Health · Medical Research · 14 hrs ago
Kaiser Permanente and MIT develop AI models to flag pregnancy risks early
Kaiser Permanente and MIT researchers developed AI models to identify patients at higher risk of pregnancy complications.
The models use routine electronic health record data, rather than information beyond standard records.
They can flag higher risk as early as 12 weeks into pregnancy.
The complications covered are gestational diabetes, gestational hypertension and preeclampsia.
The team trained 49 machine-learning models on data from 442,760 pregnancies.
The findings describe early risk prediction, but do not say what care changes may follow or what the researchers plan next.
Kaiser Permanente and Massachusetts Institute of Technology researchers built AI models to flag pregnancy complications early.
The models use only routine electronic health record data.
They identify patients at higher risk for gestational diabetes, gestational hypertension and preeclampsia as early as 12 weeks into pregnancy.
The researchers trained 49 machine learning models on data from 442,760 pregnancies.
The study was published in Communications Health.
- Who
- Researchers from Kaiser Permanente and the Massachusetts Institute of Technology.
- What
- They built AI models using routine electronic health record data to flag risk of three pregnancy complications.
- When
- The models can flag risks as early as 12 weeks into pregnancy. The article was published on November 20, 2026.
- Where
- Not stated.
- Why
- To identify patients at higher risk for gestational diabetes, gestational hypertension and preeclampsia early in pregnancy.
This story does not have two clearly opposing sides.
This story does not have a timeline yet.
- Organizations
- Kaiser Permanente and Massachusetts Institute of Technology
- Complications
- Gestational diabetes, gestational hypertension and preeclampsia
- Earliest risk flagging
- 12 weeks into pregnancy
- Models trained
- 49 machine learning models
- Pregnancies
- 442,760
- Study journal
- Communications Health