Pragmatic screening for heart failure in the general population using an electrocardiogram‐based neural network

Clinical prediction rules could help identify children at risk of slowed growth after an episode of diarrheal illness, and these rules may be generalizable to all children, regardless of diarrhea status.

Background Risk stratification of heart failure (HF) in individuals without prevalent heart disease is crucial for early intervention and prevention. Electrocardiogram (ECG)-based artificial intelligence models have shown promise when predicting HF events at long follow-up periods. However, they are still limited by imbalanced case-control data and small labelled datasets. Moreover, the electrophysiological mechanisms underlying HF remain understudied. Purpose We compared the performance of a baseline convolutional neural network (CNN) model designed to predict HF events from the ECG at 8 different timepoints (1, 2, 3, 4, 6, 8, 10 & 12 years after the ECG acquisition), with a fine-tuned self-supervised model, pre-trained using a large amount of unlabelled ECGs. Subsequently, we investigated the genetic architecture of the prediction from the strongest model. Methods The baseline model consisted of 6 convolutional layers, an attentive pooling layer and 3 fully connected layers. The fine-tuned self-supervised model was composed of a frozen self-supervised feature extraction block, and a trainable classification block. The self-supervised block, composed of 6 convolutional layers, aimed to extract relevant features from the ECG and was trained with 637,438 publicly available ECGs. The classification block, composed of 1 convolutional layer, an attentive pooling layer and 3 fully connected layers, used a loss function designed to optimize the prediction of HF at the 8 timepoints. The models were trained and tested using 86,567 ECGs from UK Biobank (69,257 for training and 17,310 for test). Then, we performed a genome-wide association study (GWAS) on the standardized output at 8 years of the model with the highest area under the curve (AUC), with sex, age, body mass index, genetic array and the RR interval. Results The fine-tuned self-supervised model showed a better performance across all follow-up times when compared to the baseline model, increasing the AUC 7% on average (Figure 1). The GWAS revealed 27 candidate genes at 23 genome-wide significant loci (P < 5 x 10-8), 19 previously proposed for ECG traits, 9 of these for cardiac structural or functional traits (6 for HF), and 4 new (PATJ, BOLL, NATA/PTGIS and CCDC63). We found these genes were significantly enriched in cardiac conduction and regulation of cardiac contraction processes (Figure 2). Conclusions The use of features derived from a pretrained self-supervised model improved the performance of the CNN model, allowing a better prediction of HF at long follow-up periods. Also, the genetics underlying the ECG-based prediction of HF indicated candidate genes previously reported for ECG and arrhythmia phenotypes, highlighting electrophysiological mechanisms implicated in HF. Our findings support the integration of deep learning with genetic analyses to develop more accurate and clinically valuable risk prediction tools for early intervention and prevention of HF.AUC of the two CNN models Significantly enriched mechanisms

Freezing of gait (FOG) is one of the most debilitating motor symptoms experienced by patients with Parkinson's disease (PD). FOG detection is possible using acceleration data from wearable sensors, and a convolutional neural network (CNN) is often used to determine the presence of FOG epochs. We compared the performance of a standard CNN for the detection of FOG with two more complex networks, which are well suited for time series data, the MiniRocket and the InceptionTime. We combined acceleration data of people with PD across four studies. The final data set was split into a training (80%) and hold-out test (20%) set. A fifth study was included as an unseen test set. The data were windowed (2 s) and five-fold cross-validation was applied. The CNN, MiniRocket, and InceptionTime models were evaluated using a receiver operating characteristic (ROC) curve and its area under the curve (AUC). Multiple sensor configurations were evaluated for the best model. The geometric mean was subsequently calculated to select the optimal threshold. The selected model and threshold were evaluated on the hold-out and unseen test set. A total of 70 participants (23.7 h, 9% FOG) were included in this study for training and testing, and in addition, 10 participants provided an unseen test set (2.4 h, 11% FOG). The CNN performed best (AUC = 0.86) in comparison to the InceptionTime (AUC = 0.82) and MiniRocket (AUC = 0.76) models. For the CNN, we found a similar performance for a seven-sensor configuration (lumbar, upper and lower legs and feet; AUC = 0.86), six-sensor configuration (upper and lower legs and feet; AUC = 0.87), and two-sensor configuration (lower legs; AUC = 0.86). The optimal threshold of 0.45 resulted in a sensitivity of 77% and a specificity of 58% for the hold-out set (AUC = 0.72), and a sensitivity of 85% and a specificity of 68% for the unseen test set (AUC = 0.90). We confirmed that deep learning can be used to detect FOG in a large, heterogeneous dataset. The CNN model outperformed more complex networks. This model could be employed in future personalized interventions, with the ultimate goal of using automated FOG detection to trigger real-time cues to alleviate FOG in daily life.

Aims: This study aimed to explore the effect of training set diversity on the performance of deep learning models for predicting early gastric cancer (EGC) under endoscopy. Methods: Images of EGC and non-cancerous lesions under narrow-band imaging (ME-NBI) and magnifying blue laser imaging (ME-BLI) were retrospectively collected. Training set 1 was composed of 150 non-cancerous and 309 EGC ME-NBI images, training set 2 was composed of 1505 non-cancerous and 309 EGC ME-BLI images, and training set 3 was the combination of training set 1 and 2. Test set 1 was composed of 376 non-cancerous and 1052 EGC ME-NBI images, test set 2 consisted of 529 non-cancerous and 71 EGC ME-BLI images, and test set 3 was the combination of test set 1 and test set 2. Three deep learning models, convolutional neural network (CNN) 1, CNN 2 and CNN 3 (CNN 1, CNN 2 and CNN 3 were independently trained using training set 1, training set 2 and training set 3, respectively), were constructed, and their performances on each test set were respectively evaluated. One hundred and thirty-eight ME-NBI videos and 17 ME-BLI videos were further collected to evaluate and compare the performance of each model in real time. Results: On the whole, the performance of CNN 3 was the best. The accuracy (Acc), sensitivity (Sn), specificity (Sp) and area under the curve (AUC) of test set 1 in CNN 3 were 87.89% (1255/1428), 90.96% (342/376), 86.79% (913/1052) and 94.60%, respectively. The Acc, Sn, Sp and AUC of test set 2 in CNN 3 were 95% (570/600), 97.92% (518/529), 73.24% (52/71) and 90.93% respectively. The Acc, Sn, Sp and AUC of test set 3 in CNN 3 were 89.99% (1825/2028), 95.03% (860/905), 85.93% (965/1123) and 94.89%, respectively. The performance of CNN 3 was also the best in videos test set. The Acc, Sn and Sp of videos test set in CNN 3 were 91.03% (142/156), 90.58% (125/138) and 94.44% (17/18), respectively. Conclusions: The deep learning model with the most diverse training data has the best diagnostic effect.

Background Deep learning may help to improve computer-aided detection of volume (CADv) measurement of pulmonary nodules at chest CT. Purpose To determine the efficacy of a deep learning method for improving CADv for measuring the solid and ground-glass opacity (GGO) volumes of a nodule, doubling time (DT), and the change in volume at chest CT. Materials and Methods From January 2014 to December 2016, patients with pulmonary nodules at CT were retrospectively reviewed. CADv without and with a convolutional neural network (CNN) automatically determined total nodule volume change per day and DT. Area under the curves (AUCs) on a per-nodule basis and diagnostic accuracy on a per-patient basis were compared among all indexes from CADv with and without CNN for differentiating benign from malignant nodules. Results The CNN training set was 294 nodules in 217 patients, the validation set was 41 nodules in 32 validation patients, and the test set was 290 nodules in 188 patients. A total of 170 patients had 290 nodules (mean size ± standard deviation, 11 mm ± 5; range, 4-29 mm) diagnosed as 132 malignant nodules and 158 benign nodules. There were 132 solid nodules (46%), 106 part-solid nodules (36%), and 52 ground-glass nodules (18%). The test set results showed that the diagnostic performance of the CNN with CADv for total nodule volume change per day was larger than DT of CADv with CNN (AUC, 0.94 [95% confidence interval {CI}: 0.90, 0.96] vs 0.67 [95% CI: 0.60, 0.74]; P < .001) and CADv without CNN (total nodule volume change per day: AUC, 0.69 [95% CI: 0.62, 0.75]; P < .001; DT: AUC, 0.58 [95% CI: 0.51, 0.65]; P < .001). The accuracy of total nodule volume change per day of CADv with CNN was significantly higher than that of CADv without CNN (P < .001) and DT of both methods (P < .001). Conclusion Convolutional neural network is useful for improving accuracy of computer-aided detection of volume measurement and nodule differentiation capability at CT for patients with pulmonary nodules. © RSNA, 2020 Online supplemental material is available for this article.

Integrating machine learning (ML) models into healthcare systems is a rapidly evolving field with the potential to revolutionize care delivery. This study aimed to classify fertility rates and identify significant predictors using ML models among reproductive women in Ethiopia. This study utilized eight ML models in 5864 reproductive-age women using Ethiopian Demographic Health Survey (EDHS), 2019 data. Phyton programming language was used to develop these models. Predictors of fertility rate were determined using the feature important techniques. The performance of models was evaluated using accuracy, area under the curve (AUC), precision, recall, F1-score, specificity, and sensitivity. The mean age of participants was 32.7 (± 5.6) years. The random forest classifier (accuracy = 0.901 and AUC = 0.961) followed by a one-dimensional convolutional neural network (accuracy = 0.899 and AUC = 0.958), logistic regression (accuracy = 0.874 and AUC = 0.937), and gradient boost classifier (accuracy = 0.851 and AUC 0.927) were the top performing ML models. Family size, age, occupation, and education with an average importance score of 0.198, 0.151, 0.118, and 0.081, respectively were the top significant predictors of the fertility rate. The best ML models to classify and predict fertility rates were random forest, one-dimensional convolutional neural network, logistic regression, and gradient boost classifier. The findings on important factors of fertility rate can inform targeted public health, programs that address disparities related to family size, occupation, education, and other socioeconomic factors.

Prostate Cancer Risk Prediction and Online Calculation Based on Machine Learning Algorithm

Using Convolutional Neural Networks for Enhanced Capture of Breast Parenchymal Complexity Patterns Associated with Breast Cancer Risk

Postherpetic neuralgia (PHN), a complication of herpes zoster, significantly impacts the quality of life of affected patients. Research indicates that early intervention for pain can reduce the occurrence or severity of PHN. This study aims to develop a predictive model and scoring table to identify patients at risk of developing PHN following acute herpetic neuralgia, facilitating informed clinical decision-making. We conducted a retrospective review of 524 hospitalized patients with herpes zoster at The First Affiliated Hospital of Zhejiang Chinese Medical University from December 2020 to December 2023 and classified them according to whether they had PHN, collecting a comprehensive set of 30 patient characteristics and disease-related indicators, 5 comorbidity indicators, 2 disease score values, and 10 serological indicators. Relevant features associated with PHN were identified using the least absolute shrinkage and selection operator (LASSO). Then, the patients were divided into a training set and a test set in a 4:1 ratio, with comparability tested using univariate analysis. Six models were established in the training set using machine learning methods: support vector machines, logistic regression, random forest, k-nearest neighbor, gradient boosting, and neural network. The performance of these models was evaluated in the test set, and a nomogram based on logistic regression was used to create a PHN prediction score table. Eight non-zero characteristic variables selected from the LASSO regression results were included in the model, including age [area under the curve (AUC)=0.812, p<0.001], Numerical Rating Scale (NRS) (AUC=0.792, p<0.001), receiving treatment time (AUC=0.612, p<0.001), rash recovery time (AUC=0.680, p<0.001), history of malignant tumor (AUC=0.539, p<0.001), history of diabetes (AUC=0.638, p<0.001), varicella-zoster virus immunoglobulin M (AUC=0.620, p<0.001), and serum nerve-specific enolase (AUC=0.659, p<0,001). The gradient boosting model outperformed other classifier models on the test set with an AUC of 0.931, 95% confidence interval (CI) (0.882-0.980), accuracy of 0.886 (95% CI 0.809-0.940). In the test set, our predictive scoring table achieved an AUC of 0.820 (95% CI 0.869-0.970) with accuracy of 0.790 (95% CI 0.700-0.864). This study presents a methodology for predicting the development of postherpetic neuralgia in shingles patients by analyzing historical case data, employing various machine learning techniques, and selecting the optimal model through comparative analysis. In addition, a logistic regression model has been used to create a scoring table for predicting the postherpetic neuralgia.

This study aims to develop and validate a machine learning-based mortality risk prediction model for V-A ECMO patients to improve the precision of clinical decision-making. This multicenter retrospective cohort study included 280 patients receiving V-A ECMO from the Second Affiliated Hospital of Guangxi Medical University, Yulin First People’s Hospital, and the MIMIC-IV database. The data from the Second Affiliated Hospital of Guangxi Medical University and the MIMIC-IV database were merged and randomly divided in a 7:3 ratio into a training set and an internal validation set, respectively. The dataset from Yulin First People’s Hospital was reserved as an external validation cohort. The primary study outcome was defined as in-hospital mortality.Feature selection was conducted using Lasso regression, followed by the development of six machine learning models: Logistic Regression, Random Forest (RF), Deep Neural Network (DNN), Support Vector Machine (SVM), LightGBM, and CatBoost. Model performance was assessed using the Area Under the Curve (AUC), accuracy, sensitivity, specificity, and F1 score. Model validation was performed through calibration and decision curve analysis. Feature importance was evaluated using SHAP, and subgroup analysis was conducted to assess the model’s applicability across different clinical scenarios. In internal validation, the Logistic Regression model performed the best, with an AUC of 0.86 (95% CI: 0.77–0.93), accuracy of 0.76, sensitivity of 0.73, specificity of 0.79, and an F1 score of 0.73. It outperformed other models (RF: AUC = 0.79, DNN: AUC = 0.78, SVM: AUC = 0.76, LightGBM: AUC = 0.71, CatBoost: AUC = 0.77). External validation yielded consistent results, with the Logistic Regression model’s AUC at 0.75 (95% CI: 0.56–0.92), accuracy of 0.69, sensitivity of 0.64, specificity of 0.73, and an F1 score of 0.66. Calibration curve analysis revealed that the Logistic Regression model had the lowest Brier score (0.1496), indicating the most reliable predicted probabilities. Decision curve analysis demonstrated that the model provided the highest net benefit across most decision thresholds. SHAP analysis identified lactate, age, and albumin as key predictors of mortality, with lactate and age positively correlated, and albumin negatively correlated. Subgroup analysis revealed better performance in the cardiac arrest group (AUC = 0.81), non-sepsis group (AUC = 0.75), and non-diabetes group (AUC = 0.78). The Logistic Regression-based mortality risk prediction model for V-A ECMO patients demonstrated comparable or even favorable performance to more complex machine learning models, with the advantage of higher interpretability.By explicitly incorporating lactate, age, and albumin as the principal predictors, this model facilitates precise risk stratification and provides practical support for clinical decision-making in ECMO management.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-25423-4.

Background Alterations in portal vein flow (PVF) are emerging as potential markers of venous congestion in congestive heart failure (HF); however, their dynamic changes during decongestion, and their utility in assessing HF status remain unclear. The aim of this study was to evaluate PVF parameter changes in HF patients during congestion and post-decongestive therapy, and their association with HF compensation and renal function. Patients and methods This cross-sectional study enrolled 35 HF patients (12 compensated, 23 noncompensated) at Assiut University Hospital’s ICU. Baseline and postdecongestion assessments included clinical evaluation, laboratory tests (e.g., eGFR, creatinine), point-of-care ultrasound (inferior vena cava size, PVF via Doppler), and echocardiography. PVF parameters (V.max, V.min, PI) were compared between groups and GFR categories, with diagnostic accuracy assessed through receiver-operating characteristic curves and predictors identified by logistic regression. Statistical significance was set at P less than 0.05. Results Mean age was 57 ± 12 years; 65.7% had noncompensated HF. At congestion, noncompensated HF showed lower V.max (15.13 ± 2.59 vs. 20.33 ± 3.34 cm/s), V.min (8.84 ± 1.64 vs. 9.47 ± 1.76 cm/s), and PI (0.41 ± 0.06 vs. 0.53 ± 0.06) than compensated HF ( P &lt;0.001). PI decreased postdecongestion (0.5 ± 0.1 to 0.4 ± 0.1). V.max [area under the curve (AUC)=0.630, specificity=100%] and V.min (AUC=0.605, specificity=91.67%) had moderate diagnostic accuracy for noncompensated HF, unlike PI (AUC=0.527). PVF varied significantly with GFR ( P &lt;0.001), with higher V.max and V.min in mild to dodtoatmoderate vs. severe GFR decline. Inferior vena cava size (OR=1.688, P =0.023) and right heart dilatation (OR=16.219, P =0.043) predicted noncompensated HF. Renal impairment was evident (eGFR 21.7 ± 8.4 ml/min/1.73 m²). Conclusion PVF parameters indicate venous congestion and renal function in HF, which change with decongestion, but are moderately diagnostically useful. POCUS-derived PVF assessment can potentially aid in the monitoring of HF, especially in the context of cardiorenal syndrome, but needs validation.

Background: Hepatocyte ballooning (HB) is a significant histological characteristic linked to the advancement of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH). Although clinicians now consider liver biopsy the most reliable method for identifying HB, its invasive nature and related dangers highlight the need for the development of non-invasive diagnostic options. Objective: This study aims to develop a novel methodology that combines deep learning and machine learning techniques to accurately identify and measure hepatobiliary abnormalities in liver ultrasound images. Methods: The research team expanded the dataset, consisting of ultrasound images, and used it for training deep convolutional neural networks (CNNs) such as InceptionV3, ResNet50, DenseNet121, and EfficientNetB0. A hybrid approach, combining InceptionV3 for feature extraction with a Random Forest classifier, emerged as the most accurate and stable method. An approach of dual dichotomy classification was used to categorize images into two stages: healthy vs. sick, and then mild versus severe ballooning.. Features obtained from CNNs were integrated with conventional machine learning classifiers like Random Forest and Support Vector Machines (SVM). Results: The hybrid approach achieved an accuracy of 97.40%, an area under the curve (AUC) of 0.99, and a sensitivity of 99% for the ‘Many’ class during the third phase of evaluation. The dual dichotomy classification enhanced the sensitivity in identifying severe instances of HB. The cross-validation process confirmed the strength and reliability of the suggested models. Conclusions: These results indicate that this combination method can decrease the need for invasive liver biopsies by providing a non-invasive and precise alternative for early identification and monitoring of NAFLD and NASH. Subsequent research will prioritize the validation of these models using larger datasets from multiple centers to evaluate their generalizability and incorporation into clinical practice.

PurposeThe aim of this study was to evaluate the value of different multiparametric MRI-based radiomics models in differentiating stage IA endometrial cancer (EC) from benign endometrial lesions.MethodsThe data of patients with endometrial lesions from two centers were collected. The radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) map, and late contrast-enhanced T1-weighted imaging (LCE-T1WI). After data dimension reduction and feature selection, nine machine learning algorithms were conducted to determine which was the optimal radiomics model for differential diagnosis. The univariate analyses and logistic regression (LR) were performed to reduce valueless clinical parameters and to develop the clinical model. A nomogram using the radscores combined with clinical parameters was developed. Two integrated models were obtained respectively by the ensemble strategy and stacking algorithm based on the clinical model and optimal radiomics model. The area under the curve (AUC), clinical decisive curve (CDC), net reclassification index (NRI), and integrated discrimination index (IDI) were used to evaluate the performance and clinical benefits of the models.ResultsA total of 371 patients were incorporated. The LR model was the optimal radiomics model with the highest average AUC (0.854) and accuracy (0.802) in the internal and external validation groups (AUC = 0.910 and 0.798, respectively), and outperformed the clinical model (AUC = 0.739 and 0.592, respectively) or the radiologist (AUC = 0.768 and 0.628, respectively). The nomogram (AUC = 0.917 and 0.802, respectively) achieved better discrimination performance than the optimal radiomics model in two validation groups. The stacking model (AUC = 0.915) and ensemble model (AUC = 0.918) had a similar performance compared with the nomogram in the internal validation group, whereas the AUCs of the stacking model (AUC = 0.792) and ensemble model (AUC = 0.794) were lower than those of the nomogram and radiomics model in the external validation group. According to the CDC, NRI, and IDI, the optimal radiomics model, nomogram, stacking model, and ensemble model achieved good net benefits.ConclusionsMultiparametric MRI-based radiomics models can non-invasively differentiate stage IA EC from benign endometrial lesions, and LR is the best machine learning algorithm. The nomogram presents excellent and stable diagnostic efficiency.

Assess the efficacy of deep convolutional neural networks (DCNNs) in detection of critical enteric feeding tube malpositions on radiographs. 5475 de-identified HIPAA compliant frontal view chest and abdominal radiographs were obtained, consisting of 174 x-rays of bronchial insertions and 5301 non-critical radiographs, including normal course, normal chest, and normal abdominal x-rays. The ground-truth classification for enteric feeding tube placement was performed by two board-certified radiologists. Untrained and pretrained deep convolutional neural network models for Inception V3, ResNet50, and DenseNet 121 were each employed. The radiographs were fed into each deep convolutional neural network, which included untrained and pretrained models. The Tensorflow framework was used for Inception V3, ResNet50, and DenseNet. Images were split into training (4745), validation (630), and test (100). Both real-time and preprocessing image augmentation strategies were performed. Receiver operating characteristic (ROC) and area under the curve (AUC) on the test data were used to assess the models. Statistical differences among the AUCs were obtained. p < 0.05 was considered statistically significant. The pretrained Inception V3, which had an AUC of 0.87 (95 CI; 0.80–0.94), performed statistically significantly better (p < .001) than the untrained Inception V3, with an AUC of 0.60 (95 CI; 0.52–0.68). The pretrained Inception V3 also had the highest AUC overall, as compared with ResNet50 and DenseNet121, with AUC values ranging from 0.82 to 0.85. Each pretrained network outperformed its untrained counterpart. (p < 0.05). Deep learning demonstrates promise in differentiating critical vs. non-critical placement with an AUC of 0.87. Pretrained networks outperformed untrained ones in all cases. DCNNs may allow for more rapid identification and communication of critical feeding tube malpositions.

Diagnostic Accuracy of D-dimer is Comparable to NT- proBNP in Decompensated Heart Failure