Haemorrhagic stroke accounts for approximately 31.52% of all stroke cases, and the most common origin is hypertension. However, little is known about the method to identify high-risk populations of hypertensive intracerebral haemorrhage. The results showed that the angle between the middle cerebral artery and the internal carotid artery (AMIC), the distance between the beginning of the median artery and superior trunk (DMS), and the density (CT value) of the lenticulostriate artery (CTL) were statistically significant enough to cause intracerebral haemorrhage. In addition, we chose these three potential features for the ensemble learning classification model. Our developed ensemble-learning method outperforms not only previous work but also three other classic classification methods based on accuracy measurements. The developed mathematical model in the present study is efficient in predicting the probability of intracerebral haemorrhage.
1 January 2021 in Mathematical Biosciences and Engineering
Estimating the binding affinity between proteins and drugs is very important in the application of structure-based drug design. Currently, applying machine learning to build the protein-ligand binding affinity prediction model, which is helpful to improve the performance of classical scoring functions, has attracted many scientists' attention. In this paper, we have developed an affinity prediction model called GAT-Score based on graph attention network (GAT). The protein-ligand complex is represented by a graph structure, and the atoms of protein and ligand are treated in the same manner. Two improvements are made to the original graph attention network. Firstly, a dynamic feature mechanism is designed to enable the model to deal with bond features. Secondly, a virtual super node is introduced to aggregate node-level features into graph-level features, so that the model can be used in the graph-level regression problems. PDBbind database v.2018 is used to train the model. Finally, the performance of GAT-Score was tested by the scheme $C_s$ (Core set as the test set) and CV (Cross-Validation). It has been found that our results are better than most methods from machine learning models with traditional molecular descriptors.
Virtual screening has become a successful alternative and complementary technique to experimental high-throughput screening technologies for drug design. This paper proposed a target-specific virtual screening method based on ensemble learning named ENS-VS.
Preoperative identification of axillary lymph node metastasis can play an important role in treatment selection strategy and prognosis evaluation. This study aimed to establish a clinical nomogram based on lymph node images to predict lymph node metastasis in breast cancer patients. A total of 193 patients with non-specific invasive breast cancer were divided into training (n = 135) and validation set (n = 58). Radiomics features were extracted from lymph node images instead of tumor region, and the least absolute shrinkage and selection operator logistic algorithm was used to select the extracted features and generate radiomics score. Then, the important clinical factors and radiomics score were integrated into a nomogram. A receiver operating characteristic curve was used to evaluate the nomogram, and the clinical benefit of using the nomogram was evaluated by decision curve analysis. We found that clinical N stage and radiomics score were independent clinical predictors. Besides, the nomogram accurately predicted axillary lymph node metastasis, yielding an area under the receiver operating characteristic curve of 0.95 (95% confidence interval 0.93-0.98) in the validation set, indicating satisfactory calibration. Decision curve analysis confirmed that the nomogram had higher clinical utility than clinical N stage or radiomics score alone. Overall, the nomogram based on radiomics features and clinical factors can help radiologists to predict axillary lymph node metastasis preoperatively and provide valuable information for individual treatment.
18 July 2025 in Journal of Computer-Aided Molecular Design
Speech interfaces are becoming ubiquitous in medical and industrial devices, yet undergraduate laboratory teaching of speech signal processing still works almost exclusively on air-conducted recordings and supervised pipelines, ignoring both the physiology of alternative sensing channels and the data scarcity that dominates real deployments. This paper reports the design of an experimental teaching system that converts an active bone-conducted speech research programme, centred on a pre-trained Transformer with low-rank driven sparsity bootstrapping (TransLRC), into a four-tier undergraduate laboratory curriculum with an engineering layer. Four principles govern the design: constructive alignment, experiential progression, the research-teaching nexus, and the CDIO lifecycle. Each tier is anchored to a bottleneck that is genuinely open in the source research: characterising the body-channel distortion of bone-conducted speech, pre-training spectral representations without paired data, pruning attention through low-rank structure, and evaluating generalisation to unseen speakers under real-time budgets. Thirty-two laboratory units (64 contact hours) are organised into six modules with staged delivery and criterion-referenced assessment that admits documented negative results, mirroring the source study’s finding that pre-training and low-rank constraints contribute separable, measurable gains. Because the source protocol runs end-to-end on a single consumer GPU with a low-cost bone-conduction sensor, the curriculum targets faithful reproduction rather than scale-reduced approximation, giving students an unusually direct path from published result to reproduced result.