Neuroanesthesia & neurocritical care
Deep white matter MRI predicts outcomes in coma of various etiologies: a cohort study
Authors' conclusion
This classifier demonstrates a high specificity to predict coma outcome while patients are still in the ICU, irrespective of coma etiology.
Study context
- Population
- Patients in coma of various etiologies not responding to simple orders at day 5 after coma onset
- Participants
- 531 vs 211
- Comparison
- Training set vs External test set
- Training setn = 531
- External test setn = 211
- Outcome
- Prediction of unfavorable outcome (UFO, GOSE 1-4) by diffusion tensor imaging (DTI)-based classifier
- Follow-up
- One year
Methodological detailsmulticentre, international, supervised machine learning, external validation
- Multicentre
- International
- Supervised machine learning
- External validation
Performance
As reported in the abstract
Classifier performance in the external test set
| Test or model | ROC AUC | Unfavorable outcome (UFO) specificity | UFO PPV | UFO sensitivity | FO specificity | FO PPV | FO NPV | No determination zone (NDZ) |
|---|---|---|---|---|---|---|---|---|
| External test set | 0.89 | 98.1% | 99.1% | 68.6% | 95.0% | 77.8% | 86.3% | 30.8% |
External test set
ROC AUC0.89
Unfavorable outcome (UFO) specificity98.1%
UFO PPV99.1%
UFO sensitivity68.6%
FO specificity95.0%
FO PPV77.8%
FO NPV86.3%
No determination zone (NDZ)30.8%
Further findings
Classifier specificity after excluding patients with withdrawal of life sustaining therapies (n = 104)
- Specificity for UFO classification
- 96.6%
- Specificity for FO classification
- 82.4%
Original publication
Deep white matter MRI predicts outcomes in coma of various etiologies: a cohort study. Crit Care. Published 4 March 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →