In this paper, near-infrared spectroscopy(NIRS) was employed to analyze 129 batches of commercial products of Reduning Injection. The batch reporting rate was estimated according to the report of Reduning Injection in the direct adverse drug reaction(ADR) reporting system of the drug marketing authorization holder of the Center for Drug Reevaluation of the National Medical Products Administration(National Center for ADR Monitoring) from August 2021 to August 2022. According to the batch reporting rate, the samples of Reduning Injection were classified into those with potential risks and those being safe. No processing, random oversampling(ROS), random undersampling(RUS), and synthetic minority over-sampling technique(SMOTE) were then employed to balance the unbalanced data. After the samples were classified according to appropriate sampling methods, competitive adaptive reweighted sampling(CARS), successive projections algorithm(SPA), uninformative variables elimination(UVE), and genetic algorithm(GA) were respectively adopted to screen the features of spectral data. Then, support vector machine(SVM), logistic regression(LR), k-nearest neighbors(KNN), naive bayes(NB), random forest(RF), and artificial neural network(ANN) were adopted to establish the risk prediction models. The effects of the four feature extraction methods on the accuracy of the models were compared. The optimal method was selected, and bayesian optimization was performned to optimize the model parameters to improve the accuracy and robustness of model prediction. To explore the correlations between potential risks of clinical use and quality test data, TreeNet was employed to identify potential quality parameters affecting the clinical safety of Reduning Injection. The results showed that the models established with the SVM, LR, KNN, NB, RF, and ANN algorithms had the F1 scores of 0.85, 0.85, 0.86, 0.80, 0.88, and 0.85 and the accuracy of 88%, 88%, 88%, 85%, 91%, and 88%, respectively, and the prediction time was less than 5 s. The results indicated that the established models were accurate and efficient. Therefore, near infrared spectroscopy combined with machine learning algorithms can quickly predict the potential risks of clinical use of Reduning Injection in batches. Three key quality parameters that may affect clinical safety were identified by TreeNet, which provided a scientific basis for improving the safety standards of Reduning Injection.
Spectroscopy, Near-Infrared/methods* ; Drugs, Chinese Herbal/administration & dosage* ; Machine Learning ; Algorithms ; Humans ; Quality Control

Cervi Cornu Pantotrichum is a precious animal-derived Chinese medicinal material, while there are often adulterants derived from animals not specified in the Chinese Pharmacopeia in the market, which disturbs the safety of medication. This study was conducted with the aim of strengthening the quality control of Cervi Cornu Pantotrichum and standardizing the medication. To achieve digital identification of Cervi Cornu Pantotrichum from different sources, a digital identification model was constructed based on ultra-high performance liquid chromatography tandem quadrupole time-of-flight mass spectrometry(UHPLC-QTOF-MS~E) combined with an adaptive boosting algorithm(Adaboost). The young furred antlers of sika deer, red deer, elk, and reindeer were processed and then subjected to polypeptide analysis by UHPLC-QTOF-MS~E. Then, the mass spectral data reflecting the polypeptide information were obtained by digital quantification. Next, the key data were obtained by feature screening based on Gini index, and the digital identification model was constructed by Adaboost. The model was evaluated based on the recall rate, F_1 composite score, and accuracy. Finally, the results of identification based on the constructed digital identification model were validated. The results showed that when the Gini index was used to screen the data of top 100 characteristic polypeptides, the digital identification model based on Adaboost had the best performance, with the recall rate, F_1 composite score, and accuracy not less than 0.953. The validation analysis showed that the accuracy of the identification of the 10 batches of samples was as high as 100.0%. Therefore, based on UHPLC-QTOF-MS~E and Adaboost algorithm, the digital identification of Cervi Cornu Pantotrichum can be realized efficiently and accurately, which can provide reference for the quality control and original animal identification of Cervi Cornu Pantotrichum.
Animals ; Algorithms ; Antlers/chemistry* ; Boosting Machine Learning Algorithms ; Chromatography, High Pressure Liquid/methods* ; Deer ; Drugs, Chinese Herbal/chemistry* ; Mass Spectrometry/methods* ; Quality Control ; Reindeer ; Tandem Mass Spectrometry/methods* ; Tissue Extracts/analysis*

OBJECTIVES: We propose a multi-scale jaw cyst segmentation model, AConvLSTM U-Net, which is based on bidirectional dense connections and attention mechanisms to achieve accurate automatic segmentation of mandibular cyst images. METHODS: A dataset consisting of 2592 jaw cyst images was used. AConvLSTM U-Net designs a MBC on the encoding path to enhance feature extraction capabilities. A DPD was used to connect the encoder and decoder, and a bidirectional ConvLSTM was introduced in the jump connection to obtain rich semantic information. A decoding block based on scSE was then used on the decoding path to enhance the focus on important information. Finally, a DS was designed, and the model was optimized by integrating a joint loss function to further improve the segmentation accuracy. RESULTS: The experiment with AConvLSTM U-Net for jaw cyst lesion segmentation showed a MCC of 93.8443%, a DSC of 93.9067%, and a JSC of 88.5133%, outperforming all the other comparison segmentation models. CONCLUSIONS The proposed algorithm shows a high accuracy and robustness on the jaw cyst dataset, demonstrating its superior performance over many existing methods for automatic segmentation of jaw cyst images and its potential to assist clinical diagnosis.
Humans ; Jaw Cysts/diagnostic imaging* ; Algorithms ; Image Processing, Computer-Assisted/methods* ; Neural Networks, Computer

OBJECTIVES: We propose a new melanoma segmentation model, SG-UNet, to enhance the precision of melanoma segmentation in dermascopy images to facilitate early melanoma detection. METHODS: We utilized a U-shaped convolutional neural network, UNet, and made improvements to its backbone, skip connections, and downsampling pooling sections. In the backbone, with reference to the structure of VGG, we increased the number of convolutions from 10 to 13 in the downsampling part of UNet to achieve a deepened network hierarchy that allowed capture of more refined feature representations. To further enhance feature extraction and detail recognition, we replaced the traditional convolution the backbone section with self-calibrated convolution to enhance the model's ability to capture both spatial and channel dimensional features. In the pooling part, the original pooling layer was replaced by Haar wavelet downsampling to achieve more effective multi-scale feature fusion and reduce the spatial resolution of the feature map. The global attention mechanism was then incorporated into the skip connections at each layer to enhance the understanding of contextual information of the image. RESULTS: The experimental results showed that the SG-UNet model achieved significantly improved segmentation accuracy on ISIC 2017 and ISIC 2018 datasets as compared with other current state-of-the-art segmentation models, with Dice reached 92.41% and 86.62% and IoU reaching 92.31% and 86.48% on the two datasets, respectively. CONCLUSIONS The proposed model is capable of effective and accurate segmentation of melanoma from dermoscopy images.
Melanoma/diagnosis* ; Humans ; Neural Networks, Computer ; Dermoscopy/methods* ; Skin Neoplasms ; Image Processing, Computer-Assisted/methods* ; Calibration ; Algorithms

OBJECTIVES: To construct a bone tumor classification model based on feature decoupling and fusion for processing modality loss and fusing multimodal information to improve classification accuracy. METHODS: A decoupling completion module was designed to extract local and global bone tumor image features from available modalities. These features were then decomposed into shared and modality-specific features, which were used to complete the missing modality features, thereby reducing completion bias caused by modality differences. To address the challenge of modality differences that hinder multimodal information fusion, a cross-attention-based fusion module was introduced to enhance the model's ability to learn cross-modal information and fully integrate specific features, thereby improving the accuracy of bone tumor classification. RESULTS: The experiment was conducted using a bone tumor dataset collected from the Third Affiliated Hospital of Southern Medical University for training and testing. Among the 7 available modality combinations, the proposed method achieved an average AUC, accuracy, and specificity of 0.766, 0.621, and 0.793, respectively, which represent improvements of 2.6%, 3.5%, and 1.7% over existing methods for handling missing modalities. The best performance was observed when all the modalities were available, resulting in an AUC of 0.837, which still reached 0.826 even with MRI alone. CONCLUSIONS The proposed method can effectively handle missing modalities and successfully integrate multimodal information, and show robust performance in bone tumor classification under various complex missing modality scenarios.
Humans ; Bone Neoplasms/diagnosis* ; Multimodal Imaging/methods* ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed ; Image Processing, Computer-Assisted/methods* ; Algorithms

OBJECTIVES: We propose a myocardial infarction (MI) detection and localization model for improving the diagnostic accuracy for MI to provide assistance to clinical decision-making. METHODS: The proposed model was constructed based on multi-scale field residual blocks fusion modified channel attention (MSF-RB-MCA). The model utilizes lead II electrocardiogram (ECG) signals to detect and localize MI, and extracts different levels of feature information through the multi-scale field residual block. A modified channel attention for automatic adjustment of the feature weights was introduced to enhance the model's ability to focus on the MI region, thereby improving the accuracy of MI detection and localization. RESULTS: A 5-fold cross-validation test of the model was performed using the publicly available Physikalisch-Technische Bundesanstalt (PTB) dataset. For MI detection, the model achieved an accuracy of 99.96% on the test set with a specificity of 99.84% and a sensitivity of 99.99%. For MI localization, the accuracy, specificity and sensitivity were 99.81%, 99.98% and 99.65%, respectively. The performances of the model for MI detection and localization were superior to those of other comparison models. CONCLUSIONS The proposed MSF-RB-MCA model shows excellent performance in AI detection and localization based on lead II ECG signals, demonstrating its great potential for application in wearable devices.
Myocardial Infarction/diagnosis* ; Humans ; Electrocardiography/methods* ; Signal Processing, Computer-Assisted ; Algorithms ; Sensitivity and Specificity

OBJECTIVES: We propose a YOLOv11-TDSP model for improving the accuracy of dental abnormality detection on panoramic oral X-ray images. METHODS: The SHSA single-head attention mechanism was integrated with C2PSA in the backbone layer to construct a new C2PSA_SHSA attention mechanism. The computational redundancy was reduced by applying single-head attention to some input channels to enhance the efficiency and detection accuracy of the model. A small object detection layer was then introduced into the head layer to correct the easily missed and false detections of small objects. Two rounds of structured pruning were implemented to reduce the number of model parameters, avoid overfitting, and improve the average precision. Before training, data augmentation techniques such as brightness enhancement and gamma contrast adjustment were employed to enhance the generalization ability of the model. RESULTS: The experiment results showed that the optimized YOLOv11-TDSP model achieved an accuracy of 94.5%, a recall rate of 92.3%, and an average precision of 95.8% for detecting dental abnormalities. Compared with the baseline model YOLOv11n, these metrics were improved by 6.9%, 7.4%, and 5.6%, respectively. The number of parameters and computational cost of the YOLOv11-TDSP model were only 12% and 13% of those of the high-precision YOLOv11x model, respectively. CONCLUSIONS The lightweight YOLOv11-TDSP model is capable of highly accurate identification of various dental diseases on panoramic oral X-ray images.
Radiography, Panoramic/methods* ; Humans ; Algorithms ; Tooth Abnormalities/diagnostic imaging*

OBJECTIVES: To develop risk prediction models for postoperative hypothermia after transurethral holmium laser enucleation of the prostate (HoLEP) using machine learning algorithms. METHODS: We retrospectively analyzed the clinical data of 403 patients from our center (283 patients in the training set and 120in the internal validation set) and 120 patients from Weifang People's Hospital (as the external validation set). The risk prediction models were built using logistic regression, decision tree and support vector machine (SVM), and model performance was evaluated in terms of accuracy, recall, precision, F1 score and AUC. RESULTS: Operation duration, prostate weight, intraoperative irrigation volume, and being underweight were identified as the predictors of postoperative hypothermia following HoLEP. Among the 3 algorithms, SVM showed the best precision rate and accuracy in all the 3 data sets and the best area under the ROC (AUC) in the training set and validation set, followed by logistic regression, which had a similar AUC in the two data sets. SVM outperformed logistic regression and decision tree models in the validation set in precision, accuracy, recall, F1 score, and AUC, and performed well in the external validation set with better precision rate and accuracy than logistic regression and decision tree models but slightly lower recall rate, F1 index, and AUC value than the decision tree model. SVM outperformed logistic regression and decision tree models in precision, accuracy, F1 score, and AUC in the training set, but had slightly lower recall rate than the decision tree. CONCLUSIONS Among the 3 models, SVM has the best performance and generalizability for predicting post-HoLEP hypothermia risk to provide support for clinical decisions.
Humans ; Male ; Retrospective Studies ; Machine Learning ; Transurethral Resection of Prostate/adverse effects* ; Hypothermia/etiology* ; Prostatic Hyperplasia/surgery* ; Algorithms ; Lasers, Solid-State ; Risk Assessment ; Postoperative Complications ; Decision Trees ; Logistic Models ; Aged ; Middle Aged ; Support Vector Machine

OBJECTIVES: To develop a heterogeneous graph prediction method based on the fusion of multi-layer semantics and topological information for addressing the challenges in drug-target interaction prediction, including insufficient modeling of high-order semantic dependencies, lack of adaptive fusion of semantic paths, and over-smoothing of node features. METHODS: A heterogeneous graph network with multiple types of entities such as drugs, proteins, side effects, and diseases was constructed, and graph embedding techniques were used to obtain low-dimensional feature representations. An adaptive metapath search module was introduced to automatically discover semantic path combinations for guiding the propagation of high-order semantic information. A semantic aggregation mechanism integrating multi-head attention was designed to automatically learn the importance of each semantic path based on contextual information and achieve differentiated aggregation and dynamic fusion among paths. A structure-aware gated graph convolutional module was then incorporated to regulate the feature propagation intensity for suppressing redundant information and redcuing over-smoothing. Finally, the potential interactions between drugs and targets were predicted through an inner product operation. RESULTS: Compared with existing drug-target interaction prediction methods, the proposed method achieved an average improvement of 3.4% and 2.4%, 3.0% and 3.8% in terms of the area under the receiver operating characteristic curve (AUC) and the area under the precision-recall curve (AUPRC) on public datasets, respectively. CONCLUSIONS The drug-target interaction prediction method developed in this study can effectively extract complex high-order semantic and topological information from heterogeneous biological networks, thereby improving the accuracy and stability of drug-target interaction prediction. This method provides technical support and theoretical foundation for precise drug target discovery and targeted treatment of complex diseases.
Semantics ; Humans ; Drug Interactions ; Neural Networks, Computer ; Algorithms

INTRODUCTION: Obstructive sleep apnoea (OSA) is a serious but underdiagnosed condition. Demand for the gold standard diagnostic polysomnogram (PSG) far exceeds its availability. More efficient diagnostic methods are needed, even in tertiary settings. Machine learning (ML) models have strengths in disease prediction and early diagnosis. We explored the use of ML with oximetry, demographic and anthropometric data to diagnose OSA. METHODS: A total of 2,996 patients were included for modelling and divided into test and training sets. Seven commonly used supervised learning algorithms were trained with the data. Sensitivity (recall), specificity, positive predictive value (PPV) (precision), negative predictive value, area under the receiver operating characteristic curve (AUC) and F1 measure were reported for each model. RESULTS: In the best performing four-class model (neural network model predicting no, mild, moderate or severe OSA), a prediction of moderate and/or severe disease had a combined PPV of 94%; one out of 335 patients had no OSA and 19 had mild OSA. In the best performing two-class model (logistic regression model predicting no-mild vs. moderate-severe OSA), the PPV for moderate-severe OSA was 92%; two out of 350 patients had no OSA and 26 had mild OSA. CONCLUSION Our study showed that the prediction of moderate-severe OSA in a tertiary setting with an ML approach is a viable option to facilitate early identification of OSA. Prospective studies with home-based oximeters and analysis of other oximetry variables are the next steps towards formal implementation.
Humans ; Oximetry/methods* ; Sleep Apnea, Obstructive/diagnosis* ; Male ; Female ; Middle Aged ; Machine Learning ; Polysomnography ; Adult ; Anthropometry ; ROC Curve ; Aged ; Algorithms ; Predictive Value of Tests ; Sensitivity and Specificity ; Neural Networks, Computer ; Demography