Hospital AI diuretic dosing model shows better kidney function in retrospective results
ShalitUri · x · 2026-11-03
The author, working with Rambam Medical Center, built a model for inpatient diuretic dosing decisions: it estimates patient-level effects of increasing/maintaining vs. decreasing doses on kidney recovery and 30-day readmission/mortality, and defers when uncertain.
Retrospective results are promising: better kidney function with no observed increase in mortality or readmission. The team is now running prospective validation with Rambam's clinical staff. The screenshot shows the recommendation interface in the hospital's EHR, comparing predicted 3-day creatinine, 30-day readmission/mortality risk, and predicted length of stay across the two options.
Related event: Causal ML for Acute HF Diuretic Dosing Enters First-in-Human Trial(2 posts)→
More from Research
- Researcher urges critics to read the prompts in the 'weight-space programming' post first — anton_lozhkov · 2026-11-03
- New paper: standard CoT monitorability metric is only an upper bound, improvable via self-distillation and RL — armancohan · 2026-11-03
- New Scaling Recipe for Data-Constrained Training Cuts Perplexity 14% via Hyperparameter Transfer and Ensembling — andrewgwils · 2026-11-03
- Purdue professor says arXiv banned him a year over 'hallucinated references' that weren't — RexDouglass · 2026-11-03
- KV cache translator proposal: small model reads 100K tokens, large model corrects — dreamai87 · 2026-11-03
- Yoav Goldberg highlights InterpBench, a benchmark of Transformers with known circuits for interpretability — yoavgo · 2026-11-03