Plant Methods. 2025 Apr 2;21(1):48. doi: 10.1186/s13007-025-01325-4.

ABSTRACT

Plant leaf diseases significantly threaten agricultural productivity and global food security, emphasizing the importance of early and accurate detection and effective crop health management. Current deep learning models, often used for plant disease classification, have limitations in capturing intricate features such as texture, shape, and color of plant leaves. Furthermore, many of these models are computationally expensive and less suitable for deployment in resource-constrained environments such as farms and rural areas. We propose a novel Lightweight Deep Learning model, Depthwise Separable Convolution with Spatial Attention (LWDSC-SA), designed to address limitations and enhance feature extraction while maintaining computational efficiency. By integrating spatial attention and depthwise separable convolution, the LWDSC-SA model improves the ability to detect and classify plant diseases. In our comprehensive evaluation using the PlantVillage dataset, which consists of 38 classes and 55,000 images from 14 plant species, the LWDSC-SA model achieved 98.7% accuracy. It presents a substantial improvement over MobileNet by 5.25%, MobileNetV2 by 4.50%, AlexNet by 7.40%, and VGGNet16 by 5.95%. Furthermore, to validate its robustness and generalizability, we employed K-fold cross-validation K=5, which demonstrated consistently high performance, with an average accuracy of 98.58%, precision of 98.30%, recall of 98.90%, and F1 score of 98.58%. These results highlight the superior performance of the proposed model, demonstrating its ability to outperform state-of-the-art models in terms of accuracy while remaining lightweight and efficient. This research offers a promising solution for real-world agricultural applications, enabling effective plant disease detection in resource-limited settings and contributing to more sustainable agricultural practices.

PMID:40176127 | DOI:10.1186/s13007-025-01325-4

Biomed Eng Online. 2025 Apr 2;24(1):39. doi: 10.1186/s12938-025-01360-1.

ABSTRACT

BACKGROUND: Hands-and-knees crawling is a promising rehabilitation intervention for infants with motor impairments, while research on assistive crawling devices for rehabilitation training was still in its early stages. In particular, precisely generating motion trajectories is a prerequisite to controlling exoskeleton assistive devices, and deep learning-based prediction algorithms, such as Long-Short-Term Memory (LSTM) networks, have proven effective in forecasting joint trajectories of gait. Despite this, no previous studies have focused on forecasting the more variable and complex trajectories of infant crawling. Therefore, this paper aims to explore the feasibility of using LSTM networks to predict crawling trajectories, thereby advancing our understanding of how to actively control crawling rehabilitation training robots.

METHODS: We collected joint trajectory data from 20 healthy infants (11 males and 9 females, aged 8-15 months) as they crawled on hands and knees. This study implemented LSTM networks to forecast bilateral elbow and knee trajectories based on corresponding joint angles. The data set comprised 58, 782 time steps, each containing 4 joint angles. We partitioned the data set into 70% for training and 30% for testing to evaluate predictive performance. We investigated a total of 24 combinations of input and output time-frames, with window sizes for input vectors ranging from 10, 15, 20, 30, 40, 50, 70, and 100 time steps, and output vectors from 5, 10, and 15 steps. Evaluation metrics included Mean Absolute Error (MAE), Mean Squared Error (MSE), and Correlation Coefficient (CC) to assess prediction accuracy.

RESULTS: The results indicate that across various input-output windows, the MAE for elbow joints ranged from 0.280 to 4.976°, MSE ranged from 0.203° to 59.186°, and CC ranged from 89.977% to 99.959%. For knee joints, MAE ranged from 0.277 to 4.262°, MSE from 0.229 to 53.272°, and CC from 89.454% to 99.944%. Results also show that smaller output window sizes lead to lower prediction errors. As expected, the LSTM predicting 5 output time steps has the lowest average error, while the LSTM predicting 15 time steps has the highest average error. In addition, variations in input window size had a minimal impact on average error when the output window size was fixed. Overall, the optimal performance for both elbow and knee joints was observed with input-output window sizes of 30 and 5 time steps, respectively, yielding an MAE of 0.295°, MSE of 0.260°, and CC of 99.938%.

CONCLUSIONS: This study demonstrates the feasibility of forecasting infant crawling trajectories using LSTM networks, which could potentially integrate with exoskeleton control systems. It experimentally explores how different input and output time-frames affect prediction accuracy and sets the stage for future research focused on optimizing models and developing effective control strategies to improve assistive crawling devices.

PMID:40176123 | DOI:10.1186/s12938-025-01360-1

J Imaging Inform Med. 2025 Apr 2. doi: 10.1007/s10278-025-01489-4. Online ahead of print.

ABSTRACT

Chronic kidney disease (CKD) remains a major public health concern, requiring better predictive models for early intervention. This study evaluates a deep learning model (DLM) that utilizes raw chest X-ray (CXR) data to predict moderate to severe kidney function decline. We analyzed data from 79,219 patients with an estimated Glomerular Filtration Rate (eGFR) between 65 and 120, segmented into development (n = 37,983), tuning (n = 15,346), internal validation (n = 14,113), and external validation (n = 11,777) sets. Our DLM, pretrained on CXR-report pairs, was fine-tuned with the development set. We retrospectively examined data spanning April 2011 to February 2022, with a 5-year maximum follow-up. Primary and secondary endpoints included CKD stage 3b progression, ESRD/dialysis, and mortality. The overall concordance index (C-index) values for the internal and external validation sets were 0.903 (95% CI, 0.885-0.922) and 0.851 (95% CI, 0.819-0.883), respectively. In these sets, the incidences of progression to CKD stage 3b at 5 years were 19.2% and 13.4% in the high-risk group, significantly higher than those in the median-risk (5.9% and 5.1%) and low-risk groups (0.9% and 0.9%), respectively. The sex, age, and eGFR-adjusted hazard ratios (HR) for the high-risk group compared to the low-risk group were 16.88 (95% CI, 10.84-26.28) and 7.77 (95% CI, 4.77-12.64), respectively. The high-risk group also exhibited higher probabilities of progressing to ESRD/dialysis or experiencing mortality compared to the low-risk group. Further analysis revealed that the high-risk group compared to the low/median-risk group had a higher prevalence of complications and abnormal blood/urine markers. Our findings demonstrate that a DLM utilizing CXR can effectively predict CKD stage 3b progression, offering a potential tool for early intervention in high-risk populations.

PMID:40175823 | DOI:10.1007/s10278-025-01489-4

Microb Ecol. 2025 Apr 2;88(1):22. doi: 10.1007/s00248-025-02516-1.

ABSTRACT

Shifts in agricultural land use over the past 200 years have led to a loss of nearly 50% of existing wetlands in the USA, and agricultural activities contribute up to 65% of the nutrients that reach the Mississippi River Basin, directly contributing to biological disasters such as the hypoxic Gulf of Mexico "Dead" Zone. Federal efforts to construct and restore wetland habitats have been employed to mitigate the detrimental effects of eutrophication, with an emphasis on the restoration of ecosystem services such as nutrient cycling and retention. Soil microbial assemblages drive biogeochemical cycles and offer a unique and sensitive framework for the accurate evaluation, restoration, and management of ecosystem services. The purpose of this study was to elucidate patterns of soil bacteria within and among wetlands by developing diversity profiles from high-throughput sequencing data, link functional gene copy number of nitrogen cycling genes to measured nutrient flux rates collected from flow-through incubation cores, and predict nutrient flux using microbial assemblage composition. Soil microbial assemblages showed fine-scale turnover in soil cores collected across the topsoil horizon (0-5 cm; top vs bottom partitions) and were structured by restoration practices on the easements (tree planting, shallow water, remnant forest). Connections between soil assemblage composition, functional gene copy number, and nutrient flux rates show the potential for soil bacterial assemblages to be used as bioindicators for nutrient cycling on the landscape. In addition, the predictive accuracy of flux rates was improved when implementing deep learning models that paired connected samples across time.

PMID:40175811 | DOI:10.1007/s00248-025-02516-1

Nat Rev Cancer. 2025 Apr 2. doi: 10.1038/s41568-025-00811-0. Online ahead of print.

NO ABSTRACT

PMID:40175619 | DOI:10.1038/s41568-025-00811-0

Sci Rep. 2025 Apr 2;15(1):11227. doi: 10.1038/s41598-025-92641-1.

ABSTRACT

Strawberry grading by picking robots can eliminate the manual classification, reducing labor costs and minimizing the damage to the fruit. Strawberry size or weight is a key factor in grading, with accurate weight estimation being crucial for proper classification. In this paper, we collected 1521 sets of strawberry RGB-D images using a depth camera and manually measured the weight and size of the strawberries to construct a training dataset for the strawberry weight regression model. To address the issue of incomplete depth images caused by environmental interference with depth cameras, this study proposes a multimodal point cloud completion method specifically designed for symmetrical objects, leveraging RGB images to guide the completion of depth images in the same scene. The method follows a process of locating strawberry pixel regions, calculating centroid coordinates, determining the symmetry axis via PCA, and completing the depth image. Based on this approach, a multimodal fusion regression model for strawberry weight estimation, named MMF-Net, is developed. The model uses the completed point cloud and RGB image as inputs, and extracts features from the RGB image and point cloud by EfficientNet and PointNet, respectively. These features are then integrated at the feature level through gradient blending, realizing the combination of the strengths of both modalities. Using the Percent Correct Weight (PCW) metric as the evaluation standard, this study compares the performance of four traditional machine learning methods, Support Vector Regression (SVR), Multilayer Perceptron (MLP), Linear Regression, and Random Forest Regression, with four point cloud-based deep learning models, PointNet, PointNet++, PointMLP, and Point Cloud Transformer, as well as an image-based deep learning model, EfficientNet and ResNet, on single-modal datasets. The results indicate that among traditional machine learning methods, the SVR model achieved the best performance with an accuracy of 77.7% (PCW@0.2). Among deep learning methods, the image-based EfficientNet model obtained the highest accuracy, reaching 85% (PCW@0.2), while the PointNet + + model demonstrated the best performance among point cloud-based models, with an accuracy of 54.3% (PCW@0.2). The proposed multimodal fusion model, MMF-Net, achieved an accuracy of 87.66% (PCW@0.2), significantly outperforming both traditional machine learning methods and single-modal deep learning models in terms of precision.

PMID:40175474 | DOI:10.1038/s41598-025-92641-1

Sci Rep. 2025 Apr 2;15(1):11349. doi: 10.1038/s41598-025-89210-x.

ABSTRACT

Deep neural networks (DNNs) have demonstrated remarkable capabilities in medical applications, including digital pathology, where they excel at analyzing complex patterns in medical images to assist in accurate disease diagnosis and prognosis. However, concerns have arisen about potential biases in The Cancer Genome Atlas (TCGA) dataset, a comprehensive repository of digitized histopathology data and serves as both a training and validation source for deep learning models, suggesting that over-optimistic results of model performance may be due to reliance on biased features rather than histological characteristics. Surprisingly, recent studies have confirmed the existence of site-specific bias in the embedded features extracted for cancer-type discrimination, leading to high accuracy in acquisition site classification. This biased behavior motivated us to conduct an in-depth analysis to investigate potential causes behind this unexpected biased ability toward site-specific pattern recognition. The analysis was conducted on two cutting-edge DNN models: KimiaNet, a state-of-the-art DNN trained on TCGA images, and the self-trained EfficientNet. In this research study, the balanced accuracy metric is used to evaluate the performance of a model trained to classify data centers, which was originally designed to learn cancerous patterns, with the aim of investigating the potential factors contributing to the higher balanced accuracy in data center detection.

PMID:40175463 | DOI:10.1038/s41598-025-89210-x

Sci Rep. 2025 Apr 2;15(1):11352. doi: 10.1038/s41598-025-91394-1.

ABSTRACT

To address the challenges associated with dense and occluded targets in small target detection utilizing unmanned aerial vehicle (UAV), we propose an enhanced detection algorithm referred as the YOLOv8n-ACW. Building upon the YOLOv8n baseline network model, we have integrated Adown into the Backbone and developed a CCDHead to further improve the drone's capability to recognize small targets. Additionally, WIoU-V3 has been introduced as the loss function. Experiment results derived from the Visdrone2019 dataset indicate that, the YOLOv8n- ACW has achieved a 4.2% increase in mAP50(%) compared to the baseline model, while simultaneously reducing the parameter count by 36.7%, exhibiting superior capabilities in detecting small targets. Furthermore, utilizing a self-constructed dataset of G5-Pro drones for target detection experiments, the results indicate that the enhanced model has robust generalization capabilities in real-world environments. The UAV target detection experiment combines experimental simulation with real-world testing, while combining scientific exploration with educational objectives. This experiment has high fidelity, excellent functional scalability, and strong practicality, aiming to cultivate students' comprehensive practical and innovative abilities.

PMID:40175443 | DOI:10.1038/s41598-025-91394-1

Invest Radiol. 2025 Apr 2. doi: 10.1097/RLI.0000000000001182. Online ahead of print.

ABSTRACT

OBJECTIVES: Respiratory motion poses a significant challenge for clinical workflows in diagnostic imaging and radiation therapy. Many technologies such as motion artifact reduction and tumor tracking have been developed to compensate for its effect. To assess these technologies, respiratory motion phantoms (RMPs) are required as preclinical testing environments, for instance, in computed tomography (CT). However, current CT RMPs are highly simplified and do not exhibit realistic tissue structures or deformation patterns. With the rise of more complex motion compensation technologies such as deep learning-based algorithms, there is a need for more realistic RMPs. This work introduces PixelPrint4D, a 3D printing method for fabricating lifelike, patient-specific deformable lung phantoms for CT imaging.

MATERIALS AND METHODS: A 4DCT dataset of a lung cancer patient was acquired. The volumetric image data of the right lung at end inhalation was converted into 3D printer instructions using the previously developed PixelPrint software. A flexible 3D printing material was used to replicate variable densities voxel-by-voxel within the phantom. The accuracy of the phantom was assessed by acquiring CT scans of the phantom at rest, and under various levels of compression. These phantom images were then compiled into a pseudo-4DCT dataset and compared to the reference patient 4DCT images. Metrics used to assess the phantom structural accuracy included mean attenuation errors, 2-sample 2-sided Kolmogorov-Smirnov (KS) test on histograms, and structural similarity index (SSIM). The phantom deformation properties were assessed by calculating displacement errors of the tumor and throughout the full lung volume, attenuation change errors, and Jacobian errors, as well as the relationship between Jacobian and attenuation changes.

RESULTS: The phantom closely replicated patient lung structures, textures, and attenuation profiles. SSIM was measured as 0.93 between the patient and phantom lung, suggesting a high level of structural accuracy. Furthermore, it exhibited realistic nonrigid deformation patterns. The mean tumor motion errors in the phantom were ≤0.7 ± 0.6 mm in each orthogonal direction. Finally, the relationship between attenuation and local volume changes in the phantom had a strong correlation with that of the patient, with analysis of covariance yielding P = 0.83 and f = 0.04, suggesting no significant difference between the phantom and patient.

CONCLUSIONS: PixelPrint4D facilitates the creation of highly realistic RMPs, exceeding the capabilities of existing models to provide enhanced testing environments for a wide range of emerging CT technologies.

PMID:40173424 | DOI:10.1097/RLI.0000000000001182

Aesthet Surg J. 2025 Apr 2:sjaf047. doi: 10.1093/asj/sjaf047. Online ahead of print.

ABSTRACT

BACKGROUND: Breast Implant Illness (BII) is a spectrum of symptoms some people attribute to breast implants. While causality remains unproven, patient interest has grown significantly. Understanding patient perceptions of BII on social media is crucial as these platforms increasingly influence healthcare decisions.

OBJECTIVES: The purpose of this study is to analyze patient perceptions and emotional responses to BII on social media using RoBERTa, a natural processing model trained on 124 million X posts.

METHODS: Posts mentioning BII from 2014-2023 were analyzed using two NLP models: one for sentiment (positive/negative) and another for emotions (fear, sadness, anger, disgust, neutral, surprise, and joy). Posts were then classified by their highest-scoring emotion. Results were compared over across 2014-2018 and 2019-2023, with correlation analysis (Pearson correlation coefficient) between published implant explantation and augmentation data.

RESULTS: Analysis of 6,099 posts over 10 years showed 75.4% were negative, with monthly averages of 50.85 peaking at 213 in March 2019. Fear and neutral emotions dominated, representing 35.9% and 35.6% respectively. The strongest emotions were neutral and fear, with an average score of 0.293 and 0.286 per post, respectively. Fear scores increased from 0.219 (2014-2018) to 0.303 (2019-2023). Strong positive correlations (r>0.70) existed between annual explantation rates/explantation-to-augmentation ratios and total, negative, neutral, and fear posts.

CONCLUSIONS: BII discourse on X peaked in 2019, characterized predominantly by negative sentiment and fear. The strong correlation between fear/negative-based posts and explantation rates suggests social media discourse significantly influences patient decisions regarding breast implant removal.

PMID:40173420 | DOI:10.1093/asj/sjaf047

PLoS One. 2025 Apr 2;20(4):e0319078. doi: 10.1371/journal.pone.0319078. eCollection 2025.

ABSTRACT

Hepatitis is a widespread inflammatory condition of the liver, presenting a formidable global health challenge. Accurate and timely detection of hepatitis is crucial for effective patient management, yet existing methods exhibit limitations that underscore the need for innovative approaches. Early-stage detection of hepatitis is now possible with the recent adoption of machine learning and deep learning approaches. With this in mind, the study investigates the use of traditional machine learning models, specifically classifiers such as logistic regression, support vector machines (SVM), decision trees, random forest, multilayer perceptron (MLP), and other models, to predict hepatitis infections. After extensive data preprocessing including outlier detection, dataset balancing, and feature engineering, we evaluated the performance of these models. We explored three modeling approaches: machine learning with default hyperparameters, hyperparameter-tuned models using GridSearchCV, and ensemble modeling techniques. The SVM model demonstrated outstanding performance, achieving 99.25% accuracy and a perfect AUC score of 1.00 with consistency in other metrics with 99.27% precision, and 99.24% for both recall and F1-measure. The MLP and Random Forest proved to be in pace with the superior performance of SVM exhibiting an accuracy of 99.00%. To ensure robustness, we employed a 5-fold cross-validation technique. For deeper insight into model interpretability and validation, we employed an explainability analysis of our best-performed models to identify the most effective feature for hepatitis detection. Our proposed model, particularly SVM, exhibits better prediction performance regarding different performance metrics compared to existing literature.

PMID:40173410 | DOI:10.1371/journal.pone.0319078

Sci Adv. 2025 Apr 4;11(14):eads5185. doi: 10.1126/sciadv.ads5185. Epub 2025 Apr 2.

ABSTRACT

Atlantic and Benguela Niño events substantially affect the tropical Atlantic region, with far-reaching consequences on local marine ecosystems, African climates, and El Niño Southern Oscillation. While accurate forecasts of these events are invaluable, state-of-the-art dynamic forecasting systems have shown limited predictive capabilities. Thus, the extent to which the tropical Atlantic variability is predictable remains an open question. This study explores the potential of deep learning in this context. Using a simple convolutional neural network architecture, we show that Atlantic/Benguela Niños can be predicted up to 3 to 4 months ahead. Our model excels in forecasting peak-season events with remarkable accuracy extending lead time to 5 months. Detailed analysis reveals our model's ability to exploit known physical precursors, such as long-wave ocean dynamics, for accurate predictions of these events. This study challenges the perception that the tropical Atlantic is unpredictable and highlights deep learning's potential to advance our understanding and forecasting of critical climate events.

PMID:40173237 | DOI:10.1126/sciadv.ads5185

Sci Adv. 2025 Apr 4;11(14):eadp0558. doi: 10.1126/sciadv.adp0558. Epub 2025 Apr 2.

ABSTRACT

Historical records of climate fields are often sparse because of missing measurements, especially before the introduction of large-scale satellite missions. Several statistical and model-based methods have been introduced to fill gaps and reconstruct historical records. Here, we use a recently introduced deep learning approach based on Fourier convolutions, trained on numerical climate model output, to reconstruct historical climate fields. Using this approach, we are able to realistically reconstruct large and irregular areas of missing data and to reproduce known historical events, such as strong El Niño or La Niña events, with very little given information. Our method outperforms the widely used statistical kriging method, as well as other recent machine learning approaches. The model generalizes to higher resolutions than the ones it was trained on and can be used on a variety of climate fields. Moreover, it allows inpainting of masks never seen before during the model training.

PMID:40173235 | DOI:10.1126/sciadv.adp0558

Global Spine J. 2025 Apr 2:21925682251332556. doi: 10.1177/21925682251332556. Online ahead of print.

ABSTRACT

Study DesignRetrospective cohort.ObjectiveDespite numerous studies highlighting patient comorbidities and surgical factors in postoperative success, the role of social determinants of health (SDH) in anterior cervical discectomy and fusion (ACDF) outcomes remains unexplored. This study evaluates the predictive impact of SDH on 90-day readmission and health utilization (HU) in ACDF patients using machine learning (ML).MethodsWe analyzed 3127 ACDF patients (2003-2023) from a multisite academic center, incorporating over 35 clinical and demographic variables. SDH characteristics were assessed using the Social Vulnerability Index. Primary outcomes included 90-day readmission and postoperative HU. ML models were developed and validated by the area under the curve (AUC) for readmission and mean absolute error (MAE) for HU. Feature importance analysis identified key predictors.ResultsBalanced Random Forest (AUC = 0.75) best predicted 90-day readmission, with length of stay, Elixhauser score, and Medicare status as top predictors. Among SDH factors, minority status & language, household composition & disability, socioeconomic status, and housing type & transportation were influential. Support Vector Regression (MAE = 1.96) best predicted HU, with perioperative duration, socioeconomic status, and minority status & language as key predictors.ConclusionsFindings highlight SDH's role in ACDF outcomes, suggesting the value of stratifying for interventions such as targeted resource allocation, language-concordant care, and tailored follow-up. While reliance on a single healthcare system and proxy SDH measures are limitations, this is the first study to apply ML to assess SDH in ACDF patients. Further validation with direct patient-reported SDH data is needed to refine predictive models.

PMID:40173192 | DOI:10.1177/21925682251332556

IEEE Trans Pattern Anal Mach Intell. 2025 Apr 2;PP. doi: 10.1109/TPAMI.2025.3557245. Online ahead of print.

ABSTRACT

This paper introduces a one-stage deep uncalibrated photometric stereo (UPS) network, namely Fourier Uncalibrated Photometric Stereo Network (FUPS-Net), for non-Lambertian objects under unknown light directions. It departs from traditional two-stage methods that first explicitly learn lighting information and then estimate surface normals. Two-stage methods were deployed because the interplay of lighting with shading cues presents challenges for directly estimating surface normals without explicit lighting information. However, these two-stage networks are disjointed and separately trained so that the error in explicit light calibration will propagate to the second stage and cannot be eliminated. In contrast, the proposed FUPS-Net utilizes an embedded Fourier transform network to implicitly learn lighting features by decomposing inputs, rather than employing a disjointed light estimation network. Our approach is motivated from observations in the Fourier domain of photometric stereo images: lighting information is mainly encoded in amplitudes, while geometry information is mainly associated with phases. Leveraging this property, our method "decomposes" geometry and lighting in the Fourier domain as guidance, via the proposed Fourier Embedding Extraction (FEE) block and Fourier Embedding Aggregation (FEA) block, which generate lighting and geometry features for the FUPS-Net to implicitly resolve the geometry-lighting ambiguity. Furthermore, we propose a Frequency-Spatial Weighted (FSW) block that assigns weights to combine features extracted from the frequency domain and those from the spatial domain for enhancing surface reconstructions. FUPS-Net overcomes the limitations of two-stage UPS methods, offering better training stability, a concise end-to-end structure, and avoiding accumulated errors in disjointed networks. Experimental results on synthetic and real datasets demonstrate the superior performance of our approach, and its simpler training setup, potentially paving the way for a new strategy in deep learning-based UPS methods.

PMID:40173071 | DOI:10.1109/TPAMI.2025.3557245

Mol Divers. 2025 Apr 2. doi: 10.1007/s11030-025-11145-2. Online ahead of print.

ABSTRACT

Farnesoid X receptor (FXR) is a key regulator of bile acid, lipid, and glucose homeostasis, making it a promising target for treating metabolic diseases. FXR antagonists have shown therapeutic potential in cholestasis, metabolic disorders, and certain cancers, while clinically approved FXR antagonists remain unavailable and underrepresented in current treatment strategies. To address this, we developed deep learning models for predicting FXR antagonistic activity (ANTCL) and toxicity (TOXCL). Screening 217,345 compounds from the HMDB database identified eleven human metabolite candidates with significant FXR binding potential. Molecular dynamics simulations and binding free energy calculations revealed five more stable complexes compared to the reference compound Gly-MCA, with HMDB0253354 (Fulvestrant) and HMDB0242367 (ZM 189154) standing out for their binding free energies. Hydrophobic interactions, particularly involving residues MET328, PHE329, and ALA291, contributed to their stability. These results demonstrate the effectiveness of deep learning in FXR antagonist discovery and highlight the potential of HMDB0253354 and HMDB0242367 as promising candidates for metabolic disease treatment.

PMID:40172823 | DOI:10.1007/s11030-025-11145-2

Med Biol Eng Comput. 2025 Apr 2. doi: 10.1007/s11517-025-03347-5. Online ahead of print.

ABSTRACT

Fetal congenital heart disease (FCHD) represents a serious and prevalent congenital malformation. However, there exist notable regional disparities in the detection rates of fetal heart abnormalities. To enhance the diagnostic capabilities of ultrasound physicians in primary hospitals regarding fetal heart structures, the adoption of artificial intelligence technology to assist in acquiring high-quality, standard fetal echocardiographic images is of paramount importance. Currently, primary hospitals face challenges in recognizing standard views in fetal echocardiography, particularly under resource-constrained conditions. Efficient and accurate identification of fetal heart structures has become an urgent issue to address. Despite existing research efforts dedicated to the recognition of standard views in fetal echocardiography, current methods still suffer from limitations in computational complexity, feature extraction capabilities, and long-distance feature capturing, hindering their widespread application in ultrasound diagnosis at primary hospitals. Specifically, the literature lacks an efficient and robust model that can effectively balance high accuracy in standard view recognition with low computational complexity and fast inference times. The need for a model that can accurately capture long-distance features while maintaining efficiency is particularly acute in the context of primary hospitals, where resources are limited and the demand for accurate fetal heart assessments is high. To address these issues, the present study proposes an efficient network based on a state-space model trained with evidence for standard view recognition in fetal echocardiography. This method integrates a visual state space (VSS) model, which boasts powerful feature extraction capabilities and effective long-distance feature capturing, while significantly reducing computational complexity and facilitating efficient model inference. In the collected dataset, the proposed model achieved an accuracy of 99.32% and an F1-score of 99.29% in identifying eight standard views of fetal echocardiography. Furthermore, the model exhibited the lowest floating point operations per second (FLOPs), parameters, and inference time, while achieving the highest frames per second (FPS). This achievement not only provides a solid technical foundation for intelligent diagnosis of FCHD but also serves as an auxiliary tool for junior or novice sonographers at primary hospitals in acquiring basic views of fetal heart structures.

PMID:40172789 | DOI:10.1007/s11517-025-03347-5

Med Biol Eng Comput. 2025 Apr 2. doi: 10.1007/s11517-025-03323-z. Online ahead of print.

ABSTRACT

Ultrasound (US) imaging enables the evaluation of vascular structures in real time, and it can provide morphological and pathological information during US-guided procedures. Automatic prediction of vascular structure boundaries can help clinicians in locating and measuring target structures more accurately and efficiently. Most existing US segmentation methods use per-pixel classification or regression, which require post-processing to obtain contour coordinates. In this work, we present ACE-Net, a novel approach that directly predicts the contour coordinates for every scanning line (A-line) in US images. ACE-Net combines two main modules: a boundary regression module that predicts the upper and lower coordinates of the target area for each A-line, and an A-line classification module that determines whether an A-line belongs to the target area or not. We evaluated our method on three clinical US datasets using, among others, dice similarity coefficient (DSC) and inference time as performance metrics. Our method outperformed state-of-the-art segmentation methods in inference time while achieving superior or comparable performance in DSC. ACE-Net is publicly available at https://github.com/bfarolabarata/ace-net .

PMID:40172788 | DOI:10.1007/s11517-025-03323-z

Eur Radiol. 2025 Apr 2. doi: 10.1007/s00330-025-11552-x. Online ahead of print.

NO ABSTRACT

PMID:40172639 | DOI:10.1007/s00330-025-11552-x

Eur Radiol. 2025 Apr 2. doi: 10.1007/s00330-025-11550-z. Online ahead of print.

NO ABSTRACT

PMID:40172638 | DOI:10.1007/s00330-025-11550-z