-
Notifications
You must be signed in to change notification settings - Fork 0
Glucose monitoring
Data not missing at random
A study entitled "Variation in Monitoring: Glucose Measurement in the ICU as a Case Study to Preempt Spurious Correlations" examines how often glucose measurements are taken for sepsis patients in the ICU setting . Utilizing the MIMIC-IV database, the study reveals no significant disparities on glocose monitoring on day 1 of ICU admission across different demographic groups. However, subsequent days showed that Hispanic and Black patients, as well as those proficient in English, received glucose measurements more frequently than their counterparts. These findings are critical as they highlight a potential source of bias in clinical data, which can inadvertently be transferred to AI models used in healthcare.
Glucose, Race, English proficiency
A MIT Critical Data Original Production
MIT Critical Data
- What Is The Data Artifacts Glossary?
- What Is The Data Artifacts Glossary Ultimate Goal?
- What Problem Does The Data Artifacts Glossary Solve?
- What Design Principles Underlie The Data Artifacts Glossary
- How Does The Data Artifacts Glossary Accomplish Its Goals?
- The Data Artifacts Glossary Initial Categories