Biomedical texts provide important data for investigating drug-drug interactions (DDIs) in the field of pharmacovigilance. Although researchers have attempted to investigate DDIs from biomedical texts and predict unknown DDIs, the lack of accurate manual annotations significantly hinders the performance of machine learning algorithms. In this study, a new DDI prediction framework, Subgraph Enhance model, was developed for DDI (SubGE-DDI) to improve the performance of machine learning algorithms. This model uses drug pairs knowledge subgraph information to achieve large-scale plain text prediction without many annotations. This model treats DDI prediction as a multi-class classification problem and predicts the specific DDI type for each drug pair (e.g. Mechanism, Effect, Advise, Interact and Negative). The drug pairs knowledge subgraph was derived from a huge drug knowledge graph containing various public datasets, such as DrugBank, TwoSIDES, OffSIDES, DrugCentral, EntrezeGene, SMPDB (The Small Molecule Pathway Database), CTD (The Comparative Toxicogenomics Database) and SIDER. The SubGE-DDI was evaluated from the public dataset (SemEval-2013 Task 9 dataset) and then compared with other state-of-the-art baselines. SubGE-DDI achieves 83.91% micro F1 score and 84.75% macro F1 score in the test dataset, outperforming the other state-of-the-art baselines. These findings show that the proposed drug pairs knowledge subgraph-assisted model can effectively improve the prediction performance of DDIs from biomedical texts.

Background: It is challenging to predict the outcome of the pregnancy when fetal heart activity is detected in early pregnancy. However, an accurate prediction is of importance for obstetricians as it helps to provide appropriate consultancy and determine the frequency of ultrasound examinations. The purpose of this study was to investigate the role of the convolutional neural network (CNN) in the prediction of spontaneous miscarriage risk through the analysis of early ultrasound gestational sac images. Methods: A total of 2196 ultrasound images from 1098 women with early singleton pregnancies of gestational age between 6 and 8 weeks were used for training a CNN for the prediction of the miscarriage in the retrospective study. The patients who had positive fetal cardiac activity on their first ultrasound but then experienced a miscarriage were enrolled. The control group was randomly selected in the same database from the fetuses confirmed to be normal during follow-up. Diagnostic performance of the algorithm was validated and tested in two separate test sets of 136 patients with 272 images, respectively. Performance in prediction of the miscarriage was compared between the CNN and the manual measurement of ultrasound characteristics in the prospective study. Results: The accuracy of the predictive model was 80.32% and 78.1% in the retrospective and prospective study, respectively. The area under the receiver operating characteristic curve (AUC) for classification was 0.857 (95% confidence interval [CI], 0.793–0.922) in the retrospective study and 0.885 (95%CI, 0.846–0.925) in the prospective study, respectively. Correspondingly, the predictive power of the CNN was higher compared with manual ultrasound characteristics, for which the AUCs of the crown-rump length combined with fetal heart rate was 0.687 (95%CI, 0.587–0.775). Conclusions: The CNN model showed high accuracy for predicting miscarriage through the analysis of early pregnancy ultrasound images and achieved better performance than that of manual measurement.

Introduction: Adverse Drug Reactions (ADRs), a widespread phenomenon in clinical drug treatment, are often associated with a high risk of morbidity and even death. Drugs and changes in gene expression are the two important factors that affect whether and how adverse reactions occur. Notably, pharmacogenomics data have recently become more available and could be used to predict ADR occurrence. However, there is a challenge in effectively analyzing the massive data lacking guidance on mutual relationship for ADRs prediction.Methods: We constructed separate similarity features for drugs and ADRs using pharmacogenomics data from the Comparative Toxicogenomics Database [CTD, including Chemical-Gene Interactions (CGIs) and Gene-Disease Associations (GDAs)]. We proposed a novel deep learning architecture, DGANet, based on the constructed features for ADR prediction. The algorithm uses Convolutional Neural Networks (CNN) and cross-features to learn the latent drug-gene-ADR associations for ADRs prediction.Results and Discussion: The performance of DGANet was compared to three state-of-the-art algorithms with different genomic features. According to the results, GDANet outperformed the benchmark algorithms (AUROC = 92.76%, AUPRC = 92.49%), demonstrating a 3.36% AUROC and 4.05% accuracy improvement over the cutting-edge algorithms. We further proposed new genomic features that improved DGANet’s predictive capability. Moreover, case studies on top-ranked candidates confirmed DGANet’s ability to predict new ADRs.

Artificial intelligence (AI)-assisted prediction of adverse drug reactions (ADRs) has significant potential for improving drug safety and reducing financial costs. Early studies often relied on limited dimensions such as the molecular structure of drugs or interactions with biomolecules. In contrast, integrating these characteristics provides valuable insights into ADR predictions from multiple perspectives, enhancing the comprehensiveness and accuracy of the prediction models. In addition, previous studies have focused on whether a specific adverse drug reaction occurs with a particular drug, ignoring the fact that multiple adverse drug reactions may occur concurrently with a single drug. To address these, we developed a predictor that identifies ADRs early in drug discovery, using a deep learning model designed to fuse multiple drug characteristics. Our approach employed four modules to extract one- and two-dimensional sequence structure information of drug molecules, drug–protein interaction data, and drug similarity. A fusion model integrated these characteristics to predict the precise probability of ADRs. The receiver operating characteristic–area under curve (ROC-AUC), area under precision–recall curve (AUPR), and F1 scores on the benchmark dataset are 0.7002, 0.6619, and 0.6330, respectively. The AUPR is significantly improved compared to the conventional multi-label classifier (from 64.02% to 66.19%). In addition, we compared the results with the state-of-the-art methods on LIU’s dataset and the AUPR increased from 34.65% to 68.82%, which shows that our model outperforms them in terms of accuracy and robustness. Ablation experiments further validated the effectiveness of the individual modules. This model accurately predicted the probability of various ADR classes by integrating comprehensive information, thereby offering significant value in enhancing monitoring measures for new drug development and clinical use.

Adverse drug reactions (ADRs) are among the global public health events that seriously endanger human life and cause high economic burdens. Therefore, predicting the possibility of their occurrence and taking early and effective response measures is of great significance. Constructing a correlation matrix between drugs and their adverse reactions, followed by effective correlation data mining, is one of the current strategies to predict ADRs using accessible public data. Since the number of known ADRs in real-world data is far less than the number of their unknown counterparts, the drug-ADR association matrix is very sparse, which greatly affects the classification performance of machine learning methods. To effectively address the problem of sparsity, we proposed a novel weighted pseudo-labeling framework that mines potential unknown drug-ADR pairs by integrating multiple weighted matrix factorization (MF) models and treating them as pseudo-labeled drug-ADR pairs. Pseudo-labeled data is added to the training set, and the MF model is fine-tuned to improve the classification performance. To prevent overfitting to easily found pseudo-labels and improve the quality of pseudo-labels, a novel weighting approach for pseudo-labels was adopted. This paper reproduces the baselines under the same experimental conditions to evaluate the performance of the proposed method on sparse data from the Side Effect Resource (SIDER) database. Experimental results showed that our method outperformed other baselines in the Area Under Precision-Recall and F1-scores and still maintained the best performance in sparser scenarios. Furthermore, we conducted a case study, and the results showed that our proposed framework efficiently predicted ADRs in the real world.