Sign in

Tomiko Oskotsky

@tomiko20.bsky.social
4 followers 4 following 0 posts
PostsRepliesMedia
Reposted by Tomiko Oskotsky
UCSF Bakar Computational Health Sciences Institute @ucsf-bchsi.bsky.social · 28/08/2026
Read Jordan Guillot, B. Miao, A. Suresh, M. Sushil, C. Williams, R. Vashisht, T. Oskotsky, M. Sirota, @atulbutte.bsky.social in @plos.org "Extracting adverse event nature, severity, timelines, interventions from clinical notes CAR-T cell therapy patients using LLM" journals.plos.org/digitalhealt...
journals.plos.org
Extracting adverse event nature, severity, timelines and resulting interventions from clinical notes of patients receiving CAR-T cell therapy using large language models
Author summary Chimeric Antigen Receptor T-cell (CAR‑T) therapy has transformed care for blood cancer but requires careful tracking of adverse events (AEs). We asked whether a large language model cou...
132
Reposted by Tomiko Oskotsky
UCSF Bakar Computational Health Sciences Institute @ucsf-bchsi.bsky.social · 25/08/2025
We were honored to share our 10 yr anniversary “A Decade of Impact: Celebrating @atulbutte.bsky.social & BCHSI@10” w/ 200+ people, faculty & community. Read our 10 Yr Report & Watch the recording! bakarinstitute.ucsf.edu/news/decade-... @ucsanfrancisco.bsky.social @ucsfmedicine.bsky.social
022
Reposted by Tomiko Oskotsky
Healthcare in Europe (DE) @health-europe-de.bsky.social · 22/08/2025
Treten bei der Verteilung medizinischer #Daten Cluster auf, kann das die Ergebnisse verfälschen. Dr. Martin Papenberg und @guwek.bsky.social von der @hhu.de zeigen mit Forschenden der @ucsf-bchsi.bsky.social unter Leitung von @tomiko20.bsky.social ein neues Verfahren, das #Clustering verhindert.
healthcare-in-europe.com
Anticlustering für bessere Verteilung medizinischer Daten
Daten in Gruppen aufzuteilen, ohne dass sich dabei ungewollte Cluster mit ähnlichen Elementen bilden, ist etwa für die Auswertung medizinischer Daten von großer Bedeutung.
011
Reposted by Tomiko Oskotsky
Gerhard Jocham @gjocham.bsky.social · 19/08/2025
Really cool work from @hhu.de colleagues Martin Papenberg )(psychology) and ‪@guwek.bsky.social‬ (computer science), together with researchers at UCSF and Stanford! 👇 www.hhu.de/en/news-arti...
hhu.de
Psychology and computer science: Publication in Cell Reports Methods
Avoiding the formation of unwanted clusters of similar elements when dividing data into groups is of great importance for the analysis of medical data. Psychologists and computer scientists from Heinr...
032
Reposted by Tomiko Oskotsky
UCSF Bakar Computational Health Sciences Institute @ucsf-bchsi.bsky.social · 14/08/2025
We look forward to seeing you at our 10 year anniversary symposium Tues., August 19th ""A Decade of Impact: Celebrating @atulbutte.bsky.social & BCHSI@10." Full agenda is up! calendar.ucsf.edu/event/a-deca... @ucsfhealth.bsky.social @ucsanfrancisco.bsky.social @ucsfmedicine.bsky.social
044
Reposted by Tomiko Oskotsky
UCSF Bakar Computational Health Sciences Institute @ucsf-bchsi.bsky.social · 10/03/2025
"Searching for the cellular underpinnings of the selective vulnerability to tauopathic insults in Alzheimer’s disease" published in Nature @CommsBio @UCSF @torok_justin @PedroDoriaMaia @RajLab_UCSF @UCSFImaging et al www.nature.com/articles/s42...
nature.com
Searching for the cellular underpinnings of the selective vulnerability to tauopathic insults in Alzheimer’s disease - Communications Biology
Meta-analysis of mouse tauopathy data sheds light on the cell types that may be involved in conferring selected regional vulnerability to pathology.
031
Reposted by Tomiko Oskotsky
joephpc.bsky.social @joephpc.bsky.social · 26/03/2025
Our manuscript is now published in Nature Medicine! This paper is a great example of how we can use AI beyond just predicting outcomes. It was a joyride working with @jrinthenicu.bsky.social @nnimaa.bsky.social and all other co-authors. Keep an eye out for what's next!
052
Reposted by Tomiko Oskotsky
Nima Aghaeepour @nnimaa.bsky.social · 25/03/2025
Many to thank including @joephpc.bsky.social @jrinthenicu.bsky.social @stanfordanes.bsky.social @stanfordmedicine.bsky.social @dbdsstanford.bsky.social @ucsfchildrens.bsky.social @neomindai.bsky.social @lorenzorighetto.bsky.social @naturemedicine.bsky.social
042
Reposted by Tomiko Oskotsky
Nima Aghaeepour @nnimaa.bsky.social · 25/03/2025
In this particular example, the AI model analyzed ~10 years worth of parenteral nutrition data and learned how to do the entire process automatically, increasing safety and reducing cost. Stay tuned for more exciting news about our next steps!
121
Reposted by Tomiko Oskotsky
Nima Aghaeepour @nnimaa.bsky.social · 25/03/2025
Excited to share our new paper in @natmedicine.bsky.social! Prematurity is the #1 cause of death in kids under 5. We show that AI can do more than predict outcomes — it can guide critical treatment decisions to improve outcomes. Press release: med.stanford.edu/news/all-new...
med.stanford.edu
AI can help doctors give intravenous nutrition to preemies, Stanford Medicine study finds
An algorithm that learned from tens of thousands of nutrition prescriptions for premature babies could reduce medical errors and better identify what nutrients the smallest patients need.
297
Reposted by Tomiko Oskotsky
bioRxivpreprint @biorxivpreprint.bsky.social · 11/03/2025
Anticlustering for Sample Allocation To Minimize Batch Effects www.biorxiv.org/content/10.1101/202…
032