Front Plant Sci. 2025 Jul 21;16:1645490. doi: 10.3389/fpls.2025.1645490. eCollection 2025.

ABSTRACT

INTRODUCTION: Microplastics (MPs), as emerging environmental contaminants, pose a significant threat to global food security. In order to rapidly screen and diagnosis rice seedling under MPs stress at an early stage, it is essential to develop efficient and non-destructive detection methods.

METHODS: In this study, rice seedlings exposed to different concentrations (0, 10, and 100 mg/L) of polyethylene terephthalate (PET), polystyrene (PS), and polyvinyl chloride (PVC) MPs stress were constructed. Two complementary spectroscopic techniques, visible/near-infrared hyperspectral imaging (VNIR-HSI) and synchrotron radiation-based Fourier Transform Infrared spectroscopy (SR-FTIR), were employed to capture the biochemical changes of leaf organic molecules.

RESULTS: The spectral information of rice seedlings under MPs stress was obtained by using VNIR-HSI, and the low-dimensional clustering distribution analysis of the original spectra was conducted. An improved SE-LSTM full-spectral detection model was proposed, and the detection accuracy rate was greater than 93.88%. Characteristic wavelengths were extracted to build a simplified detection model, and the SHapley Additive exPlanations (SHAP) framework was applied to interpret the model by identifying the bands associated with chlorophyll, carotenoids, water content, and cellulose. Meanwhile, SR-FTIR spectroscopy was used to investigate compositional changes in both leaf lamina and veins, and two-dimensional correlation spectroscopy (2DCOS) was employed to reveal the sequential interactions among molecular components.

DISCUSSION: In conclusion, the combination of spectral technology and deep learning to capture the physiological and biochemical reactions of leaves could provide a rapid and interpretable method for detecting rice seedlings under MPs stress. This method could provide a solution for the early detection of external stress on other crops.

PMID:40761567 | PMC:PMC12318996 | DOI:10.3389/fpls.2025.1645490

Front Plant Sci. 2025 Jul 21;16:1640806. doi: 10.3389/fpls.2025.1640806. eCollection 2025.

ABSTRACT

INTRODUCTION: Accurate wheat yield estimation is crucial for efficient crop management. This study introduces the Spatio-Temporal Fusion Mixture of Experts (STF-MoE) model, an innovative deep learning framework built upon an LSTM-Transformer architecture.

METHODS: The STF-MoE model incorporates a heterogeneous Mixture of Experts (MoE) mechanism with an adaptive gating network. This design dynamically processes fused multi-source remote sensing features (e.g., near-infrared vegetation reflectance, NIRv; fraction of photosynthetically active radiation absorption, Fpar) and environmental variables (e.g., relative humidity, digital elevation model) across multiple expert networks. The model was applied to estimate wheat yield in six major Chinese provinces.

RESULTS: The STF-MoE model demonstrated exceptional accuracy in the most recent estimation year (R² = 0.827, RMSE = 547.7 kg/ha) and exhibited robust performance across historical years and extreme climatic events, outperforming baseline models. Relative humidity and digital elevation model were identified as the most critical yield-influencing factors. Furthermore, the model accurately estimated yield 1-2 months before harvest by identifying key phenological stages (March to June).

DISCUSSION: STF-MoE effectively handles multi-source spatiotemporal complexity via its dynamic gating and expert specialization. While underestimation persists in extreme-yield regions, the model provides a scalable solution for pre-harvest yield estimation. Future work will optimize computational efficiency and integrate higher-resolution data.

PMID:40761564 | PMC:PMC12318938 | DOI:10.3389/fpls.2025.1640806

Front Plant Sci. 2025 Jul 21;16:1583526. doi: 10.3389/fpls.2025.1583526. eCollection 2025.

ABSTRACT

Pod numbers are important for assessing soybean yield. How to simplify the traditional manual process and determine the pod number phenotype of soybean maturity more quickly and accurately is an urgent challenge for breeders. With the development of smart agriculture, numerous scientists have explored the phenotypic information related to soybean pod number and proposed corresponding methods. However, these methods mainly focus on the total number of pods, ignoring the differences between different pod types and do not consider the time-consuming and labor-intensive problem of picking pods from the whole plant. In this study, a deep learning approach was used to directly detect the number of different types of pods on non-disassembled plants at the maturity stage of soybean. Subsequently, the number of pods wascorrected by means of a metric learning method, thereby improving the accuracy of counting different types of pods. After 200 epochs, the recognition results of various object detection algorithms were compared to obtain the optimal model. Among the algorithms, YOLOX exhibited the highest mean average precision (mAP) of 83.43% in accurately determining the counts of diverse pod categories within soybean plants. By improving the Siamese Network in metric learning, the optimal Siamese Network model was obtained. SE-ResNet50 was used as the feature extraction network, and its accuracy on the test set reached 93.7%. Through the Siamese Network model, the results of object detection were further corrected and counted. The correlation coefficients between the number of one-seed pods, the number of two-seed pods, the number of three-seed pods, the number of four-seed pods and the total number of pods extracted by the algorithm and the manual measurement results were 92.62%, 95.17%, 96.90%, 94.93%, 96.64%,respectively. Compared with the object detection algorithm, the recognition of soybean mature pods was greatly improved, evolving into a high-throughput and universally applicable method. The described results show that the proposed method is a robust measurement and counting algorithm, which can reduce labor intensity, improve efficiency and accelerate the process of soybean breeding.

PMID:40761559 | PMC:PMC12319039 | DOI:10.3389/fpls.2025.1583526

Data Brief. 2025 Jul 11;61:111867. doi: 10.1016/j.dib.2025.111867. eCollection 2025 Aug.

ABSTRACT

Beaches are recognized as major sinks of plastic litter and key sites where litter fragments into countless small pieces. Because those fine particles are almost impossible to remove from the natural environment, it is essential to monitor macroplastic litter on beaches before they degrade. To observe the distribution of this litter in detail, it is essential to have automated and objective image-processing methods that can be applied to images captured by remote sensing devices, such as web cameras and drones. To develop such an automated analysis method, a deep learning-based approach has recently become mainstream, and clarifying technical issues based on case studies is vital. The preparation of training data for those practices is critical but laborious. The BePLi Dataset v2 is updated from BePLi Dataset v1, comprises 3722 original images of beach plastic litter and 118,572 manually processed annotations. All original images were obtained from the natural coastal environment on the Northwest Japan coast, and annotation for plastic litter was provided at both the pixel and individual levels. The plastic litter objects are categorized into thirteen representative plastic object classes: "pet_bottle," "other_bottle," "plastic_bag," "box_shaped_case," "other_container," "rope," "other_string," "fishing_net," "buoy," "other_fishing_gear," "styrene_foam," "others" and "fragment." The BePLi Dataset v2 allows users to develop an instance segmentation and object detection method detecting macro beach plastic litter individually and at the pixel level. Depending on the user, this dataset can serve multiple purposes at different levels of technology development, from counting objects to estimating litter coverage, as it provides both bounding box- and pixel-based annotations.

PMID:40761540 | PMC:PMC12320089 | DOI:10.1016/j.dib.2025.111867

Bioimpacts. 2025 Jun 28;15:31226. doi: 10.34172/bi.31226. eCollection 2025.

ABSTRACT

INTRODUCTION: Identifying gene-disease associations is crucial for advancing medical research and improving clinical outcomes. Nevertheless, the rapid expansion of biomedical literature poses significant obstacles to extracting meaningful relationships from extensive text collections.

METHODS: This study uses deep learning techniques to automate this process, using publicly available datasets (EU-ADR, GAD, and SNPPhenA) to classify these associations accurately. Each dataset underwent rigorous pre-processing, including entity identification and preparation, word embedding using pre-trained Word2Vec and fastText models, and position embedding to capture semantic and contextual relationships within the text. In this research, three deep learning-based hybrid models have been implemented and contrasted, including CNN-LSTM, CNN-GRU, and CNN-GRU-LSTM. Each model has been equipped with attentional mechanisms to enhance its performance.

RESULTS: Our findings reveal that the CNN-GRU model achieved the highest accuracy of 91.23% on the SNPPhenA dataset, while the CNN-GRU-LSTM model attained an accuracy of 90.14% on the EU-ADR dataset. Meanwhile, the CNN-LSTM model demonstrated superior performance on the GAD dataset, achieving an accuracy of 84.90%. Compared to previous state-of-the-art methods, such as BioBERT-based models, our hybrid approach demonstrates superior classification performance by effectively capturing local and sequential features without relying on heavy pre-training.

CONCLUSION: The developed models and their evaluation data are available at https://github.com/NoorFadhil/Deep-GDAE.

PMID:40761527 | PMC:PMC12319213 | DOI:10.34172/bi.31226

Cogn Neurodyn. 2025 Dec;19(1):123. doi: 10.1007/s11571-025-10304-3. Epub 2025 Aug 2.

ABSTRACT

Emotion recognition is a difficult problem mainly because emotions are presented in different modalities including; speech, face, and text. In light of this, in this paper, we introduce a novel framework known as Audio, Visual, and Text Emotions Fusion Network that will enhance the approaches to analyzing emotions that can incorporate these dissimilar types of inputs efficiently for the enhancement of the existing approaches to analyzing emotions. Using specialized techniques, each modality in this framework shows Graph Attention Network-based Transformer Network by employing Graph Attention Networks to detect dependencies in facial regions; Hybrid Wav2Vec 2.0 and Convolutional Neural Network combines Wav2Vec 2.0, and Convolutional Neural Network to extract informative temporal and frequency domain audio features. Contextual and sequential text semantics are captured by Bidirectional Encoder Representations from Transformers with Bidirectional Gated Recurrent Unit. They are fused based on a novel attention-based mechanism that distributes weights depending on the emotional context and improves cross-modal interactions. Moreover, the Audio, Visual, and Text Emotions Fusion Network system effectively identifies emotions, and the result section that contains overall accuracy at 98.7%, precision at 98.2%, recall, at 97.2%, and F1-score of 97.49% makes the proposed approach strong and efficient for real-time emotion recognition strategies.

PMID:40761311 | PMC:PMC12317966 | DOI:10.1007/s11571-025-10304-3

Front Oncol. 2025 Jul 21;15:1555585. doi: 10.3389/fonc.2025.1555585. eCollection 2025.

ABSTRACT

Ovarian cancer remains one of the most lethal gynecological malignancies, posing significant challenges for early detection due to its asymptomatic nature in early stages. Accurate segmentation of ovarian tumors from ultrasound images is critical for improving diagnostic accuracy and patient outcomes. In this study, we introduce SMoFFI-SegFormer, an advanced deep learning model specifically designed to enhance multi-scale feature representation and address the complexities of ovarian tumor segmentation. Building upon the SegFormer architecture, SMoFFI-SegFormer incorporates a novel Self-modulate Fusion with Feature Inhibition (SMoFFI) module that promotes cross-scale information exchange and effectively handles spatial heterogeneity within tumors. Through extensive experimentation on two public datasets-OTU_2D and OTU_CEUS-our model demonstrates superior performance with high overall accuracy, mean Intersection over Union (mIoU), and class accuracy. Specifically, SMoFFI-SegFormer achieves state-of-the-art results, significantly outperforming existing models in both segmentation precision and efficiency. This work paves the way for more reliable and automated tools in the diagnosis and management of ovarian cancer.

PMID:40761240 | PMC:PMC12320497 | DOI:10.3389/fonc.2025.1555585

Front Immunol. 2025 Jul 21;16:1597676. doi: 10.3389/fimmu.2025.1597676. eCollection 2025.

ABSTRACT

BACKGROUND: Sepsis is a life-threatening condition with limited therapeutic options. Emerging evidence implicates gut microbial metabolites in modulating host immunity, but the specific interactions between these metabolites and host druggable targets remain poorly understood.

METHODS: We utilized a systems biology framework integrating genetic analyses, multi-omics profiling, and structure-based virtual screening to systematically map the interaction landscape between human gut microbial metabolites and druggable G-protein-coupled receptors (GPCRs), ion channels (ICs), and kinases (termed the "GIKome") in sepsis. Key findings were validated by molecular dynamics (MD) simulation, microscale thermophoresis (MST), and functional assays in a murine cecal ligation and puncture (CLP) model of sepsis.

RESULTS: We evaluated 190,950 metabolite-protein interactions, linking 114 sepsis-related GIK targets to 335 gut microbial metabolites, and prioritized indole-3-lactic acid (ILA), a metabolite enriched in Akkermansia muciniphila, as a promising therapeutic candidate. MD simulation and MST further revealed that ILA binds stably to PFKFB2, a pivotal kinase in regulating glycolytic flux and immune activation during sepsis. In vivo, ILA administration improved survival, attenuated cytokine storm, and mitigated multi-organ injury in CLP-induced septic mice.

CONCLUSIONS: This systems-level investigation unveils previously unrecognized therapeutic targets, offering a blueprint for microbiota-based precision interventions in critical care medicine.

PMID:40761792 | PMC:PMC12318984 | DOI:10.3389/fimmu.2025.1597676

Front Pharmacol. 2025 Jul 21;16:1629941. doi: 10.3389/fphar.2025.1629941. eCollection 2025.

ABSTRACT

INTRODUCTION: In the context of epidermal inflammation, the inflammatory response not only involves the release of inflammatory cytokines like interleukin 8 (IL-8), but also modulation of tight junction protein expression levels. Previous studies showed that the tight junction protein claudin 1 (CLDN1) is upregulated during tumor necrosis factor α (TNFα)-induced inflammation by capsaicin in keratinocytes in a transient receptor potential channel vanilloid 1 (TRPV1)-dependent manner. However, the caveat with TRPV1 ligands is the undesired pain response elicited by the activation of neuronal TRPV1 channels. In this study, we hypothesized that also less or non-pungent homovanillic acid esters as structural analogs of capsaicin target CLDN1 upregulation during inflammation.

METHODS: We aimed to identify beneficial structural characteristics by selecting homovanillic acid esters with different aliphatic tail structures and screening them for CLDN1 upregulation at early stages of TNFα-induced inflammation in basal keratinocytes.

RESULTS: CLDN1 expression was upregulated independently of TRPV1 by compounds with a tail of 5 or 6 C-atoms, regardless of the presence of ramifications and double bonds with a maximum fold change of 2.05 ± 0.22 against control. The induction of CLDN1 expression was accompanied by increased expression of the differentiation marker involucrin (IVL).

DISCUSSION: The results suggest that the homovanillic ester-induced CLDN1 upregulation is a result of increased differentiation of the basal keratinocytes towards the keratinocyte morphology present in the stratum granulosum (SG), where tight junctions are formed. In conclusion, homovanillic acid esters with a 5 or 6 C-atom long aliphatic chain induced CLDN1 expression, thereby stimulating keratinocyte differentiation, independent from TRPV1 activation.

PMID:40761393 | PMC:PMC12319341 | DOI:10.3389/fphar.2025.1629941

Cureus. 2025 Jul 4;17(7):e87304. doi: 10.7759/cureus.87304. eCollection 2025 Jul.

ABSTRACT

Nonsteroidal anti-inflammatory drugs (NSAIDs) are extensively utilized for their analgesic and anti-inflammatory efficacy, yet they pose a significant risk for renal adverse events, notably drug-induced acute interstitial nephritis (DI-AIN). Prompt recognition and appropriate management are paramount to prevent irreversible kidney damage. We present the case of a 46-year-old male with NSAID-induced DI-AIN, emphasizing the diagnostic utility of a specific urinary biomarker profile and the rationale for empirical steroid therapy initiated before histopathological confirmation. Our patient developed acute kidney injury (AKI) following daily ibuprofen administration for persistent fever. Despite ibuprofen discontinuation on day eight, renal function failed to improve, necessitating hospital admission on day 14. On day 15, his serum creatinine (Cr) level was 1.86 mg/dL. Urinalysis revealed mild proteinuria [urine protein-to-creatinine ratio (UPCR): 0.24 g/gCr] but strikingly elevated urinary tubular injury markers: N-acetyl-β-D-glucosaminidase (NAG): 20.9 U/L (on day one), β2-microglobulin (β2MG): 6028 μg/L, and L-type fatty acid-binding protein (L-FABP): 27.75 ng/mL. Based on a strong clinical suspicion of DI-AIN, a kidney biopsy was performed on day 15, and oral prednisolone (PSL, 0.8 mg/kg/day) was commenced the same evening before biopsy results were available. Serum creatinine improved to 1.56 mg/dL by discharge on day 23. Post-discharge, kidney biopsy results confirmed AIN. PSL was gradually tapered and discontinued after approximately 10 months, with sustained renal function recovery (serum creatinine: ~1.1 mg/dL). This report underscores the importance of suspecting DI-AIN in patients with AKI and a history of NSAID exposure. The pronounced elevation of urinary tubular markers, despite only mild proteinuria, was pivotal in raising clinical suspicion. The negative autoimmune serology further strengthened the diagnosis of a drug-induced etiology. Empirical steroid therapy, initiated due to compelling clinical evidence before histopathological confirmation, appeared to be an effective intervention. While this single case cannot establish a therapeutic standard, it illustrates a clinical scenario where early, empirically-guided treatment may be justified. Kidney biopsy remains indispensable for definitive diagnosis. The report also highlights the pressing need for enhanced patient education on appropriate NSAID utilization. Repeated use of common NSAIDs can precipitate DI-AIN. A diagnostic profile of elevated urinary tubular markers with only mild proteinuria can be a key indicator for suspecting this condition. Empirical steroid therapy, guided by strong clinical suspicion, can be an effective early intervention, with subsequent kidney biopsy providing definitive diagnostic validation. Enhanced patient education on appropriate NSAID use is essential.

PMID:40761978 | PMC:PMC12319171 | DOI:10.7759/cureus.87304

Surg Case Rep. 2025;11(1):25-0227. doi: 10.70352/scrj.cr.25-0227. Epub 2025 Jul 31.

ABSTRACT

INTRODUCTION: Tamoxifen (TAM) is a well-established treatment for hormone receptor-positive breast cancer with a known side-effect profile that includes hot flashes, genital bleeding, and diarrhea (0.1%-5%). Other notable side effects include liver damage, abnormal vaginal discharge, depression, dizziness, and headaches of unknown frequency. However, blood cell count reduction has not yet been reported as a side effect in Japan.

CASE PRESENTATION: A 46-year-old female patient was diagnosed with right breast cancer (cT1N0M0). The patient underwent partial right breast resection and sentinel lymph node biopsy. Owing to the positive surgical resection margin, a mastectomy was performed. Pathological analysis of the surgical specimen confirmed invasive ductal carcinoma (estrogen receptor [ER]: 95%, progesterone receptor [PgR]: 85%, HER2: 2+ [fluorescence in situ hybridization, FISH negative]), with macrometastasis in one sentinel lymph node. Postoperative treatment included chemotherapy (dose-dense adriamycin and cyclophosphamide [AC] to dose-dense paclitaxel [PTX]), irradiation, and TAM. While initial blood test results before starting TAM showed mild anemia (Hb: 8.9 g/dL Grade 2), a follow-up blood test 5 months after initiating TAM revealed a significant decrease in blood cell counts (white blood cell [WBC]: 2600/μL Grade 2, neutrophil [neu]: 0.55 × 10³/μL Grade 3, Hb: 7.7 g/dL Grade 2, platelet [PLT]: 13.3 × 10⁴/μL). Considering the onset of symptoms following TAM administration, drug-induced pancytopenia was suspected. TAM and its concomitant medication pregabalin were discontinued. However, the blood cell counts continued to decline, necessitating further investigation. Myelodysplastic syndrome (MDS) was suspected, leading to multiple bone marrow biopsies. However, no definitive hematological disorder was diagnosed. The patient received transfusions and granulocyte colony-stimulating factor (G-CSF) injections based on the blood cell count. Approximately 4 months after the onset of neutropenia, gradual recovery was observed and spontaneous remission occurred. Given the rarity of spontaneous recovery from MDS, TAM is considered a potential causative agent of the observed decline in blood cell counts.

CONCLUSIONS: We report a case of suspected drug-induced cytopenia associated with tamoxifen administration.

PMID:40761476 | PMC:PMC12319564 | DOI:10.70352/scrj.cr.25-0227

Medicine (Baltimore). 2025 Aug 1;104(31):e43615. doi: 10.1097/MD.0000000000043615.

ABSTRACT

BACKGROUND: The safety and efficacy of esaxerenone (ESAX), a novel nonsteroidal mineralocorticoid receptor antagonist, remain insufficiently explored in systematic reviews and meta-analyses (SR/MA). This SR/MA aimed to investigate the safety and effects of ESAX on blood pressure (BP) and renal function.

METHODS: Multiple databases and registers were systematically searched to identify randomized controlled trials and real-world studies evaluating the safety and efficacy of ESAX in various conditions. The primary outcome was the risk of adverse events (AEs); secondary outcomes included its effects on BP and renal parameters.

RESULTS: This SR/MA included 22 studies (N = 4699); 6 studies (5 randomized controlled trials and one retrospective study; n = 3211) with comparator groups were meta-analyzed. While more subjects on ESAX, especially at higher doses, experienced drug-related AEs (risk ratio [RR] 1.77) and discontinued due to these AEs (RR 6.75) compared to placebo, the number of subjects with any or serious AEs and drug-related serious AEs was similar between the 2 groups. Higher doses of ESAX were associated with increased risks of rising serum potassium levels (RR 3.30) and drug discontinuation related to these increases (RR 5.71) compared to the placebo. ESAX and active comparators exhibited comparable AEs except for a higher risk (RR 2.87) of increasing serum potassium levels with ESAX. ESAX led to larger decreases in estimated glomerular filtration rate and urine albumin-creatinine ratio than placebo. ESAX was more effective than placebo and active comparators in lowering office systolic and diastolic BP. ESAX 5 mg showed greater 24-hour average ambulatory BP reductions compared to the active comparators.

CONCLUSION: ESAX appears reasonably safe, with a modest risk of hyperkalemia and worsening of renal function, and modest efficacy in the treatment of hypertension and albuminuria.

PMID:40760566 | DOI:10.1097/MD.0000000000043615

Support Care Cancer. 2025 Aug 4;33(8):752. doi: 10.1007/s00520-025-09811-0.

ABSTRACT

PURPOSE: Over recent decades, advancements in cancer therapies have led to improved survival, resulting in a growing number of cancer patients admitted to the intensive care unit (ICU). While respiratory and hemodynamic failures have been extensively studied, less attention has been given to acute neurological dysfunction (ND) in oncology patients.

METHODS: This single-center retrospective study (2007-2020) aimed to assess the characteristics and outcomes of solid tumor patients admitted to the ICU for acute ND, distributed into cancer-related, oncology treatment-related, and non-specific causes.

RESULTS: 1845 patients with solid tumor were admitted to the ICU, including 164 (8.9%) for acute ND. Lung (21%), urinary tract (21%) and digestive tract (20%) were the main tumor sites with mostly metastatic disease (58%). Main causes of admission were seizures with or without status epilepticus (33%), coma (27%), and delirium (27%). ND was predominantly cancer-related (36%) or due to independent complications (56%), while 8% had treatment-related ND. Coma, motor deficit, and a shorter time from symptoms onset to admission were more frequent in cancer-related ND patients. Most patients with treatment-related ND had normal brain imaging (67%) with signs of encephalopathy on electroencephalogram (86%). ICU mortality rate was 18%, with coma at admission being the only significant predictor of mortality (OR 5.96 [2.06-20.4]).

CONCLUSIONS: One in ten cancer patients admitted to the ICU presents with acute ND. Among them, one-third are ultimately diagnosed with cancer-related ND, characterized by a short time from symptoms onset to ICU admission and the presence of coma or motor deficit at ICU admission. This condition does not negatively impact ICU survival.

PMID:40760226 | DOI:10.1007/s00520-025-09811-0

Sci Rep. 2025 Aug 4;15(1):28472. doi: 10.1038/s41598-025-06848-3.

ABSTRACT

This study addresses the optimization of fully convolutional networks (FCNs) for deployment on resource-limited devices in real-time scenarios. While prior research has extensively applied quantization techniques to architectures like VGG-16, there is limited exploration of comprehensive layer-wise quantization specifically within the FCN-8 architecture. To fill this gap, we propose an innovative approach utilizing full-layer quantization with an [Formula: see text] error minimization algorithm, accompanied by sensitivity analysis to optimize fixed-point representation of network weights. Our results demonstrate that this method significantly enhances sparsity, achieving up to 40%, while preserving performance, yielding an impressive 89.3% pixel accuracy under extreme quantization conditions. The findings highlight the efficacy of full-layer quantization and retraining in simultaneously reducing network complexity and maintaining accuracy in both image classification and semantic segmentation tasks.

PMID:40759661 | DOI:10.1038/s41598-025-06848-3

JMIR Med Inform. 2025 Aug 4. doi: 10.2196/73995. Online ahead of print.

ABSTRACT

This Perspective article explores the transformative role of artificial intelligence (AI) in predicting perioperative hypoxemia through the integration of deep learning (DL) with multimodal clinical data, including lung imaging, pulmonary function tests (PFTs), and arterial blood gas (ABG) analysis. Perioperative hypoxemia, defined as arterial oxygen partial pressure (PaO₂) <60 mmHg or oxygen saturation (SpO₂) <90%, poses significant risks of delayed recovery and organ dysfunction. Traditional diagnostic methods, such as radiological imaging and ABG analysis, often lack integrated predictive accuracy. AI frameworks, particularly convolutional neural networks (CNNs) and hybrid models like TD-CNNLSTM-LungNet, demonstrate exceptional performance in detecting pulmonary inflammation and stratifying hypoxemia risk, achieving up to 96.57% accuracy in pneumonia subtype differentiation and an AUC of 0.96 for postoperative hypoxemia prediction. Multimodal AI systems, such as DeepLung-Predict, unify CT scans, PFTs, and ABG parameters to enhance predictive precision, surpassing conventional methods by 22%. However, challenges persist, including dataset heterogeneity, model interpretability, and clinical workflow integration. Future directions emphasize multicenter validation, explainable AI (XAI) frameworks, and pragmatic trials to ensure equitable and reliable deployment. This AI-driven approach not only optimizes resource allocation but also mitigates financial burdens on healthcare systems by enabling early interventions and reducing ICU admission risks.

PMID:40759599 | DOI:10.2196/73995

J Exp Clin Cancer Res. 2025 Aug 4;44(1):225. doi: 10.1186/s13046-025-03478-5.

ABSTRACT

BACKGROUND: The metastatic microenvironment is often rich in tumor-associated macrophages (TAMs). In uveal melanoma (UM), high levels of TAMs positively correlate with tumor progression and poorer prognosis. We hypothesize that the immunomodulation of TAMs can remodel the UM tumor microenvironment and make it more susceptible to therapeutic interventions.

METHODS: In our work, we designed a novel computational pipeline that combines single-cell transcriptomics data, network analysis, multicriteria decision techniques, and pharmacophore-based docking simulations to select molecular targets and matching repurposable drugs for TAM immunomodulation. The method generates a ranking of drug-target interactions, the most promising of which are channeled towards experimental validation.

RESULTS: To identify potential immunomodulatory targets, we created a network-based representation of the TAM interactome and extracted a regulatory core conditioned on UM expression data. Further, we selected 13 genes from this core (NLRP3, HMOX1, CASP1, GSTP1, NAMPT, HSP90AA1, B2M, ISG15, LTA4H, PTGS2, CXCL2, PLAUR, ZFP36, TANK) for pharmacophore-based virtual screening of FDA-approved compounds, followed by flexible molecular docking. Based on the ranked docking results, we chose the interaction between caspase-1 and clindamycin for experimental validation. Functional studies on macrophages confirmed that clindamycin inhibits caspase-1 activity and thereby inflammasome activation, leading to a decrease in IL-1β, IL-18, and gasdermin D cleavage products as well as a reduction in pyroptotic cell death. This clindamycin-mediated inhibition of caspase-1 was also observable in TAMs derived from the bone marrow of multiple myeloma patients.

CONCLUSIONS: Our computational workflow for drug repurposing identified clindamycin as an efficacious inhibitor of caspase-1 that suppresses inflammasome activity and pyroptosis in vitro in TAMs.

PMID:40759978 | DOI:10.1186/s13046-025-03478-5

Inflammopharmacology. 2025 Aug 4. doi: 10.1007/s10787-025-01882-z. Online ahead of print.

ABSTRACT

This article discusses the lengthy review by Pedrag Sikiric and twenty one (21) co-authors in Inflammopharmacology (2024) 32:3119-3161.

PMID:40759852 | DOI:10.1007/s10787-025-01882-z

J Biol Chem. 2025 Aug 2:110559. doi: 10.1016/j.jbc.2025.110559. Online ahead of print.

ABSTRACT

Guanylate cyclase 2C (GCC) upon binding to the bacterial heat-stable enterotoxin ST, generates excessive cGMP, driving intestinal chloride and fluid secretion that manifests as diarrhea. We investigated the regulatory mechanism of GCC through its interactions with Scaffolding proteins sodium-hydrogen exchanger regulatory factor (NHERF)1-4. PSD95, Dlg1, ZO-1 (PDZ) domain in NHERF4 inhibited GCC catalytic activity while NHERF1-3 binary binding had no impact. NHERF4-mediated inhibition was mimicked by two synergistically acting peptides, (N4-110 [NH2-LERPRFCLL-COOH] and N4-195 [NH2-RHAHDVARAQLG-COOH]), localized in close proximity within the PDZ1 domain. These peptides, which show high sequence homology to the GCC catalytic domain, were mapped via 3-D structural modeling to the GCC dimer interface. Fluorescence resonance energy transfer (FRET) analysis confirmed that NHERF4-PDZ1 domain binding interfered with GCC oligomerization. In mouse and human enteroid models, NHERF4 peptides dose-dependently reduced GCC-mediated fluid secretion. Additionally, NHERF4-GCC interaction was enhanced upon ST stimulation, suggesting NHERF4 functions as a negative regulator of aberrant GCC activity during enterotoxin-induced diarrhea. Furthermore, we described macromolecular complex of GCC with multi-drug resistance protein 4 (MRP4), a cAMP/cGMP efflux transporter, in the regulation of fluid secretion through NHERF3 mediated assembly. Overall, our findings reveal novel regulatory mechanisms for GCC, offering insights into targeted therapies for enterotoxin-triggered diarrheas.

PMID:40759370 | DOI:10.1016/j.jbc.2025.110559

Sci Rep. 2025 Aug 4;15(1):28378. doi: 10.1038/s41598-025-11445-5.

ABSTRACT

Prostate cancer (PCa) is the most prevalent malignant tumor in males, and many patients remain at risk of biochemical recurrence (BCR) following initial treatment. Accurate prediction of BCR is vital for effective clinical management and treatment planning. This study evaluates the effectiveness of machine learning (ML) models in predicting BCR among prostate cancer patients, comparing their performance to traditional prognostic methods. We systematically searched four databases (PubMed, Web of Science, Embase, and Cochrane) for studies employing ML techniques to predict prostate cancer BCR. Data extraction included model type, sample size, and the area under the curve (AUC). A meta-analysis was conducted using AUC as the primary performance metric to assess predictive accuracy and heterogeneity across models. Sixteen studies comprising a total of 17,316 prostate cancer patients were included. The pooled AUC for ML models was 0.82 (95% CI: 0.81-0.84). Deep learning and hybrid models outperformed traditional models (AUC = 0.83). Models using imaging data showed improved performance (AUC = 0.82). ML models were most effective in predicting 1-year BCR (AUC = 0.86), with performance slightly decreasing for longer time intervals. ML models outperform traditional methods in predicting BCR, especially in the short term. Incorporating multimodal data, such as imaging, enhances predictive accuracy. Future studies should optimize and validate these models through large-scale clinical trials.

PMID:40760134 | DOI:10.1038/s41598-025-11445-5

Sci Rep. 2025 Aug 4;15(1):28467. doi: 10.1038/s41598-025-97490-6.

ABSTRACT

This paper investigates the feasibility of multi-task learning (MTL) for facial recognition on the Raspberry Pi, a low-cost single-board computer, demonstrating its ability to perform complex deep learning tasks in real time. Using MobileNet, MobileNetV2, and InceptionV3 as base models, we trained MTL models on a custom database derived from the VGGFace2 dataset, focusing on three tasks: person identification, age estimation, and ethnicity prediction. MobileNet achieved the highest accuracy, with 99% in person identification, 99.3% in age estimation, and 99.5% in ethnicity prediction. Compared to previous studies, which primarily relied on high-end hardware for MTL in facial recognition, this work uniquely demonstrates the successful deployment of efficient MTL models on resource-constrained devices like the Raspberry Pi. This advancement significantly reduces computational load and energy consumption while maintaining high accuracy, making facial recognition systems more accessible and practical for real-world applications such as security, personalized customer experiences, and demographic analytics. This study opens new avenues for innovation in resource-efficient deep learning systems.

PMID:40760089 | DOI:10.1038/s41598-025-97490-6