Road traffic accidents in Sri Lanka: A retrospective analysis and artificial intelligence-based solutions for prevention

Introduction: This study aims to conduct a retrospective, descriptive analysis to identify the factors contributing to road traffic accidents (RTAs) and propose artificial intelligence (AI)-based solutions for their prevention. Methods: The ninety-nine postmortem reports belonging to the investigators and documents connected to death investigations due to RTA were selected from fatalities reported to the Colombo North Teaching Hospital, Ragama, Sri Lanka, from 2001 to 2021. Data was gathered based on a pre-prepared questionnaire and analysed using IBM SPSS Statistics version 26. Strict confidentiality of individual information was maintained throughout the analysis and presentation of the results. Results: Among the selected population, 83.8% were male, and the majority belonged to the 51-60 year age group (19.2%). Pedestrians were the most vulnerable group (41.5%), followed by motorcycle riders (17.2%). Main roads (72.7%) were the most frequent place for RTA. The most common time for RTA was between 06.01 pm and 12.00 am (38.4%). The majority of the RTAs had contributory human factors (85.8%), but most of them did not have any identifiable contributory environmental, road-related, or vehicle-related factors. Breaking road regulations was the most common contributory human factor (27.3%). It was followed by poor judgment (25.3%). Blood for alcohol was sent in 60.6% of the selected population, and 18.2% were found to be positive with available reports. (> 80mg/dl). Blood alcohol reports could not be traced in 14.1% of victims. Head injuries (41.4%) and multiple trauma (37.3%) were the most common causes of death among the fatalities due to RTA. Conclusions: By analysing the data from history, toxicology reports, and other information provided by police, it can be inferred that the majority of RTAs had contributory human factors. It is postulated that supervised AI-based solutions could reduce the number of fatalities attributable to human errors.