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

OBJECTIVES: To analyze the differentially expressed exosomal miRNAs in patients with chronic heart failure (CHF) complicated by hyperuricemia (HUA) and explore their potential as novel diagnostic molecular markers and their target genes. METHODS: This study was conducted among 30 CHF patients with HUA (observation group) and 30 healthy volunteers (control group) enrolled between September, 2020 and September, 2023. Peripheral blood samples were collected from 6 CHF patients with HUA for analyzing exosomal miRNAs by high-throughput sequencing, and the results were validated in the remaining 24 patients using qRT-PCR. GO and KEGG enrichment analyses were performed to predict the the target genes of the identified differential miRNAs. We also validated the differentially expressed miRNAs by animal experiment. RESULTS: A total of 42 differentially expressed exosomal miRNAs were detected in observation group by high-throughput sequencing; among them, miR-27a-5p was significantly upregulated (P=0.000179), and miR-139-3p was significantly downregulated (P=0.000058). In the 24 patients with both CHF and PUA, qRT-PCR validated significant upregulation of miR-27a-5p (P=0.004) and downregulation of miR-139-3p (P=0.005) in serum exosomes. When combined, miR-27a-5p and miR-139-3p had a maximum area under the curve (AUC) of 0.899 (95% CI: 0812-0.987) for predicting CHF complicated by HUA. GO and KEGG enrichment analyses suggested that the differential expressions of miR-27a-5p and miR-139-3p was associated with the activation of the AMPK-mTOR signaling pathway to activate the autophagic response. We obtained the same conclusion from animal experiment. CONCLUSIONS Upregulated exosomal miR-27a-5p combined with downregulated exosomal miR-139-3p expression can serve as a novel molecular marker for diagnosis of CHF complicated by HUA, and their differential expression may promote autophagy in cardiomyocytes by activating the AMPK-mTOR signaling pathway.
Humans ; Hyperuricemia/diagnosis* ; Heart Failure/genetics* ; MicroRNAs/metabolism* ; Exosomes/metabolism* ; Chronic Disease ; Male ; Female ; Middle Aged ; Animals

OBJECTIVES: To evaluate the performance of a multi-constraint representation learning classification model for identifying ovarian cancer with missing laboratory indicators. METHODS: Tabular data with missing laboratory indicators were collected from 393 patients with ovarian cancer and 1951 control patients. The missing ovarian cancer laboratory indicator features were projected to the latent space to obtain a classification model using the representational learning classification model based on discriminative learning and mutual information coupled with feature projection significance score consistency and missing location estimation. The proposed constraint term was ablated experimentally to assess the feasibility and validity of the constraint term by accuracy, area under the ROC curve (AUC), sensitivity, and specificity. Cross-validation methods and accuracy, AUC, sensitivity and specificity were also used to evaluate the discriminative performance of this classification model in comparison with other interpolation methods for processing of the missing data. RESULTS: The results of the ablation experiments showed good compatibility among the constraints, and each constraint had good robustness. The cross-validation experiment showed that for identification of ovarian cancer with missing laboratory indicators, the AUC, accuracy, sensitivity and specificity of the proposed multi-constraints representation-based learning classification model was 0.915, 0.888, 0.774, and 0.910, respectively, and its AUC and sensitivity were superior to those of other interpolation methods. CONCLUSIONS The proposed model has excellent discriminatory ability with better performance than other missing data interpolation methods for identification of ovarian cancer with missing laboratory indicators.
Female ; Humans ; Ovarian Neoplasms/diagnosis* ; Machine Learning ; ROC Curve

OBJECTIVES: To explore the value of serum cystatin C (CysC) levels in evaluating renal prognosis in IgA nephropathy (IgAN) patients. METHODS: We retrospectively collected the clinical data of IgAN patients diagnosed by renal biopsy at Guangdong Provincial People's Hospital from January, 2014 to December, 2018. Based on baseline serum CysC levels, the patients were divided into high serum CysC (>1.03 mg/L) group and normal serum CysC (≤1.03 mg/L) group. The composite endpoint for poor renal prognosis was defined as ≥50% decline in estimated glomerular filtration rate (eGFR) and/or progression to end-stage renal disease (ESRD). Lasso regression, multivariate Cox regression and Kaplan-Meier survival analysis were used to identify the risk factors and compare renal survival rates between the two groups. Smooth curves fitting and threshold effect analysis were used to explore the relationship between serum CysC levels and the outcomes. A nomogram model was constructed and its predictive performance was evaluated using concordance index, calibration curve, receiver operating characteristic (ROC) curve and the area under curve (AUC). RESULTS: A total of 356 IgAN patients were enrolled, who were followed up for 4.65±0.93 years. The composite endpoint occurred in 74 patients. High serum CysC was identified as an independent risk factor for poor renal prognosis in IgAN (HR=2.142, 95% CI 1.222 to 3.755), and the patients with high serum CysC levels had a lower renal survival rate (Log-rank χ²=47.970, P<0.001). In patients with serum CysC below 2.12 mg/L, a higher CysC level was associated with an increased risk of poor renal prognosis (β=3.487, 95% CI: 2.561-4.413, P<0.001), while above this level, the increase of the risk was not significant (β=0.676, 95% CI: -0.642-1.995, P=0.315). The nomogram model based on serum CysC and 3 other independent risk factors demonstrated good internal validity with a concordance index of 0.873 (95% CI: 0.839-0.907) and an AUC of 0.909 (95% CI: 0.873-0.945). CONCLUSIONS Serum CysC levels are associated with renal prognosis in IgAN patients, and high serum CysC an independent risk factor for poor renal prognosis.
Humans ; Glomerulonephritis, IGA/diagnosis* ; Cystatin C/blood* ; Prognosis ; Risk Factors ; Retrospective Studies ; Glomerular Filtration Rate ; Kidney Failure, Chronic ; Male ; Female ; Adult ; Nomograms ; Middle Aged

OBJECTIVES: The presence of multi-scale skin lesion regions and image noise interference and limited resources of auxiliary diagnostic equipment affect the accuracy of skin disease detection in skin disease detection tasks. To solve these problems, we propose a highly efficient and lightweight skin disease detection model using an improved RT-DETR model. METHODS: A lightweight FasterNet was introduced as the backbone network and the FasterNetBlock module was parametrically refined. A Convolutional and Attention Fusion Module (CAFM) was used to replace the multi-head self-attention mechanism in the neck network to enhance the ability of the AIFI-CAFM module for capturing global dependencies and local detail information. The DRB-HSFPN feature pyramid network was designed to replace the Cross-Scale Feature Fusion Module (CCFM) to allow the integration of contextual information across different scales to improve the semantic feature expression capacity of the neck network. Finally, combining the advantages of Inner-IoU and EIoU, the Inner-EIoU was used to replace the original loss function GIOU to further enhance the model's inference accuracy and convergence speed. RESULTS: The experimental results on the HAM10000 dataset showed that the improved RT-DETR model, as compared with the original model, had increased mAP@50 and mAP@50:95 by 4.5% and 2.8%, respectively, with a detection speed of 59.1 frames per second (FPS). The improved model had a parameter count of 10.9 M and a computational load of 19.3 GFLOPs, which were reduced by 46.0% and 67.2% compared to those of the original model, validating the effectiveness of the improved model. CONCLUSIONS The proposed SD-DETR model significantly improves the performance of skin disease detection tasks by effectively extracting and integrating multi-scale features while reducing both parameter count and computational load.
Humans ; Skin Diseases/diagnosis* ; Skin/pathology* ; Neural Networks, Computer ; Algorithms

OBJECTIVES: To design a deep learning model that balances model complexity and performance to enable its integration into wearable ECG monitoring devices for automated diagnosis of atrial fibrillation. METHODS: This study was performed based on data from 84 patients with atrial fibrillation, 25 patients with atrial fibrillation, and 18 subjects without obvious arrhythmia collected from the publicly available datasets LTAFDB, AFDB, and NSRDB, respectively. A lightweight attention network based on depthwise separable convolution and fusion of channel-spatial information, namely DSC-AttNet, was proposed. Depthwise separable convolution was introduced to replace standard convolution and reduce model parameters and computational complexity to realize high efficiency and light weight of the model. The multilayer hybrid attention mechanism was embedded to compute the attentional weights of the channels and spatial information at different scales to improve the feature expression ability of the model. Ten-fold cross-validation was performed on LTAFDB, and external independent testing was conducted on AFDB and NSRDB datasets. RESULTS: DSC-AttNet achieved a ten-fold average accuracy of 97.33% and a precision of 97.30% on the test set, both of which outperformed the other 4 comparison models as well as the 3 classical models. The accuracy of the model on the external test set reached 92.78%, better than those of the 3 classical models. The number of parameters of DSC-AttNet was 1.01M, and the computational volume was 27.19G, both smaller than the 3 classical models. CONCLUSIONS This proposed method has a smaller complexity, achieves better classification performance, and has a better generalization ability for atrial fibrillation classification.
Atrial Fibrillation/diagnosis* ; Humans ; Electrocardiography ; Deep Learning ; Wearable Electronic Devices ; Neural Networks, Computer

OBJECTIVES: To construct vocal recognition classification models using 6 machine learning algorithms and vocal emotional characteristics of individuals with subthreshold depression to facilitate early identification of subthreshold depression. METHODS: We collected voice data from both normal individuals and participants with subthreshold depression by asking them to read specifically chosen words and texts. From each voice sample, 384-dimensional vocal emotional feature variables were extracted, including energy feature, Meir frequency cepstrum coefficient, zero cross rate feature, sound probability feature, fundamental frequency feature, difference feature. The Recursive Feature Elimination (RFE) method was employed to select voice feature variables. Classification models were then built using the machine learning algorithms Adaptive Boosting (AdaBoost), Random Forest (RF), Linear Discriminant Analysis (LDA), Logistic Regression (LR), Lasso Regression (LRLasso), and Support Vector Machine (SVM), and the performance of these models was evaluated. To assess generalization capability of the models, we used real-world speech data to evaluate the best speech recognition classification model. RESULTS: The AdaBoost, RF, and LDA models achieved high prediction accuracies of 100%, 100%, and 93.3% on word-reading speech test set, respectively. In the text-reading speech test set, the accuracies of the AdaBoost, RF, and LDA models were 90%, 80%, and 90%, respectively, while the accuracies of the other 3 models were all below 80%. On real-world word-reading and text-reading speech data, the classification models using AdaBoost and Random Forest still achieved high predictive accuracies (91.7% and 80.6% for AdaBoost and 86.1% and 77.8% for Random, respectively). CONCLUSIONS Analyzing vocal emotional characteristics allows effective identification of individuals with subthreshold depression. The AdaBoost and RF models show excellent performance for classifying subthreshold depression individuals, and may thus potentially offer valuable assistance in the clinical and research settings.
Humans ; Machine Learning ; Emotions ; Depression/diagnosis* ; Algorithms ; Voice ; Support Vector Machine ; Male ; Female

BACKGROUND

Six-pyruvoyl-tetrahydrobiopterin synthase (6-PTPS) deficiency is an inherited metabolic disorder which results in tetrahydrobiopterin (BH4) deficiency causing hyperphenylalaninemia.

This study aimed to describe the clinical, biochemical, and radiologic profiles, and neurologic and developmental outcomes of patients diagnosed with 6-pyruvoyl tetrahydrobiopterin (PTPS) deficiency through newborn screening and confirmed by BH4 loading test, pterin analysis, and gene sequencing who were following-up with the metabolic team.

The research was a single-center descriptive case series study design that was done at the Philippine General Hospital, a tertiary government hospital. The clinical, biochemical, radiologic profiles and neurodevelopmental evaluation of each patient were described.

Nine patients from 1 year 2 months to 14 years 5 months of age were enrolled in the study. Clinical manifestations before treatment were hypotonia, poor suck, and seizure. The most common clinical manifestation even after treatment initiation was seizure. The mean phenylalanine level on newborn screening was 990.68 umol/L, but after treatment was started, mean levels ranged from 75.69 to 385.09 umol/L. Two of the patients had focal atrophy of the posterior lobe on brain imaging. Pathogenic variants on molecular analysis were all missense, with two predominant variants, c.155A>G and c.58T>C. Eight of the nine patients had varying degrees of developmental delay or intellectual disability, while the remaining patient had signs of a learning disorder.

Newborn screening has played a crucial role in the early identification and management of patients with hyperphenylalaninemia due to 6-PTPS deficiency. Confirmation of diagnosis through determination of DHPR activity, urine pterins and/or molecular analysis is necessary for appropriate management. However, despite early initiation of treatment, neurodevelopmental findings of patients with 6-PTPS deficiency were still unsatisfactory.

Glycogen storage disease type Ⅱ （GSD Ⅱ）， also known as Pompe disease， is a common autosomal recessive lysosomal storage disease with predominantly muscle tissue involvement， and it is caused by defects in the GAA gene which encode acid α-D-glucosidase in lysosomes. According to the age of onset and the main organs involved， it is classified into infant-onset Pompe disease （IOPD） and late-onset Pompe disease（LOPD）. The diagnosis of this disease depends on the reduction in GAA enzyme activity， the detection of GAA gene mutations， and muscle tissue biopsy， and early diagnosis and treatment are crucial for prognosis. Recombinant human GAA（rhGAA） enzyme replacement therapy prepared by the gene recombination technology is currently the main disease-modifying treatment method for Pompe disease， among which the earliest drug alglucosidase α has shown good efficacy in improving muscle strength and respiratory function and prolonging survival time， and the new-generation rhGAA drugs avalglucosidase α and cipaglucosidase alfa provide new options， especially for patients with poor outcomes and severe symptoms. Substrate ablation therapy and gene therapy are still under exploration， and disease-modifying therapies combined with nutritional and exercise therapies and multidisciplinary long-term management will achieve twice the result with half the effort.
Diagnosis

Mitochondrial diseases are a group of hereditary disorders characterized by impaired oxidative phosphorylation in the mitochondrial respiratory chain caused by defects in either mitochondrial DNA or nuclear DNA， and such diseases have complex and diverse clinical manifestations and often involve multiple organs and systems， with the main manifestation of lesions in the nervous system and muscles due to their high energy demands. At present， there is still a lack of effective therapies for most mitochondrial diseases， and therefore， multidisciplinary management is essential in clinical practice， integrating various therapeutic approaches to provide personalized treatment regimens for patients with mitochondrial diseases. The primary treatment principle involves the timely correction of pathological and physiological abnormalities through pharmacological interventions， dietary modifications， and exercise management， along with the prompt treatment of system-specific impairments and the prevention of potential complications.
Diagnosis