The Yala lab develops machine learning models for personalized care and translates them to clinical practice. It focuses on designing modeling approaches that are robust to data-generation biases, offer mechanisms for clinical deployment and can adapt to diverse clinical requirements.
This project aims to improve the throughput by automated rapid scanning with minimal input from technicians and improve patient comfortableness with an ECG-free and free-breathing scan procedure with self-navigation powered by an inline AI framework.
Bakar faculty, senior author Yulin Hswen publishes on abortion access post: Dobbs v Jackson Women’s Health Decision.
Read ArticleBakar faculty Hani Goodarzi receives the NIH Director’s Transformative Research Award.
Read ArticleThe NIA NIH highlighted work by BCHSI faculty lead Alice Tang, Sirota Lab and Information Commons team.
Read ArticleDr. Rauschecker’s lab leverages modern AI methodologies to quantify, describe, and understand clinical brain MRIs. This work includes new imaging biomarker discovery, using AI for diagnostics and prognostics, and using AI for better understanding variation in the normal brain.
Program is administered in partnership with the Division of Computing, Data Science, and Society at Berkeley and UCSF BCHSI.
Read ArticleBCHSI faculty gain highest UC Tech award honor for Deidentifying Clinical Notes at Scale.
Read ArticleBCHSI faculty Marina Sirota et al will focus on data sharing, computational drug discovery, leveraging real world data to help prevent preterm birth.
Read ArticleBCHSI faculty Laura van’t Veer & team findings will guide treatment prioritization.
Read ArticleJean Feng, Romain Pirracchio, et al published on Quality Improvement for Clinical AI.
Read ArticleBCHSI Faculty, Andrej Sali, Brian Shoichet, Michael Keiser, are investigators for this QBI Coronavirus Research Group.
Read ArticleAs researchers who depend on data, join us on Monday, May 9 to learn how to deepen your research inquiry.
See DetailsAshish Raj co-led the development of a computational pipeline, Matrix Inversion and Subset Selection.
Read ArticleJimmie Ye, et al, published in Science on how they deployed mux-seq to profile 1.2M immune cells for lupus.
Read ArticleDr. Olshen has helped develop tools in such area as DNA copy number, mutation hotspot detection, and integration of data from multiple genomic assays. He is currently developing biomarkers to predict cancer outcomes in pediatric cancers.
Sharat Israni, et al and collaborators publish on knowledge networks identified by NSC as a core component of AI frameworks.
Read ArticleButte articulated priorities to Energy & Commerce Subcommittee on Health “Future of Biomedicine” Hearing.
Read ArticleUCSF BCHSI data scientist’s research draws attention to an entirely new approach to drug discovery for Alzheimer’s.
Read ArticleFranklin Huang: Providing precision oncology solutions for veterans using data science and focusing on disparities.
Read ArticleA UCSF-Stanford data analysis shows a strong association between taking SSRIs and surviving the virus.
Read ArticleDr. Lazar’s research focuses on biostatistics, data science and precision health. She has secured grant funding as PI from NIH, foundations and the University of California Office of the President, including grants for the HBCU initiative.
Ida Sim, MD, PhD and new UCSF program faculty will be affiliated with the BCHSI.
Read ArticleStuart Gansky, MS, DrPh, co-director, and William Brown III, PhD, DrPH as associate unit director.
Read ArticleMarina Sirota, PhD (co-senior author) published a study using computation to pinpoint an existing drug that may prevent Alzheimer’s Disease.
Read Article
Learn about UCSF’s knowledge network project and other innovative work supported by NSF Convergence Accelerator.
Watch VideoKatie Pollard and Duan Xu are inducted into the prestigious American Institute for Medical and Biological Engineering College of Fellows.
Read ArticleDr. Kornblith is focused on novel diagnostic strategies to enhance the care of injured children. He uses
a modern data science framework to develop accurate, consistent, and interpretable advanced analytic
models for rapid detection of intra-abdominal bleeding using clinical decision rules, computer vision,
and device design.
Dr. Fragiadakis’s research focuses on analyzing immune state in diverse sets of patient cohorts using high-dimensional single-cell technologies, including single-cell sequencing and CyTOF. She uses multi-modal data integration methods to evaluate patient differences and infer broader principles of immune organization.
Dr. Semere is using machine learning and novel computational linguistics techniques to characterize secure message communication between racially/ethnically diverse patients with diabetes, their caregivers, and providers. She is applying this information to develop family-centered technology based communication strategies that promote effective diabetes management.
We focus on understanding how the immune system coordinates responses with an emphasis on tumor immunology. We combine experimental and computational methods to understand how the immune system responds to tumors and to rationally initiate curative immune responses against cancer.
Dr. Feng has developed machine learning and deep learning algorithms that accurately quantify their uncertainty and are appropriate for high dimensional datasets. Her current research is in developing safe machine learning algorithms that learn and recalibrate using streaming Electronic Health Records data.
The objective of this NIAID-funded K01 award is to identify the residual drivers of HIV infection in Fast Track cities. Methods from semi-parametric
statistical modeling, HIV phylogenetics, and minority stress theory will be used to augment surveillance data.
Dr. Sarkar’s research focuses on: (1) Ambulatory patient safety, (2) Digital health innovations to improve the safety and quality of outpatient care, (3) Social media research for behavior change, (4) Safety-net implementation of evidence-based digital health in real-world settings.
Dr. Hswen current research seeks to identify authentic attitudes, feelings and beliefs that influence behaviors and drive populations. Through the collection of unconventional and underground online social networks, Dr. Hswen captures unfiltered conversations to further understand the connections between social experiences and health.
Pathology and laboratory medicine provide 70% of the information in most patient care settings. Therefore, using the best software is the best way to deliver that information to the clinicians on the front lines. I am interested in developing these software solutions by utilizing the best practices from the cutting edge artificial intelligence technologies to traditional methods.
Dani Ushizima explores algorithms and a data analysis pipeline to help accurately distinguish COVID-19 abnormalities in CT scans and chest X-rays.
Read ArticleFacilitated by BCHSI, AWS connected with the laboratory of Charles Chiu to enable two COVID-19 projects.
Read ArticleDr. Ushizima research focuses on computer vision algorithms for monitoring diseases progression while exploring information from multiple imaging techniques. Together with Grinberg’s lab, developed quantitative histological analysis of whole human brains using multiple instruments.
UC Health will distribute daily updates about SARS-CoV-2 testing volume, the # of positive tests & age distribution of confirmed cases gathered from its 5 medical centers.
Read ArticleButte lab developed a COVID-19 County Tracker app to track cases nationwide. It features plots of total cases by states and county, with interactive exploration.
See DetailsMarina Sirota uses a computational study to identify a dozen of other drugs to reduce inflammation.
Read ArticleDr. Jiang’s research interest lies in machine learning methods, high dimensional models, functional data analysis and their applications in analyzing neurological, image, genetics data, and in designing adaptive randomization clinical trials. Her efforts yield high quality statistical and computational publications, which focus on addressing practical problems in the medical domain.
Dr. Scheffler develops statistical methods for high-dimensional signals produced by his collaborators in neurology and orthopedics. The proper interrogation of this data will lead to improved health outcomes at both the patient and population level.
The Abbasi Lab aims to understand functions of the brain and related disorders by leveraging principles in machine learning and statistics. The lab studies large-scale single-cell level neurophysiological datasets as well as medical images and structured and unstructured healthcare data.
Vijay Ramani has been named to Forbes magazine’s annual 30 Under 30 list of rising stars in healthcare.
Read ArticleSergio Baranzini and Sharat Israni will lead the NSF award & collaborate with Google, Lawrence Livermore Library, and Institute Systems for Biology.
Read ArticleFour UCSF faculty members are among the 100 new members elected to the National Academy of Medicine this year, including BCHSI affiliate faculty Julia Adler-Milstein.
Read ArticleWe are developing & testing smartphone apps that use reinforcement learning models to personalize text-messages to encourage physical activity. The apps are designed for English & Spanish speaking individuals diagnosed with depression and diabetes, being treated at the ZSFGH.
Dr. Rankin investigates the underpinnings of human socioemotional behavior in aging and neurodegenerative disease with quantitative brain imaging, and develops tools for harmonizing cross-disciplinary data and analytic processes to facilitate scientific collaboration and clinical care.
Dr. Phillips focuses on the value of new technologies and how to effectively and efficiently implement them into health care. Her core specialty is precision medicine. Her work spans multiple disciplines, including basic, clinical & social sciences
Dr. Murray is leading the Advanced Analytics & Innovation team, she is involved in large analytic projects using EHR data to inform quality and value improvement efforts at the medical center. She is interested in predictive analytics and has done research in EHR phenotyping.
Sanders Lab aims to identify the etiology of developmental disorders through the discovery of genetic risk factors. They aim to continue progress, leverage findings to build a complete understanding of ASD, & extend this approach to other human disorders, including congenital malformations.
Dr. Hong’s research program focuses on combining clinical domain knowledge with data science to generate insights from real world data, develop actionable computational tools, and evaluate the benefit of these advances in personalized cancer care.
The main theme of our research is large-scale analysis of disease genomes by integrating multi-omics data, evolutionary insights, electronic health records, as well as digitized clinical traits from imaging and wearable sensor readouts. The ultimate goal is to build data-driven frameworks to detect diseases before symptoms emerge and to achieve precision health management.
Seeley and colleagues used maps of brain connections to predict how brain atrophy would spread in individual patients with frontotemporal dementia.
Read ArticleMajumdar will run the Center for Intelligent Imaging’s operations to accelerate the application of artificial intelligence (AI) technology to radiology.
Read ArticleCallcut led the product development of the new AI screening tool, known as Critical Care Suite, which is being licensed by UCSF Innovation Venture to GE Healthcare.
Read ArticleAlejandro Sweet-Cordero, explains how the UCSF500 test can identify inherited predispositions to cancer and help patients design prevention and surveillance strategies.
Read ArticleUCSF, Cornell Tech, Sage Bionetworks, Open mHealth and The Commons Project are collaborating to integrate EHR data to Android smartphones.
Read ArticleStudy suggests tau tangles, not amyloid plaques, drive daytime napping that precedes dementia.
Read ArticleBCHSI faculty Marina Sirota led the first cohort of the AI4All program, focused on promoting greater diversity and inclusion in the field of AI in biomedicine.
Read ArticleDr. Tison applies machine learning and deep-learning techniques to large-scale electronic health data from heterogeneous sources in order to achieve the goal of personalized cardiovascular prognosis and disease prevention.
Audit-log data captures moment to moment clinical care processes within EHR. It makes these labor-intensive approaches obsolete, providing more comprehensive data faster with more accuracy. Taking this audit-log data and combining it with clinical data enables us to research the impact of provider workflows, behaviors and interactions with the EHR on patient outcomes.
New fellowship program from UCSF’s BCHSI, UC Berkeley’s BIDS, and Janssen Research & Development to recruit data scientists for innovative, high-impact, data-driven healthcare research
Read ArticleRamani Lab invents molecular tools to study biology, e.g. devising ways to molecularly tag nucleic acids and proteins with unique genomic or proteomic identifiers, then using these to quantify biological phenomena at the level of single cells and single molecules, using high-throughput sequencing and cutting-edge mass spectrometric techniques.
Dr. Huang studies biological processes relating to cancer disparities with a focus in prostate cancer. His lab uses cancer genomics including single-cell approaches to understand mechanisms that drive lethal, aggressive disease. A major focus is to identify and uncover roles for cancer genes and identify new cancer vulnerabilities.
Dr. Whooley’s work focuses on applying the methods of health services research (including data science, implementation science, and program evaluation) to accelerate the adoption of evidence-based practices within the context of a national learning healthcare system. One of her specific interest areas is in the domain of cardiovascular health outcomes.
Dr. Larson’s research program focuses on developments aimed at several applications: Metabolic imaging methods using hyperpolarized carbon-13 MRI; Semi-solid tissue MRI, for imaging of tendons, cortical bone, myelin, and lung tissue; and PET/MRI systems that combine the exceptional soft-tissue contrast of MRI with the functional contrast of PET.
Dr. Pirracchio focuses on three clinical research areas including evaluation and optimization of daily cares in the intensive care unit. He is also broadly interested in problems of causal inference and prediction, particularly developing novel methodologies for addressing scientific questions using complex observational data subject to sampling biases.
Julia Adler-Milstein will lead the new partnership between UCSF and The Doctor’s Company to make substantive advances in patient safety and digital health.
Read ArticleDr. Seo applies his expertise in radionuclide and x-ray imaging physics and instrumentation to develop quantitative imaging techniques for everything from small animal research to analysis of clinical research data.
Dr. Pedoia develops analytics to model the complex interactions between morphological, biochemical and biomechanical aspects of the knee joint as a whole; deep learning convolutional neural network for musculoskeletal tissue segmentation and for the extraction of silent features from quantitative relaxation maps for a comprehensive study of the biochemical articular cartilage composition; with ultimate goal of developing a completely data-driven model that is able to extract imaging features and use them to identify risk factors and predict outcomes for Osteoarthritis.
Dr. Goldstien uses the tools of big data, statistics and machine learning to answer questions related to Precision Medicine such as: How can we learn from the inventory of genomic test and knowledge in the EMR about patient outcomes and the data associated with individual patients to better direct their care? How can we better use repeatable animal models to translate knowledge to human patients? How can we integrate existing knowledge and high throughput data? Can we use genomic data to bring new therapies to bear?
The Zaitlen lab develops statistical and computational tools to discover the genetic basis of complex phenotypes, with particular interest in human disease, variation in drug/treatment response, and disease outcomes. Current projects primarily focus on incorporating environmental context into medical genetics.
Dr. van ‘t Veer’s research focuses on personalized medicine, to advance patient management based on knowledge of the genetic make-up of the tumor as well as the genetic make-up of the patient. This allows clinicians to optimally assign systemic therapy for those patients in need of such treatment, and to ensure the selection of the therapy that is most effective.
Dr. Hess’s research interests lie in the development and translational application of magnetic resonance imaging techniques for diagnosis and treatment monitoring in neurologic disease. His scientific background is in MRI acquisition, reconstruction and image analysis, focusing on diffusion and high-field MRI. His primary clinical interests are in neurovascular disease, dementia, brain development, and epilepsy.
The Goodarzi laboratory employs a systems biological and multidisciplinary approach that integrates computational and experimental strategies to identify and characterize key regulatory programs that underlie cancer progression.
Dr. Arnaout’s lab is currently developing computational methods to bring precision phenotyping to echocardiography, and also using the zebrafish animal model to study cardiovascular developmental gene function and to model human cardiovascular disease.
Ashish’s team develops novel image processing and analysis algorithms for MRI. His lab also works to model brain connectivity networks using graph theory, and investigates how these networks are disrupted with disease and trauma.
Dr Baranzini’s current research involves immunological studies using the EAE model, sequencing of whole genomes and transcriptomes from patients with multiple sclerosis and developing bioinformatics tools to integrate this information with that coming from other high throughput technologies. Dr Baranzini uses a combination of “wet lab” methods including DNA microarrays, proteomics, and laser capture microdissection, in combination with “dry lab” analytical approaches encompassing bioinformatics, complexity theory, and mathematical modeling.
Dr. Adler-Milstein’s research assesses the progress of health IT adoption; the impact of such adoption on healthcare costs and quality; and the relationships between market, organizational, and team structure and health IT use. A core focus of her work is on health information exchange and interoperability.
The Grinberg Lab processes whole human brains for state-of-the-art quantitative histological analysis, digitize all of the results, and precisely registers to MRI. They are developing advanced tools for analysis of microscopic images that enable more comprehensive and higher-throughput studies of human brain tissue.
Dr. Brown uses knowledge engineering, health informatics, comparative-effectiveness research, semantic harmonization, and integration of datasets (including EHR) to examine health disparities and develop patient-centered health information tools.
The lab seeks to discover new therapeutic approaches to target the genetic mutations and altered signaling networks that are specific to cancer cells. Using functional genomics applied to mouse and human systems, we work to understand the transcriptional networks that regulate the outcome of specific oncogenic mutations and to understand how cancers become resistant to chemotherapy. This work relies heavily on computational genomic analysis, generating and using high-throughput datasets and next-generation sequencing for gene and network discovery. Our primary disease focus is lung cancer and pediatric sarcomas.
Dr. Kober uses ‘omics data (i.e., genotype and expression arrays, DNAseq — genome, exome, RNAseq, methylation arrays) to improve our understanding of the molecular mechanisms underlying common symptoms (e.g., fatigue, pain) or treatment failure experienced by patients with chronic medical conditions (e.g., cancer, HIV infection).
