Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0729
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-08-10T06:35:54.781025Z
status: published
lens: trace
sector: health
headline: Mortality prediction of road traffic crash with artificial intelligence: a systematic review
dek: Background Road traffic crashes cause substantial global mortality and disability. Conventional injury severity scores may not fully capture the complex interactions among demographic, clinical, crash and environmental factors. Artificial intelligence and machine learning may improve mortality prediction by modelling non-linear patterns in traffic crash data. Methods This systematic review followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidance. PubMed/MEDLINE, Web of S…
gain_title: Systematic review found AI and machine learning may improve mortality prediction after road traffic crashes by modelling non-linear patterns among demographic, clinical and crash factors.
problem_title: Clinical translation of AI mortality prediction after road traffic crashes remains limited by insufficient external validation, inconsistent handling of class imbalance, and incomplete reporting of tuning and missing data strategies.
trace_subject: AI and machine learning for mortality prediction after road traffic crashes
gain_reading: Systematic review found AI and machine learning may improve mortality prediction after road traffic crashes by modelling non-linear patterns among demographic, clinical and crash factors.
gain_evidence: Artificial intelligence and machine learning show promise for traffic crash mortality prediction | may improve mortality prediction by modelling non-linear patterns in traffic crash data
problem_reading: Clinical translation of AI mortality prediction after road traffic crashes remains limited by insufficient external validation, inconsistent handling of class imbalance, and incomplete reporting of tuning and missing data strategies.
problem_evidence: clinical translation remains limited by insufficient external validation | inconsistent handling of class imbalance, incomplete reporting of tuning and missing data strategies and limited use of explainability methods
quick_read: A systematic review published 7 August 2026 examined 18 retrospective studies from 2014-2025 that used AI or machine learning to predict death after road traffic crashes, drawing mostly on national or regional databases, hospital records, and police or insurance tabular data.

The review matters because better mortality prediction could inform triage and resource allocation, but the evidence as of the August 2025 search shows limited external validation and inconsistent reporting of imbalance handling, tuning, and explainability, leaving prospective, interpretable validation as an unresolved need.
limitation: Findings were synthesised narratively due to heterogeneity, and only two studies reported external validation, limiting generalizability.
tag: Dual reading
key_points: Systematic review screened PubMed/MEDLINE, Web of Science, Scopus and IEEE Xplore on 7 August 2025 for 2014-2025 studies. | 18 studies met inclusion criteria, most retrospective using structured tabular data from national or regional databases, hospital records, and police or insurance datasets. | Frequently reported predictors included age, Injury Severity Scores, body region injured, crash mechanism, temporal factors and geographical characteristics.
rundown: The review followed PRISMA 2020 guidance and searched four databases on 7 August 2025 for English-language peer-reviewed studies from 2014 to 2025 applying AI or machine learning to predict mortality after road traffic crashes.

Across the 18 included studies, regression-based models and decision trees remained common while ensemble methods including random forest and gradient boosting increased in recent years, with objectives including binary mortality prediction and multiclass injury severity prediction including death.
sources:
- peer_reviewed | Injury Prevention | https://doi.org/10.1136/ip-2025-046038 | 2026-08-07
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

b61e2f507dfc6340ce65c51df669300c30a54b684fd7666bf4da4673a53e43f5

previous

0000000000000000000000000000000000000000000000000000000000000000