Remote Photoplethysmography and Motion Tracking Convolutional Neural Network with Bidirectional Long Short-Term Memory: Non-Invasive Fatigue Detection Method Based on Multi-Modal Fusion.

Existing vision-based fatigue detection methods commonly utilize RGB cameras to extract facial and physiological features for monitoring driver fatigue. These features often include single indicators such as eyelid movement, yawning frequency, and heart rate. However, the accuracy of RGB cameras can be affected by factors like varying lighting conditions and motion. To address these challenges, we propose a non-invasive method for multi-modal fusion fatigue detection called RPPMT-CNN-BiLSTM. This method incorporates a feature extraction enhancement module based on the improved Pan-Tompkins algorithm and 1D-MTCNN. This enhances the accuracy of heart rate signal extraction and eyelid features. Furthermore, we use one-dimensional neural networks to construct two models based on heart rate and PERCLOS values, forming a fatigue detection model. To enhance the robustness and accuracy of fatigue detection, the trained model data results are input into the BiLSTM network. This generates a time-fitting relationship between the data extracted from the CNN, allowing for effective dynamic modeling and achieving multi-modal fusion fatigue detection. Numerous experiments validate the effectiveness of the proposed method, achieving an accuracy of 98.2% on the self-made MDAD (Multi-Modal Driver Alertness Dataset). This underscores the feasibility of the algorithm. In comparison with traditional methods, our approach demonstrates higher accuracy and positively contributes to maintaining traffic safety, thereby advancing the field of smart transportation.

Construction of a Nomogram Model for Predicting Pleural Effusion Secondary to Severe Acute Pancreatitis.

Background: This study aims to investigate the risk factors of pleural effusion (PE) secondary to severe acute pancreatitis (SAP) and to build a nomogram model. Methods: The clinical parameters of SAP patients admitted to the emergency department of the First Affiliated Hospital of Bengbu Medical College from January 2019 to August 2021 were retrospectively collected. The independence risk factors of PE secondary to SAP were analyzed by univariate analysis and multivariate logistic regression analysis. A nomogram risk prediction model was established and validated through the area under the ROC curve. Result: Two hundred twenty-two SAP patients were included for analysis, of which 65 patients experienced secondary PE. The incidence of PE secondary to SAP was 29.28% (65/222). Logistic regression analysis showed that serum albumin (ALB) (OR = 0.830, 95% CI: 0.736∼0.936), fibrinogen (FIB) (OR = 4.573, 95% CI: 1.795∼11.648), C-reactive protein (CRP) (OR = 1.046, 95% CI: 1.009∼1.083), acute physiology, chronic health score system (APACHE-II) score (OR = 1.484, 95% CI: 1.106∼1.990), and sequential organ failure score (SOFA) (OR = 43.038, 95% CI: 2.030∼4.548) were independent risk factors for PE secondary to SAP (P < 0.05) and entered into the nomogram. The nomogram showed robust discrimination with an index of concordance of 0.755 and an area under the receiver operating characteristic curve of 0.837 (95% CI: 0.779∼0.894). Conclusion: We developed a nomogram model for PE secondary to SAP with ALB, FIB, CRP, APACHE-II scores, and SOFA scores. The nomogram model showed good discrimination and consistency, and it can better predict the risk of PE secondary to SAP.