arXiv subjects

B M Tazbiul Hassan Anik

Publications and source records attributed to B M Tazbiul Hassan Anik.

3 recordsLinked to original sources

arXiv2024

Evaluating Traumatic Brain Injury Outcomes in Road Traffic Crashes

Traumatic Brain Injuries (TBIs) resulting from Road Traffic Crashes (RTCs) can have fatal and disabling effects on patients. In this study, we evaluated the TBIs outcomes of patients involved in RTCs and identify key contributing factors affecting these outcomes. Data on 207 patients recorded by physicians at a tertiary hospital in Bangladesh was collected. A random parameters multinomial logit model with heterogeneity in the means was utilized to assess patients outcomes in three categories: Non-surgical, Surgical, and Fatal. From the random parameters, the study found that male patients (55.48%) are more likely to experience surgical and fatal outcomes. Male motorcycle users have a higher probability of experiencing fatal consequences. Additionally, 60.94% of incidents on rural roads result in surgeries and fatalities, with nighttime crashes on these roads significantly increasing the likelihood of fatal outcomes. Key factors impacting the likelihood of TBIs outcomes include older age, pedestrian involvement, bus and truck crashes, speeding, wet pavements, overtaking, low visibility, and weekday crashes. The study identified two significant interaction variables that increase the probability of fatal outcomes from TBIs: the interactions between low visibility and bus involvement, and between overtaking and wet pavements. While these factors individually had a higher probability of leading to both surgical and fatal outcomes, together these factors increase the risks to fatalities. Overall, our findings provide more detailed insights about the impact of TBIs outcomes resulting from RTCs and emphasize the need to develop more effective measures to improve road safety and patient outcomes.

stat.AP↗

arXiv2023

Prediction of Crash Injury Severity in Florida's Interstate-95

Drivers can sustain serious injuries in traffic accidents. In this study, traffic crashes on Florida's Interstate-95 from 2016 to 2021 were gathered, and several classification methods were used to estimate the severity of driver injuries. In the feature selection method, logistic regression was applied. To compare model performances, various model assessment matrices such as accuracy, recall, and area under curve (AUC) were developed. The Adaboost algorithm outperformed the others in terms of recall and AUC. SHAP values were also generated to explain the classification model's results. This analytical study can be used to examine factors that contribute to the severity of driver injuries in crashes.

cs.LG↗

arXiv2023

inTformer: A Time-Embedded Attention-Based Transformer for Crash Likelihood Prediction at Intersections Using Connected Vehicle Data

The real-time crash likelihood prediction model is an essential component of the proactive traffic safety management system. Over the years, numerous studies have attempted to construct a crash likelihood prediction model in order to enhance traffic safety, but mostly on freeways. In the majority of the existing studies, researchers have primarily employed a deep learning-based framework to identify crash potential. Lately, Transformer has emerged as a potential deep neural network that fundamentally operates through attention-based mechanisms. Transformer has several functional benefits over extant deep learning models such as LSTM, CNN, etc. Firstly, Transformer can readily handle long-term dependencies in a data sequence. Secondly, Transformers can parallelly process all elements in a data sequence during training. Finally, a Transformer does not have the vanishing gradient issue. Realizing the immense possibility of Transformers, this paper proposes inTersection-Transformer (inTformer), a time-embedded attention-based Transformer model that can effectively predict intersection crash likelihood in real-time. The proposed model was evaluated using connected vehicle data extracted from Signal Analytics Platform. Acknowledging the complex traffic operation mechanism at intersection, this study developed zone-specific models by dividing the intersection region into two distinct zones: within-intersection and approach zone. The best inTformer models in 'within-intersection,' and 'approach' zone achieved a sensitivity of 73%, and 70%, respectively. The zone-level models were also compared to earlier studies on crash likelihood prediction at intersections and with several established deep learning models trained on the same connected vehicle dataset.

cs.LG↗