University of California. Ophthalmol Retina. 2025 Mar 10:S2468-6530(25)00100-9. doi: 10.1016/j.oret.2025.03.003. Online ahead of print. PMID: 40074061

Kepner W, Yang KH, Dionicio P, Bailey K, Satybaldiyeva N, Moore A, Han BH, Palamar JJ. Prevalence and Correlates of Lifetime Ecstasy/MDMA Use Among Asian American and Pacific Islander Adult Populations in the United States, 2015-2020. J Psychoactive Drugs. 2025 Mar 3:1-9. doi: 10.1080/02791072.2025.2474243. Online ahead of print. PMID: 40033160

Parsons GR, Parvathinathan G, Etemadi A, Liu S, Ross E, Jones WS, Stedman MR, Chang TI. Identifying peripheral artery disease in persons with and without chronic kidney disease from electronic health records.Vasc Med. 2025 Mar 3:1358863X251322182. doi: 10.1177/1358863X251322182. Online ahead of print. PMID: 40028756 No abstract available.

Arseniev-Koehler A, Tai-Seale M, Cené CW, Grunvald E, Sitapati A. Leveraging diagnosis and biometric data from the All of Us Research Program to uncover disparities in obesity diagnosis. Obes Pillars. 2025 Feb 7;13:100165. doi: 10.1016/j.obpill.2025.100165. eCollection 2025 Mar. PMID: 40028616 Free PMC article.

Chilakapati N, Timple L, Pizzi S, Fine JF, Makhija H, Bruce LK, Owens RL, Alden D, Malhotra A, Martin JL, Kamdar BB. Light levels in a modern intensive care unit: Impact of time of year, window directionality, and outdoor light levels. Chronobiol Int. 2025 Feb 24:1-9. doi: 10.1080/07420528.2025.2469885. Online ahead of print. PMID: 39994888

Pekar JE, Moshiri N, Lemey P, Crits-Christoph A, Débarre F, Goldstein SA, Hensel Z, Rambaut A, Worobey M, Holmes EC, Wertheim JO. Recently reported SARS-CoV-2 genomes suggested to be intermediate between the two early main lineages are instead likely derived.Virus Evol. 2025 Feb 22;11(1):veaf008. doi: 10.1093/ve/veaf008. eCollection 2025. PMID: 40040925 Free PMC article.

Lubrano S, Cervantes-Villagrana RD, Faraji F, Ramirez S, Sato K, Adame-Garcia SR, Officer A, Arang N, Rigiracciolo DC, Anguiano Quiroz PY, Martini C, Wang Y, Ferguson FM, Bacchiocchi A, Halaban R, Coma S, Holmen SL, Pachter JA, Aplin AE, Gutkind JS. FAK inhibition combined with the RAF-MEK clamp avutometinib overcomes resistance to targeted and immune therapies in BRAF V600E melanoma. Cancer Cell. 2025 Feb 20:S1535-6108(25)00053-4. doi: 10.1016/j.ccell.2025.02.001. Online ahead of print. PMID: 40020669

Cai CX, Hribar M, Baxter S, Goetz K, Swaminathan SS, Flowers A, Brown EN, Toy B, Xu B, Chen J, Chen A, Wang S, Lee C, Leng T, Ehrlich JR, Barkmeier A, Armbrust KR, Boland MV, Dorr D, Boyce D, Alshammari T, Swerdel J, Suchard MA, Schuemie M, Bu F, Sena AG, Hripcsak G, Nishimura A, Nagy P, Falconer T, DuVall SL, Matheny M, Viernes B, O'Brien W, Zhang L, Martin B, Westlund E, Mathioudakis N, Fan R, Wilcox A, Lai A, Stocking JC, Takkouche S, Lee LH, Xie Y, Humes I, McCoy DB, Adibuzzaman M, Areaux RG Jr, Rojas-Carabali W, Brash J, Lee DA, Weiskopf NG, Mawn L, Agrawal R, Morgan-Cooper H, Desai P, Ryan PB. Semaglutide and Nonarteritic Anterior Ischemic Optic Neuropathy.JAMA Ophthalmol. 2025 Feb 20. doi: 10.1001/jamaophthalmol.2024.6555. Online ahead of print. PMID: 39976940

Arthur TD, Nguyen JP, Henson BA, D'Antonio-Chronowska A, Jaureguy J, Silva N; iPSCORE Consortium; Panopoulos AD, Izpisua Belmonte JC, D'Antonio M, McVicker G, Frazer KA. Multiomic QTL mapping reveals phenotypic complexity of GWAS loci and prioritizes putative causal variants. Cell Genom. 2025 Feb 16:100775. doi: 10.1016/j.xgen.2025.100775. Online ahead of print. PMID: 39986281

Ford JS, Morrison JC, Kyaw M, Hewlett M, Tahir P, Jain S, Nemati S, Malhotra A, Wardi G. The Effect of Severe Sepsis and Septic Shock Management Bundle (SEP-1) Compliance and Implementation on Mortality Among Patients With Sepsis : A Systematic Review. Ann Intern Med. 2025 Feb 18. doi: 10.7326/ANNALS-24-02426. Online ahead of print. PMID: 39961104 Review.

Rountree L, Lin YT, Liu C, Salvatore M, Admon A, Nallamothu BK, Singh K, Basu A, Bu F, Mukherjee B. Reporting of Fairness Metrics in Clinical Risk Prediction Models: A Call for Change to Ensure Equitable Precision Health Benefits for All. Online J Public Health Inform. 2025 Feb 17. doi: 10.2196/66598. Online ahead of print. PMID: 39962044

Syal G, Barnes E, Raffals L, Al Kazzi E, Haydek J, Agrawal M, Singh S. Correction to: Medical Therapies for Prevention and Treatment of Inflammatory Pouch Disorders-A Systematic Review and Meta-Analysis. Am J Gastroenterol. 2025 Feb 17. doi: 10.14309/ajg.0000000000003285. Online ahead of print. PMID: 39950961 No abstract available.

Alina Arseniev-Koehler, Ming Tai-Seale, Crystal W. Cené, Eduardo Grunvald, Amy Sitapati,
Leveraging diagnosis and biometric data from the All of Us Research Program to uncover disparities in obesity diagnosis, Obesity Pillars, Volume 13, 2025, 100165, ISSN 2667-3681,

D'Antonio M, Arthur TD, Gonzalez Rivera WG, Wu X, Nguyen JP, Gymrek M, Woo-Yeong P, Frazer KA. Genetic analysis of elevated levels of creatinine and cystatin C biomarkers reveals novel genetic loci associated with kidney function.  Hum Mol Genet. 2025 Feb 10:ddaf018. doi: 10.1093/hmg/ddaf018. Online ahead of print.  PMID: 39927731

Owsley C, Matthies DS, McGwin G, Edberg JC, Baxter SL, Zangwill LM, Owen JP, Lee CS; AI-READI Consortium. Cross-sectional design and protocol for Artificial Intelligence Ready and Equitable Atlas for Diabetes Insights (AI-READI). BMJ Open. 2025 Feb 6;15(2):e097449. doi: 10.1136/bmjopen-2024-097449. PMID: 39915016

Kuo TT, Gabriel RA, Koola J, Schooley RT, Ohno-Machado L. Distributed cross-learning for equitable federated models - privacy-preserving prediction on data from five California hospitals. Nat Commun. 2025 Feb 5;16(1):1371. doi: 10.1038/s41467-025-56510-9. PMID: 39910076 Free PMC article.

Baxter SL. Precision Population-Based Screening for Glaucoma. JAMA Netw Open. 2025 Feb 3;8(2):e2457849. doi: 10.1001/jamanetworkopen.2024.57849. PMID: 39913142 No abstract available.

Unsubscribe

This email was sent to nherbst@ucsd.edu.

To continue receiving our emails, add us to your address book.

9500 Gilman Drive | La Jolla, CA 92093 US

Copyright 2024 UCSD DBMI. All rights reserved.