Traditional Chinese medicine(TCM) decoction pieces are a core carrier for the inheritance and innovation of TCM, and their quality and safety are critical to public health and the sustainable development of the industry. Conventional quality control models, while having established a well-developed system through long-term practice, still face challenges such as relatively long inspection cycles, insufficient objectivity in characterizing complex traits, and urgent needs for improving the efficiency of integrating multidimensional quality information when confronted with the dual demands of large-scale production and precision quality control. With the rapid development of artificial intelligence, machine learning can deeply analyze multidimensional data of the morphology, spectroscopy, and chemical fingerprints of decoction pieces by constructing high-dimensional feature space analysis models, significantly improving the standardization level and decision-making efficiency of quality evaluation. This article reviews the research progress in the application of machine learning in the processing, production, and rapid quality evaluation of TCM decoction pieces. It further analyzes current challenges in technological implementation and proposes potential solutions, offering theoretical and technical references to advance the digital and intelligent transformation of the industry.
Machine Learning ; Drugs, Chinese Herbal/standards* ; Quality Control ; Medicine, Chinese Traditional/standards* ; Humans

Cardiovascular diseases and psychological disorders represent two major threats to human physical and mental health. Research on electrocardiogram (ECG) signals offers valuable opportunities to address these issues. However, existing methods are constrained by limitations in understanding ECG features and transferring knowledge across tasks. To address these challenges, this study developed a multi-resolution feature encoding network based on residual networks, which effectively extracted local morphological features and global rhythm features of ECG signals, thereby enhancing feature representation. Furthermore, a model compression-based continual learning method was proposed, enabling the structured transfer of knowledge from simpler tasks to more complex ones, resulting in improved performance in downstream tasks. The multi-resolution learning model demonstrated superior or comparable performance to state-of-the-art algorithms across five datasets, including tasks such as ECG QRS complex detection, arrhythmia classification, and emotion classification. The continual learning method achieved significant improvements over conventional training approaches in cross-domain, cross-task, and incremental data scenarios. These results highlight the potential of the proposed method for effective cross-task knowledge transfer in ECG analysis and offer a new perspective for multi-task learning using ECG signals.
Humans ; Electrocardiography/methods* ; Mental Health ; Algorithms ; Signal Processing, Computer-Assisted ; Machine Learning ; Arrhythmias, Cardiac/diagnosis* ; Cardiovascular Diseases ; Neural Networks, Computer ; Mental Disorders

Emotion classification and recognition is a crucial area in emotional computing. Physiological signals, such as electroencephalogram (EEG), provide an accurate reflection of emotions and are difficult to disguise. However, emotion recognition still faces challenges in single-modal signal feature extraction and multi-modal signal integration. This study collected EEG, electromyogram (EMG), and electrodermal activity (EDA) signals from participants under three emotional states: happiness, sadness, and fear. A feature-weighted fusion method was applied for integrating the signals, and both support vector machine (SVM) and extreme learning machine (ELM) were used for classification. The results showed that the classification accuracy was highest when the fusion weights were set to EEG 0.7, EMG 0.15, and EDA 0.15, achieving accuracy rates of 80.19% and 82.48% for SVM and ELM, respectively. These rates represented an improvement of 5.81% and 2.95% compared to using EEG alone. This study offers methodological support for emotion classification and recognition using multi-modal physiological signals.
Humans ; Emotions/physiology* ; Electroencephalography ; Support Vector Machine ; Electromyography ; Signal Processing, Computer-Assisted ; Galvanic Skin Response/physiology* ; Machine Learning ; Male

The risk prediction of paroxysmal atrial fibrillation (PAF) is a challenge in the field of biomedical engineering. This study integrated the advantages of machine learning feature engineering and end-to-end modeling of deep learning to propose a PAF risk prediction method based on multimodal feature fusion. Additionally, the study utilized four different feature selection methods and Pearson correlation analysis to determine the optimal multimodal feature set, and employed random forest for PAF risk assessment. The proposed method achieved accuracy of (92.3 ± 2.1)% and F1 score of (91.6 ± 2.9)% in a public dataset. In a clinical dataset, it achieved accuracy of (91.4 ± 2.0)% and F1 score of (90.8 ± 2.4)%. The method demonstrates generalization across multi-center datasets and holds promising clinical application prospects.
Humans ; Atrial Fibrillation/diagnosis* ; Machine Learning ; Deep Learning ; Risk Assessment/methods*

Alzheimer's disease (AD) classification models usually segment the entire brain image into voxel blocks and assign them labels consistent with the entire image, but not every voxel block is closely related to the disease. To this end, an AD auxiliary diagnosis framework based on weakly supervised multi-instance learning (MIL) and multi-scale feature fusion is proposed, and the framework is designed from three aspects: within the voxel block, between voxel blocks, and high-confidence voxel blocks. First, a three-dimensional convolutional neural network was used to extract deep features within the voxel block; then the spatial correlation information between voxel blocks was captured through position encoding and attention mechanism; finally, high-confidence voxel blocks were selected and combined with multi-scale information fusion strategy to integrate key features for classification decision. The performance of the model was evaluated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) and Open Access Series of Imaging Studies (OASIS) datasets. Experimental results showed that the proposed framework improved ACC and AUC by 3% and 4% on average compared with other mainstream frameworks in the two tasks of AD classification and mild cognitive impairment conversion classification, and could find the key voxel blocks that trigger the disease, providing an effective basis for AD auxiliary diagnosis.
Alzheimer Disease/diagnosis* ; Humans ; Neuroimaging/methods* ; Neural Networks, Computer ; Brain/diagnostic imaging* ; Magnetic Resonance Imaging ; Deep Learning ; Machine Learning

Retrieving keyframes most relevant to text from small intestine videos with given labels can efficiently and accurately locate pathological regions. However, training directly on raw video data is extremely slow, while learning visual representations from image-text datasets leads to computational inconsistency. To tackle this challenge, a small bowel video keyframe retrieval based on multi-modal contrastive learning (KRCL) is proposed. This framework fully utilizes textual information from video category labels to learn video features closely related to text, while modeling temporal information within a pretrained image-text model. It transfers knowledge learned from image-text multimodal models to the video domain, enabling interaction among medical videos, images, and text data. Experimental results on the hyper-spectral and Kvasir dataset for gastrointestinal disease detection (Hyper-Kvasir) and the Microsoft Research video-to-text (MSR-VTT) retrieval dataset demonstrate the effectiveness and robustness of KRCL, with the proposed method achieving state-of-the-art performance across nearly all evaluation metrics.
Humans ; Video Recording ; Intestine, Small/diagnostic imaging* ; Machine Learning ; Image Processing, Computer-Assisted/methods* ; Algorithms

Neurodegenerative diseases (NDDs) are a group of heterogeneous neurological disorders that can cause progressive loss of neurons in the central nervous system or peripheral nervous system, resulting in a decline in motor function. Motion capture, as a high-precision and high-resolution technology for capturing human motion data, drives NDDs motor assessment to effectively extract kinematic features and thus assess the patient's motor ability or disease severity. This paper focuses on the recent research progress in motor assessment of NDDs driven by motion capture data. Based on a brief introduction of NDDs motor assessment datasets, we categorized the assessment methods into three types according to the way of feature extraction and processing: NDDs motor assessment methods based on statistical analysis, machine learning and deep learning. Then, we comparatively analyzed the technical points and characteristics of the three types of methods from the aspects of data composition, data preprocessing, assessment methods, assessment purposes and effects. Finally, we discussed and prospected the development trends of NDDs motor assessment.
Humans ; Neurodegenerative Diseases/diagnosis* ; Machine Learning ; Biomechanical Phenomena ; Deep Learning ; Motion ; Motion Capture

Brain age prediction, as a significant approach for assessing brain health and early diagnosing neurodegenerative diseases, has garnered widespread attention in recent years. Electroencephalogram (EEG), an non-invasive, convenient, and cost-effective neurophysiological signal, offers unique advantages for brain age prediction due to its high temporal resolution and strong correlation with brain functional states. Despite substantial progress in enhancing prediction accuracy and generalizability, challenges remain in data quality and model interpretability. This review comprehensively examined the advancements in EEG-based brain age prediction, detailing key aspects of data preprocessing, feature extraction, model construction, and result evaluation. It also summarized the current applications of machine learning and deep learning methods in this field, analyzed existing issues, and explored future directions to promote the widespread application of EEG-based brain age prediction in both clinical and research settings.
Humans ; Electroencephalography/methods* ; Brain/physiology* ; Machine Learning ; Aging/physiology* ; Deep Learning ; Signal Processing, Computer-Assisted

Protein lysine β-hydroxybutyrylation (Kbhb) is a newly discovered post-translational modification associated with a wide range of biological processes. Identifying Kbhb sites is critical for a better understanding of its mechanism of action. However, biochemical experimental methods for probing Kbhb sites are costly and have a long cycle. Therefore, a feature embedding learning method based on the Transformer encoder was proposed to predict Kbhb sites. In this method, amino acid residues were mapped into numerical vectors according to their amino acid class and position in a learnable feature embedding method. Then the Transformer encoder was used to extract discriminating features, and the bidirectional long short-term memory network (BiLSTM) was used to capture the correlation between different features. In this paper, a benchmark dataset was constructed, and a Kbhb site predictor, AutoTF-Kbhb, was implemented based on the proposed method. Experimental results showed that the proposed feature embedding learning method could extract effective features. AutoTF-Kbhb achieved an area under curve (AUC) of 0.87 and a Matthews correlation coefficient (MCC) of 0.37 on the independent test set, significantly outperforming other methods in comparison. Therefore, AutoTF-Kbhb can be used as an auxiliary means to identify Kbhb sites.
Protein Processing, Post-Translational ; Lysine/chemistry* ; Proteins/chemistry* ; Machine Learning ; Algorithms

With the rising incidence of breast cancer among women, neoadjuvant chemotherapy (NAC) is becoming increasingly crucial as a preoperative treatment modality, enabling tumor downstaging and volume reduction. However, its efficacy varies significantly among patients, underscoring the importance of predicting pathological complete response (pCR) following NAC. Early research relied on statistical methods to integrate clinical data for predicting treatment outcomes. With the advent of artificial intelligence (AI), traditional machine learning approaches were subsequently employed for efficacy prediction. Deep learning emerged to dominate this field, and demonstrated the capability to automatically extract imaging features and integrate multimodal data for pCR prediction. This review comprehensively examined the applications and limitations of these three methodologies in predicting breast cancer pCR. Future efforts must prioritize the development of superior predictive models to achieve precise predictions, integrate them into clinical workflows, enhance patient care, and ultimately improve therapeutic outcomes and quality of life.
Humans ; Breast Neoplasms/pathology* ; Neoadjuvant Therapy ; Artificial Intelligence ; Female ; Machine Learning ; Deep Learning ; Chemotherapy, Adjuvant ; Treatment Outcome