Search Results (171)

This paper investigates the feasibility of diagnosing abdominal aortic aneurysm (AAA) via deep learning (DL)-enabled analysis of non-invasive arterial pulse waveform signals. We generated arterial blood pressure (BP) and pulse volume recording (PVR) waveform signals across a diverse synthetic patient cohort using a systemic arterial circulation model coupled with a viscoelastic model relating arterial BP to PVR while simulating a range of AAA severity levels. We confirmed the plausibility of the synthetic data by comparing the alterations in the simulated waveform signals due to AAA against previously reported in vivo findings. Then, we developed a convolutional neural network (CNN) with continuous property-adversarial regularization that can estimate AAA severity from brachial and tibial PVR signals. We evaluated the algorithm’s performance in comparison with an identical CNN trained on invasive arterial BP waveform signals. The DL-enabled PVR-based algorithm achieved robust AAA detection across different severity thresholds with area under the ROC curve values >0.89, and showed reasonable accuracy in severity estimation, though slightly lower than its invasive BP counterpart (MAE: 12.6% vs. 10.3%). These findings suggest that DL-enabled analysis of PVR waveform signals offers a non-invasive and cost-effective approach for AAA diagnosis, potentially enabling accessible screening through operator-agnostic and point-of-care technologies. Full article

Background/Objectives: Preterm birth (PTB), defined as birth before 37 completed weeks of gestation, is a major global health challenge and a leading cause of neonatal mortality. PTB is broadly classified into spontaneous and medically indicated (iatrogenic), which have distinct etiologies. While prediction is key to improving outcomes, there is a lack of models that specifically differentiate between spontaneous and iatrogenic PTB subtypes. This study aimed to develop and validate predictive models for the prediction of spontaneous and iatrogenic PTB at <32, <34, and <37 weeks’ gestation using medical history and readily available second-trimester data. Methods: This was a retrospective cohort study on singleton pregnancies from a single tertiary institution (2012–2025). Predictor variables included maternal characteristics, obstetric history, and second-trimester ultrasound markers. Four algorithms, including multivariable Logistic Regression and three machine learning methods (Random Forest, XGBoost, and a Neural Network), were trained and evaluated on a held-out test set (20% of the data). Model performance was primarily assessed by the Area Under the Curve (AUC). Results: In total, 9805 singleton pregnancies were included. The models performed significantly better for iatrogenic PTB than for spontaneous PTB. For delivery <37 weeks, the highest AUC for iatrogenic PTB was 0.764 (Random Forest), while for spontaneous PTB it was 0.609 (Neural Network). Predictive accuracy improved for earlier gestations; for delivery <32 weeks, the best model for iatrogenic PTB achieved an AUC of 0.862 (Neural Network), and the best model for spontaneous PTB achieved an AUC of 0.749 (Random Forest). Model interpretation revealed that iatrogenic PTB was primarily driven by markers of placental dysfunction, such as estimated fetal weight by ultrasound scan and uterine artery pulsatility index, while spontaneous PTB was most associated with a history of PTB and a short cervical length. Conclusions: Models using routine mid-gestation data demonstrate effective prediction for iatrogenic PTB, with accuracy improving for earlier, more severe cases. In contrast, performance for spontaneous PTB was modest. Traditional Logistic Regression performed comparably to complex machine learning algorithms, highlighting that the clinical value is rooted in the subtype-specific modeling approach rather than in algorithmic complexity. Full article

Background: The Fontan procedure is a palliative surgery for patients with single-ventricle congenital heart disease (CHD), but it is associated with postoperative and long-term mortality and morbidity. Accurate, individualized risk stratification remains a challenge with traditional models. This study aimed to develop and validate a deep learning (DL) model to predict postoperative mortality after the Fontan procedure and to identify key predictive factors. Methods: We retrospectively analysed data from 230 patients who underwent the Fontan procedure between 2010 and 2024. A Deep Neural Network (DNN) model was developed using comprehensive preoperative, intraoperative, and postoperative clinical, biochemical, and hemodynamic variables. The dataset was split using five-fold cross-validation, with 80% for training and 20% for testing in each fold. The Synthetic Minority Over-sampling Technique (SMOTE) was used to fix class imbalance. Model performance was evaluated using five-fold stratified cross-validation. We assessed accuracy, precision, recall, F1-score, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC). SHapley Additive exPlanations (SHAP) analysis was employed to enhance model interpretability and identify the importance of features. A user-friendly clinical application interface was developed using Streamlit. This study was reported in accordance with the TRIPOD + AI reporting guidelines. Results: The DNN model demonstrated superior performance in predicting postoperative mortality, achieving an overall accuracy of 91.5% (95% CI: 87.2–94.8%), precision of 83.3% (95% CI: 76.5–89.1%), recall (sensitivity) of 90.9% (95% CI: 85.2–95.1%), specificity of 92.5% (95% CI: 88.3–95.7%), F1-score of 87.0% (95% CI: 82.1–91.3%), and an AUC-ROC of 0.94 (95% CI: 0.88–0.99). SHAP analysis identified key predictors of mortality, such as pulmonary artery pressure, ventricular end-diastolic pressure, preoperative BNP levels, and severity of AV valve regurgitation. The Streamlit application offered a user-friendly interface for personalized risk evaluation. Conclusions: A deep learning model that incorporates detailed clinical data can precisely forecast postoperative mortality in patients undergoing Fontan surgeries. This AI-based method, combined with interpretability techniques, provides a valuable tool for personalized risk assessment. It has the potential to improve preoperative counseling, optimize perioperative care, and enhance patient outcomes. However, additional external validation is needed to verify its broader applicability and clinical usefulness. Full article

Finite element analysis (FEA)-based computational fluid dynamics (CFD) simulations are essential in biomedical engineering for studying haemodynamics, yet their high computational cost limits large-scale parametric studies. This paper presents a comparative analysis of FEA and surrogate modelling techniques applied to renal artery haemodynamics. The aortic–renal bifurcation strongly influences renal perfusion, affecting conditions such as hypertension, infarction, and transplant rejection. This study evaluates GPU-accelerated voxel simulations (Ansys 2024 R2 Discovery), 2D and 3D FEA simulations (COMSOL Multiphysics 6.3), finite volume CFD (Ansys 2020 R2 Fluent), and deep neural networks (DNNs) as surrogate models. Branching angles and blood pressure were systematically varied, and their effects on velocity, pressure, and turbulent kinetic energy were assessed in a time-dependent framework. Fluent provided accurate baseline results, while COMSOL 2D gave sufficient accuracy with much lower runtimes. In contrast, COMSOL 3D required over 160 times longer, making it prohibitive. Surrogate models trained on 6500 or more FEA-derived samples achieved high predictive accuracy (R² > 0.98 for velocity and pressure), balancing training cost and result quality. Cost analysis showed surrogate models become advantageous after 76–93 simulations. Symmetry is expressed in balancing model fidelity and computational efficiency, providing a resource-effective methodology with broad potential in vascular applications. Full article

Background/Objectives: Chronic wounds of the lower extremities, particularly arterial and venous ulcers, represent a significant and costly challenge in medical care. To assist in differential diagnosis, we aim to evaluate various advanced deep-learning models for classifying arterial and venous ulcers and visualize their decision-making processes. Methods: A retrospective dataset of 607 images (198 arterial and 409 venous ulcers) was used to train five convolutional neural networks: ResNet50, ResNeXt50, ConvNeXt, EfficientNetB2, and EfficientNetV2. Model performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC. Grad-CAM was applied to visualize image regions contributing to classification decisions. Results: The models demonstrated high classification performance, with accuracy ranging from 72% (ConvNeXt) to 98% (ResNeXt50). Precision and recall values indicated strong discrimination between arterial and venous ulcers, with EfficientNetV2 achieving the highest precision. Conclusions: AI-assisted classification of venous and arterial ulcers offers a valuable method for enhancing diagnostic efficiency. Full article

Human perception of urban streetscapes plays a crucial role in shaping human-centered urban planning and policymaking. Traditional studies on safety perception often rely on labor-intensive field surveys with limited spatial coverage, hindering large-scale assessments. To address this gap, this study constructs a street safety perception dataset for Wuhan, classifying street scenes into three perception levels. A convolutional neural network model based on transfer learning is developed, achieving a classification accuracy of 78.3%. By integrating image-based prediction with spatial clustering and correlation analysis, this study demonstrates that safety perception displays a distinctly clustered and uneven spatial distribution, primarily concentrated along major arterial roads and rail transit corridors by high safety levels. Correlation analysis indicates that higher safety perception is moderately associated with greater road grade, increased road width, and lower functional level while showing a weak negative correlation with housing prices. By presenting a framework that integrates transfer learning and geospatial analysis to connect urban street imagery with human perception, this study advances the assessment of spatialized safety perception and offers practical insights for urban planners and policymakers striving to create safer, more inclusive, and sustainable urban environments. Full article

Background: Pulmonary arterial pressure is a key parameter for diagnosing cardiovascular and pulmonary diseases. Its measurement through right heart catheterization is considered the gold standard, and it is an invasive procedure that entails significant risks for patients. This has motivated the development of non-invasive techniques based on patient-specific imaging, such as Physics-Informed Neural Networks (PINNs), which integrate clinical measurements with physical models, such as the 1D reduced Navier–Stokes model, enabling biologically plausible predictions with limited data. Methods: This work implements a PINN model that uses velocity and area measurements in the main bifurcation of the pulmonary artery, comprising the main artery and its secondary branches, to predict pressure, velocity, and area variations throughout the bifurcation. The model training includes penalties to satisfy the laws of flow and momentum conservation. Results: The results show that, using 4D Flow MRI images from a healthy patient as clinical data, the pressure estimates provided by the model are consistent with the expected ranges reported in the literature, reaching a mean arterial pressure of 21.5 mmHg. Conclusions: This model presents an innovative approach that avoids invasive methods, being the first study to apply PINNs to estimate pulmonary arterial pressure in bifurcations. In future work, we aim to validate the model in larger populations and confirm pulmonary hypertension cases diagnosed through catheterization. Full article

Objectives: This study aimed to develop and validate a machine learning (ML) algorithm to predict 30-day mortality following isolated coronary artery bypass grafting (CABG) and to compare its performance against the widely used European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) risk prediction model. Methods: In this retrospective study, we included consecutive adult patients who underwent isolated CABG between January 2009 and December 2022. Three predictive models were compared: (1) EuroSCORE II variables alone (baseline), (2) EuroSCORE II combined with additional preoperative variables (Model I), and (3) EuroSCORE II plus preoperative and postoperative variables available within five days after surgery (Model II). Logistic Regression (LR), Random Forest (RF), and Neural Network (NN) were employed and validated. Predictive accuracy was assessed using the area under the receiver operating characteristic curve (AUC) and specificity at 85% sensitivity. Results: Among the 3483 patients included, the mean age was 66.2 years (SD 10.3), with an overall 30-day mortality rate of 2.5%. The mean EuroSCORE II was 3.12 (SD 4.8). Integrating additional preoperative variables significantly improved specificity at 85% sensitivity for both random forest (from 42% to 51%; p < 0.001) and NN (from 28% to 43%; p < 0.001) but not for LR. Incorporating preoperative along with postoperative data (Model II) further improved specificity to approximately 70% across all ML methods (p < 0.001). The most influential postoperative predictors included kidney failure, pulmonary complications, and myocardial infarction. Conclusions: ML models incorporating preoperative and postoperative variables significantly outperform the traditional EuroSCORE II in predicting short-term mortality following isolated CABG. Full article

Background/Objectives: Invasive coronary angiography (ICA) is the gold standard for the diagnosis of coronary artery disease (CAD), with various non-invasive imaging modalities also available. Machine learning (ML) methods are increasingly applied to overcome the limitations of diagnostic imaging by improving accuracy and observer independent performance. Methods: This meta-analysis (PRISMA method) summarizes the evidence for ML-based analyses of coronary imaging data from ICA, coronary computed tomography angiography (CT), and nuclear stress perfusion imaging (SPECT) to predict clinical outcomes and performance for precise diagnosis. We searched for studies from Jan 2012–March 2023. Study-reported c index values and 95% confidence intervals were used. Subgroup analyses separated models by outcome. Combined effect sizes using a random-effects model, test for heterogeneity, and Egger’s test to assess publication bias were considered. Results: In total, 46 studies were included (total subjects = 192,561; events = 31,353), of which 27 had sufficient data. Imaging modalities used were CT (n = 34), ICA (n = 7) and SPECT (n = 5). The most frequent study outcome was detection of stenosis (n = 11). Classic deep neural networks (n = 12) and convolutional neural networks (n = 7) were the most used ML models. Studies aiming to diagnose CAD performed best (0.85; 95% CI: 82, 89); models aiming to predict clinical outcomes performed slightly lower (0.81; 95% CI: 78, 84). The combined c-index was 0.84 (95% CI: 0.81–0.86). Test of heterogeneity showed a high variation among studies (I² = 97.2%). Egger’s test did not indicate publication bias (p = 0.485). Conclusions: The application of ML methods to diagnose CAD and predict clinical outcomes appears promising, although there is lack of standardization across studies. Full article

This study explores the integration of advanced artificial intelligence (AI) techniques with infrared thermography for diagnosing diabetic peripheral neuropathy (DPN) and peripheral arterial disease (PAD). Diabetes-related foot complications, including DPN and PAD, are leading causes of morbidity and disability worldwide. Traditional diagnostic methods, such as the monofilament test for DPN and ankle–brachial pressure index for PAD, have limitations in sensitivity, highlighting the need for improved solutions. Thermographic imaging, a non-invasive, cost-effective, and reliable tool, captures temperature distributions of the patient plantar surface, enabling the detection of physiological changes linked to these conditions. This study collected thermographic data from diabetic patients and employed convolutional neural networks (CNNs) and vision transformers (ViTs) to classify individuals as healthy or affected by DPN or PAD (not healthy). These neural networks demonstrated superior diagnostic performance, compared to traditional methods (an accuracy of 95.00%, a sensitivity of 100.00%, and a specificity of 90% in the case of the ResNet-50 network). The results underscored the potential of combining thermography with AI to provide scalable, accurate, and patient-friendly diagnostics for diabetic foot care. Future work should focus on expanding datasets and integrating explainability techniques to enhance clinical trust and adoption. Full article

Ischemic stroke (IS) is a high-mortality, multi-complication cardiovascular disease. Reducing brain injury and promoting neuronal repair after IS onset remain important challenges for current treatments. Our team previously found that PAS840, an extract from Periplaneta americana (L.), protects nerve function; this study further uses LC-MS/MS and peptidomics to analyze PAS840’s components and network pharmacology to predict its ischemic stroke (IS) therapeutic targets. We then employed Transwell, a biochemical kit, real-time quantitative polymerase chain reaction (RT-qPCR), and transcriptomics to investigate PAS840’s effects on migration ability, oxidative stress levels, and cellular pathways in mouse microglial cells (BV-2) following oxygen–glucose deprivation/reoxygenation (OGD/R) injury. Finally, using Evans blue staining, immunohistochemical analysis, and RT-qPCR, we investigated PAS840’s effects on the blood–brain barrier, inflammation pathways, and neural function in a transient middle cerebral artery occlusion (tMCAO) rat model. PAS840 components target multiple IS pathways, effectively inhibit NF-κB/NLRP3/Caspase-1/IL-1β inflammasome pathway activation in BV-2 cells following OGD/R, reduce cellular oxidative stress, inflammation, and pyroptosis, and improve cell viability and migration ability. PAS840 decreases NF-κB/NLRP3/Caspase-1/IL-1β inflammasome pathway expression in tMCAO rat brains, reduces inflammation, activates BDNF/VGF/NGR1/Erbb4 neurotrophic factor and vascular endothelial growth factor pathways, enhances neuronal cell viability, and effectively protects and repairs the blood–brain barrier. Full article

Background: Ventilator-associated pneumonia (VAP) is a common and serious ICU complication, affecting up to 40% of mechanically ventilated patients. The diagnosis of VAP currently relies on retrospective clinical, radiological, and microbiological criteria, which often delays targeted treatment and promotes the overuse of broad-spectrum antibiotics. The early prediction of VAP is crucial to improve outcomes and guide antimicrobial use related to this disease. This study aimed to develop and validate PREDICT (Pneumonia Risk Evaluation and Diagnostic Intelligence via Computational Technology), a deep learning algorithm for early VAP prediction that is based solely on vital signs. Methods: We conducted a retrospective cohort study using the MIMIC-IV database, which includes ICU patients who were ventilated for at least 48 h. Five vital signs (respiratory rate, SpO₂, heart rate, temperature, and mean arterial pressure) were structured into 24 h temporal windows. The PREDICT model, based on a long short-term memory neural network, was trained to predict the onset of VAP 6, 12, and 24 h in the future. Its performance was compared to that of conventional machine learning models (random forest, XGBoost, logistic regression) using their AUPRC, sensitivity, specificity, and predictive values. Results: PREDICT achieved high predictive accuracy with AUPRC values of 96.0%, 94.1%, and 94.7% at 6, 12, and 24 h before the onset of VAP, respectively. Its sensitivity and positive predictive values exceeded 85% across all horizons. Traditional ML models showed a drop in performance over longer timeframes. Analysis of the model’s explainability highlighted the respiratory rate, SpO₂, and temperature as key predictive features. Conclusions: PREDICT is the first deep learning model specifically designed for early VAP prediction in ICUs. It represents a promising tool for timely clinical decision-making and improved antibiotic stewardship. Full article

Arterial stiffness (AS) is a well-established predictor of cardiovascular events, including myocardial infarction and stroke. One of the most recognized methods for assessing AS is through arterial pulse wave velocity (aPWV), which provides valuable clinical insights into vascular health. However, its measurement typically requires specialized equipment, making it inaccessible in primary healthcare centers and low-resource settings. In this study, we developed and validated different machine learning models to estimate aPWV using common clinical markers routinely collected in standard medical examinations. Thus, we trained five regression models: Linear Regression, Polynomial Regression (PR), Gradient Boosting Regression, Support Vector Regression, and Neural Networks (NNs) on the EVasCu dataset, a cohort of apparently healthy individuals. A 10-fold cross-validation demonstrated that PR and NN achieved the highest predictive performance, effectively capturing nonlinear relationships in the data. External validation on two independent datasets, VascuNET (a healthy population) and ExIC-FEp (a cohort of cardiopathic patients), confirmed the robustness of PR and NN ($R^2>0.90$) across different vascular conditions. These results indicate that by using easily accessible clinical variables and AI-driven insights, it is possible to develop a cost-effective tool for aPWV estimation, enabling early cardiovascular risk stratification in underserved and rural areas where specialized AS measurement devices are unavailable. Full article

Compartment boundaries divide the embryo into segments with distinct fates and functions. In the vascular system, compartment boundaries organize endothelial cells into arteries, capillaries, and veins that are the fundamental units of a circulatory network. For vascular smooth muscle cells (SMCs), such boundaries produce mosaic patterns of investment based on embryonic origins with important implications for the non-uniform distribution of vascular disease later in life. The morphogenesis of blood vessels requires vascular cell movements within compartments as highly-sensitive responses to changes in fluid flow shear stress and wall strain. These movements underline the remodeling of primitive plexuses, expansion of lumen diameters, regression of unused vessels, and building of multilayered artery walls. Although the loss of endothelial compartment boundaries can produce arterial–venous malformations, little is known about the consequences of mislocalization or the failure to form SMC-origin-specific boundaries during vascular development. We propose that the failure to establish a normal compartment boundary between cardiac neural-crest-derived SMCs of the 6th pharyngeal arch artery (future ductus arteriosus) and paraxial-mesoderm-derived SMCs of the dorsal aorta in mid-gestation embryos leads to aortic coarctation observed at birth. This model raises new questions about the effects of fluid flow dynamics on SMC investment and the formation of SMC compartment borders during pharyngeal arch artery remodeling and vascular development. Full article

The development of non-invasive blood pressure monitoring systems remains a critical challenge, particularly in resource-constrained settings. This study proposes an efficient deep learning framework integrating Edge Artificial Intelligence for continuous blood pressure estimation using photoplethysmography (PPG) signals. We evaluate three architectures: a residual-enhanced convolutional neural network, a transformer-based model, and an attentive BPNet. Using the MIMIC-IV waveform database, we implement a signal processing pipeline with adaptive filtering, statistical normalization, and peak-to-peak alignment. Experiments assess varying temporal windows (10 s, 20 s, 30 s) to optimize predictive accuracy and computational efficiency. Attentive BPNet achieves the best performance, with systolic blood pressure (SBP) estimation yielding a mean absolute error (MAE) of 6.36 mmHg, diastolic blood pressure (DBP) an MAE of 4.09 mmHg, and mean arterial pressure (MBP) an MAE of 4.56 mmHg. Post-training quantization reduces the model size by 90.71% (to 0.13 MB), enabling deployment on Edge devices. These findings demonstrate the feasibility of deploying deep learning-based continuous blood pressure monitoring on edge devices. The proposed framework provides a scalable and computationally efficient solution, offering real-time, accessible monitoring that could enhance hypertension management and optimize healthcare resource utilization. Full article