Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 2 appraisals
PloS one
Data fusion-based traffic prediction and software decision support for recreational suburban roads
Predicting traffic flow on mountainous suburban roads is challenging due to highly variable environmental, temporal, and traffic-related conditions. This study focuses on Kandovan Road, a critical route with complex behavioral patterns influenced by weather conditions, calendar events, and road-specific characteristics. To improve forecasting accuracy, eight machine learning and deep learning models were implemented, including Deep LSTM, Random Forest Regressor, XGBRegressor, Transformer, ST-ResNet, Conv-LSTM, Bidirectional LSTM, and LSTM-GAN. The models were trained and evaluated using traffic, weather, and event datasets from 2017 to 2023, with performance measured through MAE, RMSE, MSE, and MAPE metrics. Among the evaluated models, the Random Forest Regressor achieved the highest accuracy with an R² score of 0.88 and a low average error. This result demonstrates its strong ability to model non-linear and dynamic traffic patterns. The results indicate that integrating diverse data sources significantly enhances traffic prediction performance on mountain roads. Additionally, a dedicated traffic forecasting software system was developed to visualize real-time predictions and provide an operational decision-support tool for traffic authorities. The outcomes of this work support more efficient traffic management, improved road safety, and sustainable transportation planning in challenging terrains.
11 May 2026
Read appraisal →PloS one
SSH-YOLO: YOLOv8 improved model based on Object Detection in Complex Road Scenes
To address the problem of insufficient detection accuracy for dense targets, small targets and partially occluded objects in complex road scenarios, an improved object detection model, SSH-YOLO, is proposed. On the basis of YOLOv8n, the model optimizes and improves performance through a three-level collaborative architecture: 1) introduce the spatial and deep conversion (SPDConv) module in the backbone network to replace the traditional step downsampling with nonstep convolution, retain the fine-grained features of small targets, and solve the feature loss problem of low-resolution images; 2) embed the spatial and channel collaborative attention module (SCSA), through cross-scale feature fusion (SMSA) and channel weight progressive optimization (PCSA), focus on the key visible areas of the occluded target and suppress background interference such as roadside vegetation; and 3) add a new 160 × 160 resolution small object detection head, combined with the original P3‒P5 layer to form a four-level detection system, covering long-distance small targets < 32 × 32 pixels. The experimental results show that the improved model performs well on the self-built RoadScene-Complex dataset and four public datasets BDD100K: 0.729 (12.4% higher than YOLOv8n) on the RoadScene-Complex dataset [email protected] (12.4% higher than YOLOv8n) and 0.868 (7.6% higher than the KITTI dataset) [email protected]). COCO small target subset [email protected] to 0.585 (up 16.5%), and CityPersons occluded scene [email protected] to 0.739 (up 22.8%). At the same time, it maintains lightweight characteristics and has an inference speed of up to 60 FPS to meet the needs of real-time on-board detection. The research results provide a balanced solution of "accuracy-speed-lightweight" for high-precision target detection in complex traffic scenarios, especially in small target and occlusion scenarios.
17 Apr 2026
Read appraisal →