Sci Data. 2025 Apr 15;12(1):628. doi: 10.1038/s41597-025-04938-5.

ABSTRACT

Ontologies have gained popularity in the scientific community as a way to standardize terminologies in organizations' data. Although certain cohorts have created frameworks with rules and guidelines on creating ontologies, there exist significant variations in how Materials Science ontologies are currently developed. We seek to provide guidance in the form of a unified automated framework for developing interoperable and modular ontologies for Materials Data Science that simplifies the ontology terms matching by establishing a semantic bridge up to the Basic Formal Ontology(BFO). This framework provides key recommendations on how ontologies should be positioned within the semantic web, what knowledge representation language is recommended, and where ontologies should be published online to boost their findability and interoperability. Two fundamental components of the MDS-Onto framework are the bilingual package called FAIRmaterials for ontology creation and FAIRLinked, for FAIR data creation. To showcase the practical capabilities of FAIRmaterials, we present two exemplar domain ontologies of MDS-Onto: Synchrotron X-Ray Diffraction and Photovoltaics.

PMID:40234492 | DOI:10.1038/s41597-025-04938-5

BMJ Open Qual. 2025 Apr 15;14(2):e002847. doi: 10.1136/bmjoq-2024-002847.

ABSTRACT

Aminoglycoside antibiotics cause ototoxicity for which baseline audiometric testing is recommended but often not done. Barriers to successful implementation include limited availability of sound-booths and audiologists. An ototoxicity monitoring programme (OMP) was implemented using tablet-based audiometry (TBA) by non-audiologists.A quality improvement project conducted over 1 year (19 April 2021 to 18 April 2022), using Plan Do Study Act (PDSA) cycles, monitored the adherence to the OMP using Shoebox Standard Edition application on iPads. Barriers to adoption were identified to determine potential solutions for improved adherence. Adult respiratory patients (cystic fibrosis (CF), bronchiectasis, non-tuberculosis mycobacteria (NTM) infection) aged 17-82 years receiving >1 day of intravenous aminoglycosides (IVAGs) at a single tertiary-referral hospital were included. Other reported outcomes were patient characteristics, risk factors associated with abnormal hearing and ototoxic shift.73 patients were tested in the OMP (46 received ≥2 hearing tests) giving an overall adherence rate of 69% after 12 months. Patient identification using referral and reporting systems initially improved adherence from 36% to 88% (p=0.03) during PDSA 1. Barriers to successful adherence were staff availability and COVID-19 infection outbreaks (p=0.057). Older age (p<0.001), higher Body Mass Index (p=0.041), non-CF bronchiectasis (p=0.01), non-CF NTM (p=0.028) and higher Hearing Handicap Inventory for Adults scores (p=0.002) were significantly associated with abnormal baseline hearing. 78% with hearing loss were asymptomatic. Ototoxic shift was associated with gentamicin use compared to amikacin/tobramycin (p=0.027). TBA was associated with high usability in patients ≤50 years old.TBA by non-audiologists was feasible and demonstrated good patient usability, permitting screening of patients within 72 hours of starting IVAG and earlier referrals for formal audiometry. Hearing loss and ototoxicity were detected at earlier stages, enabling more rapid decision-making and treatment modification.

PMID:40233985 | DOI:10.1136/bmjoq-2024-002847

Eur J Med Res. 2025 Apr 15;30(1):288. doi: 10.1186/s40001-025-02501-x.

ABSTRACT

BACKGROUND: The diagnosis of idiopathic pulmonary fibrosis (IPF) is complex, which requires lung biopsy, if necessary, and multidisciplinary discussions with specialists. Clinical diagnosis of the two ailments is particularly challenging due to the impact of interobserver variability. Several studies have endeavored to utilize image-based machine learning to diagnose IPF and its subtype of usual interstitial pneumonia (UIP). However, the diagnostic accuracy of this approach lacks evidence-based support.

OBJECTIVE: We conducted a systematic review and meta-analysis to explore the diagnostic efficiency of image-based machine learning (ML) for IPF.

DATA SOURCES AND METHODS: We comprehensively searched PubMed, Cochrane, Embase, and Web of Science databases up to August 24, 2024. During the meta-analysis, we carried out subgroup analyses by imaging source (computed radiography/computed tomography) and modeling type (deep learning/other) to evaluate its diagnostic performance for IPF.

RESULTS: The meta-analysis findings indicated that in the diagnosis of IPF, the C-index, sensitivity, and specificity of ML were 0.93 (95% CI 0.89-0.97), 0.79 (95% CI 0.73-0.83), and 0.84 (95% CI 0.79-0.88), respectively. The sensitivity of radiologists/clinicians in diagnosing IPF was 0.69 (95% CI 0.56-0.79), with a specificity of 0.93 (95% CI 0.74-0.98). For UIP diagnosis, the C-index of ML was 0.91 (95% CI 0.87-0.94), with a sensitivity of 0.92 (95% CI 0.80-0.97) and a specificity of 0.92 (95%CI 0.82-0.97). In contrast, the sensitivity of radiologists/clinicians in diagnosing UIP was 0.69 (95% CI 0.50-0.84), with a specificity of 0.90 (95% CI 0.82-0.94).

CONCLUSIONS: Image-based machine learning techniques demonstrate robust data processing and recognition capabilities, providing strong support for accurate diagnosis of idiopathic pulmonary fibrosis and usual interstitial pneumonia. Future multicenter large-scale studies are warranted to develop more intelligent evaluation tools to further enhance clinical diagnostic efficiency. Trial registration This study protocol was registered with PROSPERO (CRD42022383162).

PMID:40235000 | DOI:10.1186/s40001-025-02501-x

Eur J Med Res. 2025 Apr 15;30(1):293. doi: 10.1186/s40001-025-02553-z.

ABSTRACT

BACKGROUND: There is no standard practice for intensive care admission after non-small cell lung cancer surgery. In this study, we aimed to determine the need for intensive care admission after non-small cell lung cancer surgery with deep learning models.

METHODS: The data of 953 patients who were operated for non-small cell lung cancer between January 2001 and 2023 was analyzed. Clinical, laboratory, respiratory, tumor's radiological and surgical features were included as input data in the study. The outcome data was intensive care unit admission. Deep learning was performed with the Fully Connected Neural Network algorithm and k-fold cross validation method.

RESULTS: The training accuracy value was 92.0%, the training F1 1 score of the algorithm was 86.7%, the training F1 0 value was 94.2%, and the training F1 average score was 90.5%. The test sensitivity value of the algorithm was 67.7%, the test positive predictive value was 84.0%, and the test accuracy value was 85.3%. Test F1 1 score was 75.0%, test F1 0 score was 89.5%, and test F1 average score was 82.3%. The AUC in the ROC curve created for the success analysis of the algorithm's test data was 0.83.

CONCLUSIONS: Using our method deep learning models predicted the need for intensive care unit admission with high success and confidence values. The use of artificial intelligence algorithms for the necessity of intensive care hospitalization will ensure that postoperative processes are carried out safely using objective decision mechanisms.

PMID:40234958 | DOI:10.1186/s40001-025-02553-z

BMC Med Inform Decis Mak. 2025 Apr 15;25(1):165. doi: 10.1186/s12911-025-02981-1.

ABSTRACT

BACKGROUND: Mortality prediction is critical in clinical care, particularly in intensive care units (ICUs), where early identification of high-risk patients can inform treatment decisions. While deep learning (DL) models have demonstrated significant potential in this task, most suffer from limited generalizability, which hinders their widespread clinical application. Additionally, the class imbalance in electronic health records (EHRs) complicates model training. This study aims to develop a causally-informed prediction model that incorporates underlying causal relationships to mitigate class imbalance, enabling more stable mortality predictions.

METHODS: This study introduces the CRISP model (Causal Relationship Informed Superior Prediction), which leverages native counterfactuals to augment the minority class and constructs patient representations by incorporating causal structures to enhance mortality prediction. Patient data were obtained from the public MIMIC-III and MIMIC-IV databases, as well as an additional dataset from the West China Hospital of Sichuan University (WCHSU).

RESULTS: A total of 69,190 ICU cases were included, with 30,844 cases from MIMIC-III, 27,362 cases from MIMIC-IV, and 10,984 cases from WCHSU. The CRISP model demonstrated stable performance in mortality prediction across the 3 datasets, achieving AUROC (0.9042-0.9480) and AUPRC (0.4771-0.7611). CRISP's data augmentation module showed predictive performance comparable to commonly used interpolation-based oversampling techniques.

CONCLUSION: CRISP achieves better generalizability across different patient groups, compared to various baseline algorithms, thereby enhancing the practical application of DL in clinical decision support.

TRIAL REGISTRATION: Trial registration information for the WCHSU data is available on the Chinese Clinical Trial Registry website ( http://www.chictr.org.cn ), with the registration number ChiCTR1900025160. The recruitment period for the data was from August 5, 2019, to August 31, 2021.

PMID:40234903 | DOI:10.1186/s12911-025-02981-1

Plant Methods. 2025 Apr 15;21(1):51. doi: 10.1186/s13007-025-01369-6.

ABSTRACT

Posture is a critical phenotypic trait that reflects crop growth and serves as an essential indicator for both agricultural production and scientific research. Accurate pose estimation enables real-time tracking of crop growth processes, but in field environments, challenges such as variable backgrounds, dense planting, occlusions, and morphological changes hinder precise posture analysis. To address these challenges, we propose PFLO (Pose Estimation Model of Field Maize Based on YOLO Architecture), an end-to-end model for maize pose estimation, coupled with a novel data processing method to generate bounding boxes and pose skeleton data from a"keypoint-line"annotated phenotypic database which could mitigate the effects of uneven manual annotations and biases. PFLO also incorporates advanced architectural enhancements to optimize feature extraction and selection, enabling robust performance in complex conditions such as dense arrangements and severe occlusions. On a fivefold validation set of 1,862 images, PFLO achieved 72.2% pose estimation mean average precision (mAP50) and 91.6% object detection mean average precision (mAP50), outperforming current state-of-the-art models. The model demonstrates improved detection of occluded, edge, and small targets, accurately reconstructing skeletal poses of maize crops. PFLO provides a powerful tool for real-time phenotypic analysis, advancing automated crop monitoring in precision agriculture.

PMID:40234900 | DOI:10.1186/s13007-025-01369-6

Sci Rep. 2025 Apr 15;15(1):12899. doi: 10.1038/s41598-025-97258-y.

ABSTRACT

The current study investigates the application of artificial intelligence (AI) techniques, including machine learning (ML) and deep learning (DL), in predicting the ultimate load-carrying capacity and ultimate strain ofboth hollow and solid hybrid elliptical fiber-reinforced polymer (FRP)-concrete-steel double-skin tubular columns (DSTCs) under axial loading. Implemented AI techniques include five ML models - Gene Expression Programming (GEP), Artificial Neural Network (ANN), Random Forest (RF), Adaptive Boosting (ADB), and eXtreme Gradient Boosting (XGBoost) - and one DL model - Deep Neural Network (DNN).Due to the scarcity of experimental data on hybrid elliptical DSTCs, an accurate finite element (FE) model was developed to provide additional numerical insights. The reliability of the proposed nonlinear FE model was validated against existing experimental results. The validated model was then employed in a parametric study to generate 112 data points.The parametric study examined the impact of concrete strength, the cross-sectional size of the inner steel tube, and FRP thickness on the ultimate load-carrying capacity and ultimate strain of both hollow and solid hybrid elliptical DSTCs.The effectiveness of the AI application was assessed by comparing the models' predictions with FE results.Among the models, XGBoost and RF achieved the best performance in both training and testing with respect to the determination coefficient (R2), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) values. The study provided insights into the contributions of individual features to predictions using the SHapley Additive exPlanations (SHAP) approach. The results from SHAP, based on the best prediction performance of the XGBoost model, indicate that the area of the concrete core has the most significant effect on the load-carrying capacity of hybrid elliptical DSTCs, followed by the unconfined concrete strength and the total thickness of FRP multiplied by its elastic modulus. Additionally, a user interface platform was developed to streamline the practical application of the proposed AI models in predicting the axial capacity of DSTCs.

PMID:40234698 | DOI:10.1038/s41598-025-97258-y

Sci Rep. 2025 Apr 15;15(1):13039. doi: 10.1038/s41598-025-96949-w.

ABSTRACT

Osteoporosis, a bone disease characterized by decreased bone mineral density (BMD) resulting in decreased mechanical strength and an increased fracture risk, remains poorly understood in children. Herein, we developed/validated a deep learning-based model to predict pediatric BMD using plain spine radiographs. Using a two-stage model, Yolov8 was applied for vertebral body detection to predict BMD values using a regression model based on ResNet-18, from which a low-BMD group was classified based on Z-scores of predicted BMD. Patients aged 10-20-years who underwent dual-energy X-ray absorptiometry and radiography within 6 months at our hospital were enrolled. Ultimately, 601 patients (mean age, 14 years 4 months [SD 2 years]; 276 males) were included. The model achieved robust performance in detecting vertebral bodies (average precision [AP] 50 = 0.97, AP [50:95] = 0.68) and predicting BMD, with significant correlation (r = 0.72), showing consistency across different vertebral segments and agreement (intraclass correlation coefficient: 0.64). Moreover, it successfully classified low-BMD groups (area under the receiver operating characteristic curve = 0.85) with high sensitivity (0.76) and specificity (0.87). This deep-learning approach shows promise for BMD prediction and classification, with potential to enhance early detection and streamline bone health management in high-risk pediatric populations.

PMID:40234697 | DOI:10.1038/s41598-025-96949-w

Sci Rep. 2025 Apr 15;15(1):13032. doi: 10.1038/s41598-025-97391-8.

ABSTRACT

Glycogen storage disease (GSD) is a group of rare inherited metabolic disorders characterized by abnormal glycogen storage and breakdown. These disorders are caused by mutations in G6PC1, which is essential for proper glucose storage and metabolism. With the advent of continuous glucose monitoring systems, development of algorithms to analyze and predict glucose levels has gained considerable attention, with the aim of preemptively managing fluctuations before they become problematic. However, there is a lack of research focusing specifically on patients with GSD. Therefore, this study aimed to forecast glucose levels in patients with GSD using state-of-the-art deep-learning (DL) algorithms. This retrospective study utilized blood glucose data from patients with GSD who were either hospitalized or managed at Yonsei University Wonju Severance Christian Hospital, Korea, between August 2020 and February 2024. In this study, three state-of-the-art DL models for time-series forecasting were employed: PatchTST, LTSF N-Linear, and TS Mixer. First, the models were used to predict the patients' Glucose levels for the next hour. Second, a binary classification task was performed to assess whether hypoglycemia could be predicted alongside direct glucose levels. Consequently, this is the first study to demonstrate the capability of forecasting glucose levels in patients with GSD using continuous glucose-monitoring data and DL models. Our model provides patients with GSD with a more accessible tool for managing glucose levels. This study has a broader effect, potentially serving as a foundation for improving the care of patients with rare diseases using DL-based solutions.

PMID:40234688 | DOI:10.1038/s41598-025-97391-8

Sci Rep. 2025 Apr 15;15(1):13038. doi: 10.1038/s41598-025-97351-2.

ABSTRACT

Membranous nephropathy (MN) is one of the most common glomerular diseases. Although the diagnostic method based on serum PLA2R antibodies has gradually been applied in clinical practice, only 52-86% of PLA2R-associated MN patients show positive anti-PLA2R antibodies. Therefore, renal biopsy remains the gold standard for diagnosing MN. However, the renal biopsy procedure is highly complex and involves multiple steps, including tissue sampling, fixation, dehydration, embedding, sectioning, PAS staining, Masson trichrome staining, and silver staining. Each step requires precise technique from laboratory personnel, as any error can affect the quality of the final tissue sections and, consequently, the diagnosis. As a result, there is an urgent need to develop a method that enables rapid diagnosis after renal biopsy. Previous studies have shown that Raman spectroscopy offers promising results for diagnosing MN, exhibiting high sensitivity and specificity when applied to human serum and urine samples. In this study, we propose a rapid diagnostic method combining Raman spectroscopy of mouse kidney tissue with a CNN-BiLSTM deep learning model. The model achieved 98% accuracy, with specificity and sensitivity of 98.3%, providing a novel auxiliary tool for the pathological diagnosis of MN.

PMID:40234682 | DOI:10.1038/s41598-025-97351-2

Sci Rep. 2025 Apr 15;15(1):12966. doi: 10.1038/s41598-025-97132-x.

ABSTRACT

Pavement crack serves as a crucial indicator of road condition, directly associated with subsequent pavement deterioration. To address the demand for large-scale real-time pavement damage assessment, this study proposes a lightweight pavement damage detection model based on YOLOv5s (LPDD-YOLO). Initially, a lightweight feature extraction network, FasterNet, is adopted to reduce the number of parameters and computational complexity. Secondly, to mitigate the reduction in accuracy resulting from the usage of lightweight network, the attention-based downsampling module and the neural network cognitive module are introduced. These modules aim to enhance the feature extraction robustness and to eliminate interference from irrelevant features. In addition, considering the significant variation in aspect ratios and diverse morphologies of pavement damages, K-Means clustering and the deformable convolution module are employed. These mechanisms ensure dynamic anchor feature selection and extend the scope of deformation ability, respectively. According to the ablation experiment on a self-built dataset, LPDD-YOLO demonstrates notable improvements in both accuracy and efficiency compared to the original model. Specifically, the mAP increases by 4.1%, and the F1 score rises by 5.3%. Moreover, LPDD-YOLO can obtain a 47.3% reduction in parameters and a 54.4% decrease in GFLOPs. It is noteworthy that LPDD-YOLO achieves real-time and accurate damage detection, with a speed of up to 85 FPS. The effectiveness and superiority of LPDD-YOLO are further substantiated through comparisons with other state-of-the-art algorithms.

PMID:40234635 | DOI:10.1038/s41598-025-97132-x

Sci Rep. 2025 Apr 15;15(1):12927. doi: 10.1038/s41598-025-97563-6.

ABSTRACT

Accurate prediction of maize yields is crucial for effective crop management. In this paper, we propose a novel deep learning framework (CNNAtBiGRU) for estimating maize yield, which is applied to typical black soil areas in Northeast China. This framework integrates a one-dimensional convolutional neural network (1D-CNN), bidirectional gated recurrent units (BiGRU), and an attention mechanism to effectively characterize and weight key segments of input data. In the predictions for the most recent year, the model demonstrated high accuracy (R² = 0.896, RMSE = 908.33 kg/ha) and exhibited strong robustness in both earlier years and during extreme climatic events. Unlike traditional yield estimation methods that primarily rely on remote sensing vegetation indices, phenological data, meteorological data, and soil characteristics, this study innovatively incorporates anthropogenic factors, such as Degree of Cultivation Mechanization (DCM), reflecting the rapid advancement of agricultural modernization. The relative importance analysis of input variables revealed that Enhanced Vegetation Index (EVI), Sun-Induced Chlorophyll Fluorescence (SIF), and DCM were the most influential factors in yield prediction. Furthermore, our framework enables maize yield prediction 1-2 months in advance by leveraging historical patterns of environmental and agricultural variables, providing valuable lead time for decision-making. This predictive capability does not rely on forecasting future weather conditions but rather captures yield-relevant signals embedded in early-season data.

PMID:40234562 | DOI:10.1038/s41598-025-97563-6

Pediatr Infect Dis J. 2025 May 1;44(5):428-430. doi: 10.1097/INF.0000000000004683. Epub 2025 Jan 10.

ABSTRACT

We investigated the steady-state pharmacokinetics of generic abacavir (ABC)/lamivudine (3TC) dispersible tablets (DTs) in young children living with HIV aged 3 months to <7 years, weighing 6 to <20 kg. Twenty-eight Thai children were enrolled and received ABC/3TC-DT plus pediatric dolutegravir-DT once daily. ABC/3TC was administered using WHO weight band (WB) doses: 180/90 mg, 240/120 mg and 300/150 mg for children weighing 6 to <10 kg (WB 1, n = 7), 10 to <14 kg (WB2, n = 9) and 14 to <20 kg (WB3, n = 12), respectively. ABC geometric mean (GM) AUC 0-24 h (CV%) was 14.2 (50.5%), 15.6 (32.6%) and 20.7 (28.3%) mg.h/L, respectively, and 3TC GM AUC 0-24 h was 14.8 (44.1%), 18.2 (28.0%) and 19.9 (26.2%) mg.h/L, respectively. ABC and 3TC exposures were within target ranges across WBs. These data supported current ABC/3TC-DT WHO-weight band dosing guidance for young children in Thailand.

PMID:40232882 | DOI:10.1097/INF.0000000000004683

Curr Opin Allergy Clin Immunol. 2025 Apr 11. doi: 10.1097/ACI.0000000000001077. Online ahead of print.

ABSTRACT

PURPOSE OF REVIEW: Allergen immunotherapy (AIT) is increasingly popular as a treatment strategy for food allergy. Unfortunately, there is significant heterogeneity in reported outcomes, specifically in the dose-thresholds selected for evaluation and the symptoms used to define a "tolerated dose". These considerations are often investigator-driven and do not consider patient perspectives.

RECENT FINDINGS: A systematic review by the EAACI CO-FAITH taskforce recently flagged the need to better standardize and harmonize outcomes used in clinical trials of food-AIT. Using less objective symptoms to define dose-limiting symptoms can underestimate the reaction threshold determined at baseline food challenge. As a consequence, this can overestimate the efficacy of food-AIT by 15%. In this review, we perform an individual patient data (IPD) meta-analysis using data from three randomized-controlled trials and one real-world registry, to evaluate how the definition of dose "tolerance" impacts upon reported desensitization rates.

SUMMARY: This analysis provides insight into how clinical efficacy rates for food-AIT are impacted by using different dose thresholds and definitions for when a dose might be consider tolerated. Using more patient-centric outcomes may be a more useful metric to harmonize reporting of outcomes and inform clinical practice, paving the way towards reaching a consensus on outcome reporting in trials of food-AIT.

PMID:40233247 | DOI:10.1097/ACI.0000000000001077

Comput Methods Programs Biomed. 2025 Apr 12;266:108753. doi: 10.1016/j.cmpb.2025.108753. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: Due to the presence of charge traps in amorphous silicon flat-panel detectors, lag signals are generated in consecutively captured projections. These signals lead to ghosting in projection images and severe lag artifacts in cone-beam computed tomography (CBCT) reconstructions. Traditional Linear Time-Invariant (LTI) correction need to measure lag correction factors (LCF) and may leave residual lag artifacts. This incomplete correction is partly attributed to the lack of consideration for exposure dependency.

METHODS: To measure the lag signals more accurately and suppress lag artifacts, we develop a novel hardware correction method. This method requires two scans of the same object, with adjustments to the operating timing of the CT instrumentation during the second scan to measure the lag signal from the first. While this hardware correction significantly mitigates lag artifacts, it is complex to implement and imposes high demands on the CT instrumentation. To enhance the process, We introduce a deep learning method called Lag-Net to remove lag signal, utilizing the nearly lag-free results from hardware correction as training targets for the network.

RESULTS: Qualitative and quantitative analyses of experimental results on both simulated and real datasets demonstrate that deep learning correction significantly outperforms traditional LTI correction in terms of lag artifact suppression and image quality enhancement. Furthermore, the deep learning method achieves reconstruction results comparable to those obtained from hardware correction while avoiding the operational complexities associated with the hardware correction approach.

CONCLUSION: The proposed hardware correction method, despite its operational complexity, demonstrates superior artifact suppression performance compared to the LTI algorithm, particularly under low-exposure conditions. The introduced Lag-Net, which utilizes the results of the hardware correction method as training targets, leverages the end-to-end nature of deep learning to circumvent the intricate operational drawbacks associated with hardware correction. Furthermore, the network's correction efficacy surpasses that of the LTI algorithm in low-exposure scenarios.

PMID:40233441 | DOI:10.1016/j.cmpb.2025.108753

PLOS Digit Health. 2025 Apr 15;4(4):e0000792. doi: 10.1371/journal.pdig.0000792. eCollection 2025 Apr.

ABSTRACT

Recent advancements in deep learning have shown promise in enhancing the performance of medical image analysis. In pathology, automated whole slide imaging has transformed clinical workflows by streamlining routine tasks and diagnostic and prognostic support. However, the lack of transparency of deep learning models, often described as black boxes, poses a significant barrier to their clinical adoption. This study evaluates various explainability methods for Vision Transformers, assessing their effectiveness in explaining the rationale behind their classification predictions on histopathological images. Using a Vision Transformer trained on the publicly available CAMELYON16 dataset comprising of 399 whole slide images of lymph node metastases of patients with breast cancer, we conducted a comparative analysis of a diverse range of state-of-the-art techniques for generating explanations through heatmaps, including Attention Rollout, Integrated Gradients, RISE, and ViT-Shapley. Our findings reveal that Attention Rollout and Integrated Gradients are prone to artifacts, while RISE and particularly ViT-Shapley generate more reliable and interpretable heatmaps. ViT-Shapley also demonstrated faster runtime and superior performance in insertion and deletion metrics. These results suggest that integrating ViT-Shapley-based heatmaps into pathology reports could enhance trust and scalability in clinical workflows, facilitating the adoption of explainable artificial intelligence in pathology.

PMID:40233316 | DOI:10.1371/journal.pdig.0000792

Dentomaxillofac Radiol. 2025 Apr 15:twaf029. doi: 10.1093/dmfr/twaf029. Online ahead of print.

ABSTRACT

OBJECTIVES: Deep Learning-driven Super Resolution (SR) aims to enhance the quality and resolution of images, offering potential benefits in dental imaging. Although extensive research has focused on deep learning based dental classification tasks, the impact of applying super-resolution techniques on classification remains underexplored. This study seeks to address this gap by evaluating and comparing the performance of deep learning classification models on dental images with and without super-resolution enhancement.

METHODS: An open-source dental image dataset was utilized to investigate the impact of SR on image classification performance. SR was applied by two models with a scaling ratio of 2 and 4, while classification was performed by four deep learning models. Performances were evaluated by well-accepted metrics like SSIM, PSNR, accuracy, recall, precision, and F1-score. The effect of SR on classification performance is interpreted through two different approaches.

RESULTS: Two SR models yielded average SSIM and PSNR values of 0.904 and 36.71 for increasing resolution with two scaling ratios. Average accuracy and F-1 score for the classification trained and tested with two SR-generated images were 0.859 and 0.873. In the first of the comparisons carried out with two different approaches, it was observed that the accuracy increased in at least half of the cases (8 out of 16) when different models and scaling ratios were considered, while in the second approach, SR showed a significantly higher performance for almost all cases (12 out of 16).

CONCLUSION: This study demonstrated that the classification with SR-generated images significantly improved outcomes.

ADVANCES IN KNOWLEDGE: For the first time, the classification performance of dental radiographs with improved resolution by SR has been investigated. Significant performance improvement was observed compared to the case without SR.

PMID:40233244 | DOI:10.1093/dmfr/twaf029

PLoS Comput Biol. 2025 Apr 15;21(4):e1012964. doi: 10.1371/journal.pcbi.1012964. Online ahead of print.

ABSTRACT

A generalizable computational platform, CTRL-V (Computational TRacking of Likely Variants), is introduced to design selective binding (dual bait) biosensor proteins. The iteratively evolving receptor binding domain (RBD) of SARS-CoV-2 spike protein has been construed as a model dual bait biosensor which has iteratively evolved to distinguish and selectively bind to human entry receptors and avoid binding neutralizing antibodies. Spike RBD prioritizes mutations that reduce antibody binding while enhancing/ retaining binding with the ACE2 receptor. CTRL-V's through iterative design cycles was shown to pinpoint 20% (of the 39) reported SARS-CoV-2 point mutations across 30 circulating, infective strains as responsible for immune escape from commercial antibody LY-CoV1404. CTRL-V successfully identifies ~70% (five out of seven) single point mutations (371F, 373P, 440K, 445H, 456L) in the latest circulating KP.2 variant and offers detailed structural insights to the escape mechanism. While other data-driven viral escape variant predictor tools have shown promise in predicting potential future viral variants, they require massive amounts of data to bypass the need for physics of explicit biochemical interactions. Consequently, they cannot be generalized for other protein design applications. The publicly availably viral escape data was leveraged as in vivo anchors to streamline a computational workflow that can be generalized for dual bait biosensor design tasks as exemplified by identifying key mutational loci in Raf kinase that enables it to selectively bind Ras and Rap1a GTP. We demonstrate three versions of CTRL-V which use a combination of integer optimization, stochastic sampling by PyRosetta, and deep learning-based ProteinMPNN for structure-guided biosensor design.

PMID:40233103 | DOI:10.1371/journal.pcbi.1012964

Bioinformatics. 2025 Apr 15:btaf170. doi: 10.1093/bioinformatics/btaf170. Online ahead of print.

ABSTRACT

MOTIVATION: Understanding the protein sequence-function relationship is essential for advancing protein biology and engineering. However, less than 1% of known protein sequences have human-verified functions. While deep learning methods have demonstrated promise for protein function prediction, current models are limited to predicting only those functions on which they were trained.

RESULTS: Here, we introduce ProtNote, a multimodal deep learning model that leverages free-form text to enable both supervised and zero-shot protein function prediction. ProtNote not only maintains near state-of-the-art performance for annotations in its training set, but also generalizes to unseen and novel functions in zero-shot test settings. ProtNote demonstrates superior performance in prediction of novel GO annotations and EC numbers compared to baseline models by capturing nuanced sequence-function relationships that unlock a range of biological use cases inaccessible to prior models. We envision that ProtNote will enhance protein function discovery by enabling scientists to use free text inputs without restriction to predefined labels - a necessary capability for navigating the dynamic landscape of protein biology.

AVAILABILITY AND IMPLEMENTATION: The code is available on GitHub: https://github.com/microsoft/protnote; model weights, datasets, and evaluation metrics are provided via Zenodo: https://zenodo.org/records/13897920.

SUPPLEMENTARY INFORMATION: Supplementary Information is available at Bioinformatics online.

PMID:40233101 | DOI:10.1093/bioinformatics/btaf170

STAR Protoc. 2025 Apr 14;6(2):103768. doi: 10.1016/j.xpro.2025.103768. Online ahead of print.

ABSTRACT

Here, we present a protocol for using SIMS (scalable, interpretable machine learning for single cell) to transfer cell type labels in single-cell RNA sequencing data. This protocol outlines data preparation, model training with labeled data or inference using pretrained models, and methods for visualizing, downloading, and interpreting predictions. We provide stepwise instructions for accessing SIMS through the application programming interface (API), GitHub Codespaces, and a web application. For complete details on the use and execution of this protocol, please refer to Gonzalez-Ferrer et al.1.

PMID:40232935 | DOI:10.1016/j.xpro.2025.103768