Sam Fenske
@samfenske.bsky.social
PhD student @yalecbb.bsky.social in the lab of @hattaca.bsky.social | interested in single-cell genomics, drug discovery, and AI | former data analyst at Northwestern PCCM | alum of WashU BME
In failed extubation cases, the model would’ve advised against extubation 35% of the time, showing promise as a second opinion tool 💡. Top predictors from SHAP and ablation testing include plateau pressure, heart rate, PaCO2, aligning with clinical intuition.
July 30, 2025 at 7:16 PM
In failed extubation cases, the model would’ve advised against extubation 35% of the time, showing promise as a second opinion tool 💡. Top predictors from SHAP and ablation testing include plateau pressure, heart rate, PaCO2, aligning with clinical intuition.
We trained models on 37 clinical features (vitals, labs, meds, vent settings)🫁 collected from midnight–8 AM ⏰ , so predictions are ready for morning rounds. We carefully annotated data, reviewing hundreds of charts.
July 30, 2025 at 7:16 PM
We trained models on 37 clinical features (vitals, labs, meds, vent settings)🫁 collected from midnight–8 AM ⏰ , so predictions are ready for morning rounds. We carefully annotated data, reviewing hundreds of charts.
In failed extubation cases, the model would’ve advised against it 35% of the time, showing promise as a second opinion tool 💡. Top predictors from SHAP and ablation testing include plateau pressure, heart rate, PaCO2, aligning with clinical intuition.
July 30, 2025 at 6:59 PM
In failed extubation cases, the model would’ve advised against it 35% of the time, showing promise as a second opinion tool 💡. Top predictors from SHAP and ablation testing include plateau pressure, heart rate, PaCO2, aligning with clinical intuition.
We trained models on 37 clinical features (vitals, labs, meds, vent settings)🫁 collected from midnight–8 AM ⏰ , so predictions are ready for morning rounds. We carefully annotated data, reviewing hundreds of charts.
July 30, 2025 at 6:59 PM
We trained models on 37 clinical features (vitals, labs, meds, vent settings)🫁 collected from midnight–8 AM ⏰ , so predictions are ready for morning rounds. We carefully annotated data, reviewing hundreds of charts.