Bringing AI & machine learning to the clinic

We are pioneers in end-to-end development, validation, and clinical trials to build and demonstrate artificial intelligence tools improve outcomes for patients.

Predicting & reducing the side effects of cancer therapies

We predict side effects with computational approaches to enable early management and reduction to improve cancer care.

Improving treatment of prostate cancer

We create algorithms to predict outcomes and identify the best personalized treatments for patients with prostate cancer.

Optimizing the delivery of cancer care

We develop approaches to anticipate the needs of patients with cancer to ensure they have the resources they need before they need them.

Integrating clinical knowledge & data science

We integrate knowledge of the underlying details of clinical data with machine learning expertise to guide algorithmic development.

The Hong Lab is apart of

News

We're a go! The Wearable Activity Tracking to Curb Hospitalizations (WATCH) trial is now open! This open study will evaluate our prior models in predicting unplanned acute care for patients who are undergoing cancer therapy.
Hot off the press, congrats to Travis Zack, now on faculty at UCSF, and CPH student Li-Ching Chen leading our most recent publication "Assessing Large Language Models for Oncology Data Inference From Radiology Reports," investigating the use of large language models to "read" radiology reports in JCO Clincal Cancer Informatics!
We had a strong showing at ASTRO 2024! Congrats to Sarah Hampson and Michelle Zhao for their oral presentations on applying natural language processing to predicting emergency visits and hospitalizations and analyzing the impact of distress on cancer treatment symptomology! Congrats to Jonathan Ejie and Nathan Magalit on their posters describing the relationship between distress and acute care and patient portal utilization!
Congratulations to Ali Sabbagh for winning an ASCO Merit Award! He will be presenting the preliminary results from our Prostate Cancer Foundation-supported study applying machine learning to help accelerate the read-out of clinical trials at the ASCO Annual Meeting next week! This study leveraged seven key trials in metastatic prostate cancer and we developed a modeling approach to predict likely outcomes of the trials within the first several months.
Julian collaborated with Dr. James Yu to review AI applications in prostate cancer for the journal Oncology.
Congratulations to our collaborative team led by Izzy Friesner alongside Jean Feng and our collaborators from Montefiore Einstein Comprehensive Cancer Center for our latest publication in JAMA Oncology, building the first wearables-based clinical model for predicting adverse events during cancer therapy! The next step is validation on the recently opened NRGF-001 national clinical trial.
Congratulations to Divya Natesan and the SHIELD-RT team on our latest publication in the New England Journal of Medicine AI, one of the first cost analyses in healthcare AI, demonstrating the cost benefits of our machine learning approaches for predicting adverse events during radiation treatment!
Congrats to Hui Lin, Lisa Ni, and Christina Phuong on their recent publication, Natural Language Processing for Radiation Oncology: Personalizing Treatment Pathways, in Pharmacogenomics and Personalized Medicine!
Congratulations to Jane Chen for receiving the American Association for Women in Radiology (AAWR) Member-in-Training Awards for Outstanding ASTRO Presentations!
Congratulations to Sumi Sinha on her paper describing oncologist utilization of the electronic health record in JNCI Spectrum! This collaborative effort was also highlighted by media coverage in Becker's Hospital Review and Cancer Therapy Advisor.
Congratulations to Mike Waters on his paper, Unlocking the Power of ChatGPT, Artificial Intelligence, and Large Language Models: Practical Suggestions for Radiation Oncologists in Practical Radiation Oncology!
We are excited to have received an NIH/NCI R01, which will support our work in the generalizability of electronic health record and wearable device-based machine learning strategies to predict and reduce acute care in patients undergoing cancer therapies. We are looking forward to working with our wonderful collaborators at Duke, Beth Israel Deaconness, Essentia Health, and Washington Hospital!
Congrats to Lisa Ni, Christina Phuong, and Gabe Vidal on their chapters on natural lanugage processing (NLP) and the future of radiation oncology in Artificial Intelligence in Radiation Oncology!
We are excited to have our latest SHIELD-RT-related study published in BMJ Health & Care Informatics, describing healthcare provider reactions to the implementation of our machine learning-directed care strategy.
Our publication describing challenges involved in implementing machine learning in the clinic based on our experience on SHIELD-RT was published in BMC Bioinformatics.
We are back from ASTRO and proud of our team!
- Izzy Friesner gave an oral presentation about our machine learning work to predict hospitalization based on wearable activity monitoring in collaboration with Montefiore Einstein Cancer Center. Media coverage in the ASTRO Press Release, Medscape, Aunt Minnie, Inside Precision Medicine, and Health IT Analytics.
- Jane Chen gave an oral presentation about our natural language processing work to understand patterns of documented symptoms following metastatic cancer diagnoses and prior to palliative radiotherapy.
- Daphna Spiegel gave an oral presentation on our natural language procesing work to characterize conversations on social media around breast cancer treatments.
- Will Chen gave an oral presentation on our work with collaborators investigating deep learning applications in predicting cardiac events based on radiotherapy treatment planning scans.

News archive

The Hong Lab @ UCSF