Clin Exp Metastasis. 2025 May 15;42(3):30. doi: 10.1007/s10585-025-10349-y.

ABSTRACT

Local recurrence and distant metastasis were a common manifestation of locoregionally advanced nasopharyngeal carcinoma (LA-NPC) after neoadjuvant chemoradiotherapy (NACT). To validate the clinical value of MRI radiomic models based on deep learning for predicting the recurrence of LA-NPC patients. A total of 328 NPC patients from four hospitals were retrospectively included and divided into the training(n = 229) and validation (n = 99) cohorts randomly. Extracting 975 traditional radiomic features and 1000 deep radiomic features from contrast enhanced T1-weighted (T1WI + C) and T2-weighted (T2WI) sequences, respectively. Least absolute shrinkage and selection operator (LASSO) was applied for feature selection. Five machine learning classifiers were conducted to develop three models for LA-NPC prediction in training cohort, namely Model I: traditional radiomic features, Model II: combined the deep radiomic features with Model I, and Model III: combined Model II with clinical features. The predictive performance of these models were evaluated by receive operating characteristic (ROC) curve analysis, area under the curve (AUC), accuracy, sensitivity and specificity in both cohorts. The clinical characteristics in two cohorts showed no significant differences. Choosing 15 radiomic features and 6 deep radiomic features from T1WI + C. Choosing 9 radiomic features and 6 deep radiomic features from T2WI. In T2WI, the Model II based on Random forest (RF) (AUC = 0.87) performed best compared with other models in validation cohort. Traditional radiomic model combined with deep radiomic features shows excellent predictive performance. It could be used assist clinical doctors to predict curative effect for LA-NPC patients after NACT.

PMID:40369240 | DOI:10.1007/s10585-025-10349-y

Sci Rep. 2025 May 14;15(1):16815. doi: 10.1038/s41598-025-99460-4.

ABSTRACT

Knee osteoarthritis (KOA) is a severe arthrodial joint condition with significant global socioeconomic consequences. Early recognition and treatment of KOA is critical for avoiding disease progression and developing effective treatment programs. The prevailing method for knee joint analysis involves manual diagnosis, segmentation, and annotation to diagnose osteoarthritis (OA) in clinical practice while being highly laborious and a susceptible variable among users. To address the constraints of this method, several deep learning techniques, particularly the deep convolutional neural networks (CNNs), were applied to increase the efficiency of the proposed workflow. The main objective of this study is to create advanced deep learning (DL) approaches for risk assessment to forecast the evolution of pain for people suffering from KOA or those at risk of developing it. The suggested methodology applies a collective transfer learning approach for extracting accurate deep features using four pre-trained models, VGG19, ResNet50, AlexNet, and GoogleNet, to extract features from KOA images. The numeral of extracted features was reduced for identifying the most appropriate feature attributes for the disease. The binary Greylag Goose (bGGO) optimizer was employed to perform this task, with an average fitness of 0.4137 and a best fitness of 0.3155. The chosen features were categorized utilizing both deep learning and machine learning approaches. Finally, a CNN hyper-parameter algorithm was performed utilizing GGO. The suggested model outperformed previous models with accuracy, sensitivity, and specificity of 0.988692, 0.980156, and 0.990089, respectively. A comprehensive statistical analysis test was performed to confirm the validity of our findings.

PMID:40369219 | DOI:10.1038/s41598-025-99460-4

Sci Rep. 2025 May 14;15(1):16768. doi: 10.1038/s41598-025-91212-8.

ABSTRACT

Underwater computer vision faces significant challenges from light scattering, absorption, and poor illumination, which severely impact underwater vision tasks. To address these issues, ViT-Clarity, an underwater image enhancement module, is introduced, which integrates vision transformers with a convolutional neural network for superior performance. For comparison, ClarityNet, a transformer-free variant of the architecture, is presented to highlight the transformer's impact. Given the limited availability of paired underwater image datasets (clear and degraded), BlueStyleGAN is proposed as a generative model to create synthetic underwater images from clear in-air images by simulating realistic attenuation effects. BlueStyleGAN is evaluated against existing state-of-the-art synthetic dataset generators in terms of training stability and realism. Vit-ClarityNet is rigorously tested on five datasets representing diverse underwater conditions and compared with recent state-of-the-art methods as well as ClarityNet. Evaluations include qualitative and quantitative metrics such as UCIQM, UCIQE, and the deep learning-based URanker. Additionally, the impact of enhanced images on object detection and SIFT feature matching is assessed, demonstrating the practical benefits of image enhancement for underwater computer vision tasks.

PMID:40369132 | DOI:10.1038/s41598-025-91212-8

Sci Rep. 2025 May 14;15(1):16681. doi: 10.1038/s41598-025-01841-2.

ABSTRACT

In Wireless Sensor Networks (WSNs), achieving optimal coverage in dynamic environments remains a significant challenge. Traditional optimization techniques, such as genetic algorithms, particle swarm optimization, and ant colony optimization, have demonstrated adaptability in node placement but struggle with real-time self-learning capabilities, requiring frequent retraining to handle continuously changing conditions. To address these limitations, this research introduces a novel hybrid model that integrates Deep Reinforcement Learning (DRL) with Graph Neural Networks (GNN). The DRL component enables adaptive decision-making, allowing real-time sensor node adjustments based on network performance feedback. Simultaneously, the GNN model enhances spatial awareness by capturing relational dependencies among sensor nodes, optimizing coverage efficiency. This integration significantly improves network adaptability and operational efficiency. Extensive simulations demonstrate that the proposed DRL-GNN model achieves a coverage ratio of up to 96.4%, energy efficiency of 95.8%, and minimizes overlap to 5.2%, outperforming traditional methods. These results validate the effectiveness of the proposed approach in enhancing WSN coverage while maintaining energy efficiency and minimal redundancy.

PMID:40369115 | DOI:10.1038/s41598-025-01841-2

Sci Rep. 2025 May 14;15(1):16825. doi: 10.1038/s41598-025-99792-1.

ABSTRACT

The coexistence of chronic fibrosing idiopathic interstitial pneumonia (cf-IIP) and lung cancer (LC) in the same patient raises many doubts regarding patient prognosis and complicates decision-making in multidisciplinary thoracic committees. To provide new insights into the management and prognosis of patients with cf-IIP, we assessed the sociodemographic and clinical differences between patients with and without LC to evaluate their role in their mortality. Other factors were also studied. A longitudinal study was conducted from January 1, 2001, to December 31, 2020, in our hospital. All patients who attended the interstitial lung disease multidisciplinary unit and had a medical diagnosis of cf-IIP were included. The primary outcome was all-cause mortality. The independent variable was the presence of LC. Covariates: sociodemographic, clinical, and therapeutic variables. The mortality rate (MR) was expressed per 1000 patient-years with a 95% confidence interval (CI). Cox multiple regression analysis examined the influence of LC and other covariates on mortality. The results are expressed as hazard ratios (HRs) with CIs. A total of 313 patients with cf-IIP were included, with a follow-up of 1589.6 patient-years. There were 209 (66.8%) deaths. The MR was 131.5 (114.8-150.6), and 50% of patients died at 5.58 years from cf-IIP diagnosis. After adjusting for confounders, LC was associated with a significantly increased risk of mortality (HR 4.11; 2.50-6.77; p = 0.000), whereas antifibrotic treatment was the only factor that decreased the risk of mortality (HR 0.60; 0.43-0.84; p = 0.003). The prevalence of LC was 15.3%. The most common histopathological type was squamous cell carcinoma (39.6%). At diagnosis, 68.7% of LC cases were in stages III-IV. The most widely used treatment was chemotherapy and its combination (43.7%). Patients with cf-IIP who developed LC had a fourfold increased risk of mortality and shorter mean survival (2 years vs. 6 years) compared with those without LC. Therefore, we recommend LC screening in these patients.

PMID:40369244 | DOI:10.1038/s41598-025-99792-1

Acta Pharmacol Sin. 2025 May 14. doi: 10.1038/s41401-025-01558-y. Online ahead of print.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive interstitial lung disease characterized by complex aetiologies involving the accumulation of inflammatory cells, such as macrophages, in the alveoli. This process is driven by uncontrolled extracellular matrix (ECM) deposition and the development of fibrous connective tissues. Here, we observed that the mRNA expression of Ffar1, the gene encoding G protein-coupled receptor 40 (GPR40), is repressed, while Cd36 is increased in the bronchoalveolar lavage fluid (BALF), which is predominantly composed of alveolar macrophages, of IPF patients. Furthermore, the GPR40 protein was found to be largely adhered to macrophages and was pathologically downregulated in the lungs of bleomycin (BLM)-induced PF model mice (PF mice) compared with those of control mice. Specific knockdown of GPR40 in pulmonary macrophages by adeno-associated virus 9-F4/80-shGPR40 (AAV9-shGPR40) exacerbated the fibrotic phenotype in the PF mice, and activation of GPR40 by its determined agonist compound SC (1,3-dihydroxy-8-methoxy-9H-xanthen-9-one) effectively protected the PF mice from pathological exacerbation. Moreover, Ffar1 or Cd36 gene knockout mouse-based assays were performed to explore the mechanism underlying the regulation of GPR40 activation in pulmonary macrophages with compound SC as a probe. We found that compound SC mitigated pulmonary fibrosis progression by preventing M2 macrophage polarization from exerting profibrotic effects through the GPR40/PKD1/CD36 axis. Our results strongly support the therapeutic potential of targeting intrinsic GPR40 activation in pulmonary macrophages for IPF and highlight the potential of compound SC in treating this disease.

PMID:40369224 | DOI:10.1038/s41401-025-01558-y

Sci Rep. 2025 May 14;15(1):16801. doi: 10.1038/s41598-025-01759-9.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a progressive lung disease. Recent evidence suggests that the pathogenesis of IPF may involve abnormalities in mitochondrial energy metabolism. This study aimed to identify mitochondrial energy metabolism related differentially expressed genes (MEMRDEGs) and to elucidate their potential mechanistic involvement in IPF. We employed a multistep bioinformatics approach, including data extraction from the Gene Expression Omnibus database, removal of batch effects, and normalization and differential gene expression analyses. We then conducted Gene Ontology, Kyoto Encyclopedia of Genes and Genomes enrichment, and gene set enrichment analyses. A protein-protein interaction network was constructed from the STRING database, and hub genes were identified. Receiver operating characteristic curve analysis was performed to evaluate immune infiltration. Our integrated analysis of IPF datasets identified 25 MEMRDEGs. Nine hub genes emerged as central to mitochondrial energy metabolism in IPF. COX5A, EHHADH, and SDHB are potential biomarkers for diagnosing IPF with high accuracy. Single-sample gene set enrichment analysis revealed significant differences in the abundances of specertainfic immune cell types between IPF samples and controls. In conclusion, COX5A, EHHADH, and SDHB are potential biomarkers for the high-accuracy diagnosis of IPF. These findings pave the way for further investigations into the molecular mechanisms underlying IPF.

PMID:40369105 | DOI:10.1038/s41598-025-01759-9

Intern Med. 2025;64(10):1552-1562. doi: 10.2169/internalmedicine.4493-24. Epub 2025 May 15.

ABSTRACT

Nintedanib inhibits disease progression in patients with idiopathic pulmonary fibrosis (IPF) and dramatically improves the lung function in patients known to be super-responders (SRs). However, there are no reports on quantitative high-resolution computed tomography (HRCT), and the HRCT imaging characteristics of SRs remain unknown. We herein present the case of a 66-year-old man with IPF who was a SR to nintedanib treatment, which showed a marked improvement compared to other patients with IPF upon quantitative HRCT evaluation using the artificial intelligence (AI) software 3D Slicer. This study is worth reporting because a quantitative HRCT assessment using AI may be necessary to understand the SR characteristics.

PMID:40368830 | DOI:10.2169/internalmedicine.4493-24

Intern Med. 2025 May 15. doi: 10.2169/internalmedicine.5575-25. Online ahead of print.

NO ABSTRACT

PMID:40368795 | DOI:10.2169/internalmedicine.5575-25

Intern Med. 2025 May 15. doi: 10.2169/internalmedicine.5447-25. Online ahead of print.

ABSTRACT

Objectives To analyze the influence of age of the onset on myositis organ damage and to identify the factors influencing myositis organ damage, as clinical manifestations of myopathies differ by the age of onset and the background of patients. Methods Factors influencing organ damage [SLICC/ACR Damage Index (SDI)] were identified using the Japanese multicenter myositis registry (MYKO, n=220). Factors influencing organ damage were identified using a multivariate analysis. SDI was compared among juvenile-onset (<20 years old), adolescent-onset (20-64 years), and elderly-onset (>64 years) groups. Results There was a correlation between the age at onset and the SDI score (Spearman's rank correlation coefficient ρ=0.28). Elderly patients exhibited more widespread organ damage, including neuropsychiatric, renal, pulmonary, cardiovascular, peripheral vascular, gastrointestinal, skin, and diabetes, whereas juvenile-onset patients exhibited musculoskeletal damage. Adolescent-onset patients had the lowest incidence of ocular and malignant damage. A regression analysis revealed that an older onset age (coefficient, β=0.03), longer disease duration (β=0.05), and total dose of glucocorticoid (β=3.35x10-5) influenced SDI. After adjusting for disease duration, the influences of anti-MDA5 [hazard ratio (95% confidence interval), 4.47 (2.17-9.21)] on pulmonary fibrosis and a history of steroid pulse [2.16 (1.16-4.05)] on muscle atrophy or weakness were shown. Conclusions There were associations between the age of onset and autoantibodies with myositis organ damage. Musculoskeletal damage was greater in patients with a juvenile onset. An older age of onset is associated with severe organ damage. These findings highlight the importance of considering the age of onset and autoantibodies for assessing the prognosis and developing treatment plans for myopathies.

PMID:40368794 | DOI:10.2169/internalmedicine.5447-25

Nature. 2025 May 14. doi: 10.1038/s41586-025-09012-z. Online ahead of print.

ABSTRACT

Metastasis to the cerebrospinal-fluid-filled leptomeninges, or leptomeningeal metastasis, represents a fatal complication of solid tumours1. Multimodal analyses of clinical specimens reveal substantial inflammatory infiltrate in leptomeningeal metastases with enrichment of IFNγ and resulting downstream signalling. Here, to investigate and overcome this futile anti-tumour response within the leptomeninges, we developed syngeneic lung cancer, breast cancer and melanoma leptomeningeal-metastasis mouse models. We show that transgenic host mice lacking IFNγ or its receptor fail to control the growth of leptomeningeal metastases growth. Leptomeningeal overexpression of Ifng through a targeted adeno-associated-virus-based system controls cancer cell growth independent of adaptive immunity. Using a suite of transgenic hosts, we demonstrate that leptomeningeal T cells generate IFNγ to actively recruit and activate peripheral myeloid cells, generating a diverse spectrum of dendritic cell subsets. Independent of antigen presentation, migratory CCR7+ dendritic cells orchestrate the influx, proliferation and cytotoxic action of natural killer cells to control cancer cell growth in the leptomeninges. This study identifies unique, leptomeninges-specific IFNγ signalling and suggests an immune-therapeutic approach against tumours within this space.

PMID:40369076 | DOI:10.1038/s41586-025-09012-z

Nat Commun. 2025 May 14;16(1):4428. doi: 10.1038/s41467-025-59059-9.

ABSTRACT

Embryonic macrophages emerge before the onset of definitive hematopoiesis, seed into discrete tissues and contribute to specialized resident macrophages throughout life. Presence of embryonic macrophages in the bone marrow and functional impact on hematopoietic stem cells (HSC) or the niche remains unclear. Here we show that bone marrow macrophages consist of two ontogenetically distinct cell populations from embryonic and adult origin. Newborn mice lacking embryonic macrophages have decreased HSC numbers in the bone marrow suggesting an important function for embryo-derived macrophages in orchestrating HSC trafficking around birth. The establishment of a normal cellular niche space in the bone marrow critically depends on embryonic macrophages that are important for the development of mesenchymal stromal cells, but not other non-hematopoietic niche cells, providing evidence for a specific role for embryo-derived macrophages in the establishment of the niche environment pivotal for the establishment of a normally sized HSC pool.

PMID:40368907 | DOI:10.1038/s41467-025-59059-9

Trends Plant Sci. 2025 May 13:S1360-1385(25)00085-8. doi: 10.1016/j.tplants.2025.03.011. Online ahead of print.

ABSTRACT

The mathematics of mass screening is crucial for understanding results from imperfect detection methods in large datasets. Although this framework is well-established, it is not commonly applied in plant biology. Here, we view the identification of messenger RNAs that travel over long distances in plants (mobile mRNAs) through the lens of mass screening statistics. RNA-Seq analyses have identified thousands of mobile mRNAs. Consideration of the detection accuracy and prevalence, however, cast doubt on these numbers. The presented methodology is relevant to all areas of research where detection tests with less than 100% accuracy are applied to find rare events in large datasets.

PMID:40368680 | DOI:10.1016/j.tplants.2025.03.011

Antiviral Res. 2025 May 12:106187. doi: 10.1016/j.antiviral.2025.106187. Online ahead of print.

ABSTRACT

Lipovirtide, originally designated as LP-80, is a stearic acid-modified lipopeptide HIV fusion inhibitor with highly potent and long-lasting anti-HIV activity, and it has already progressed to phase II clinical trials. In this report, we investigated the pharmacokinetics and safety profile of LP-80 in Sprague Dawley (SD) rats. LP-80 was absorbed rapidly following subcutaneous injection, exhibiting high absolute bioavailability (F): 92.32% in male and 84.74% in female. The time to reach maximum plasma concentration (Tmax) ranged from 5.5 to 8 hours (h), and the elimination half-life (T1/2) was between 6.26 and 7.47 h, indicating a relatively long-lasting presence in the bloodstream. LP-80 was widely distributed across various tissues, with the highest concentration observed in serum, suggesting effective systemic delivery and potential for targeting HIV in different compartments. Only a minimal amount of the parent drug was excreted in feces and urine, which indicates that LP-80 is metabolically stable and not rapidly cleared from the body. Acute, subchronic, and chronic toxicity studies demonstrated that LP-80 was well tolerated in animals, with no significant adverse effects observed. No anti-drug antibodies (ADA) were detected, suggesting low immunogenicity. Furthermore, LP-80 showed no toxic effects on fertility, embryo-fetal development, or offspring development. Collectively, our studies demonstrate that LP-80 is metabolically stable and exhibits a favorable safety profile, supporting its advancement into clinical trials for HIV treatment.

PMID:40368113 | DOI:10.1016/j.antiviral.2025.106187

Int J Antimicrob Agents. 2025 May 12:107538. doi: 10.1016/j.ijantimicag.2025.107538. Online ahead of print.

NO ABSTRACT

PMID:40368009 | DOI:10.1016/j.ijantimicag.2025.107538

PLoS One. 2025 May 14;20(5):e0323597. doi: 10.1371/journal.pone.0323597. eCollection 2025.

ABSTRACT

Kawasaki disease (KD) is an acute vasculitis that primarily affects children under five and is a leading cause of acquired heart disease in this age group. Despite the standard treatment with intravenous immunoglobulin (IVIG), approximately 10-20% of patients exhibit IVIG resistance, leading to persistent inflammation and an increased risk of coronary artery aneurysms（CAA）. The underlying molecular mechanisms driving KD, particularly the role of pyroptosis, remain incompletely understood. In this study, we employed integrative bioinformatics approaches to investigate the mechanistic role of pyroptosis in KD. By analyzing transcriptomic datasets, we identified differentially expressed genes (DEGs) associated with pyroptosis and immune dysregulation. Weighted Gene Co-Expression Network Analysis (WGCNA) was utilized to uncover key co-expressed gene modules, followed by functional enrichment analyses to explore the biological significance of these genes. Through machine learning-based biomarker discovery, we identified MYD88 and S100A12 as critical pyroptosis-related genes in KD. Their diagnostic potential was validated using external datasets, and their involvement in immune cell infiltration was assessed through computational deconvolution techniques. Furthermore, drug repurposing analysis and molecular docking simulations suggested that Atogepant, Ubrogepant, and Zanubrutinib could serve as potential therapeutic candidates targeting S100A12 and MYD88. These findings provide novel insights into the molecular pathogenesis of KD and highlight potential biomarkers and therapeutic targets for improving KD diagnosis and treatment strategies.

PMID:40367231 | DOI:10.1371/journal.pone.0323597

J Med Internet Res. 2025 May 14;27:e70733. doi: 10.2196/70733.

ABSTRACT

BACKGROUND: Online health information is widely available, but a substantial portion of it is inaccurate or misleading, including exaggerated, incomplete, or unverified claims. Such misinformation can significantly influence public health decisions and pose serious challenges to health care systems. With advances in artificial intelligence and natural language processing, pretrained large language models (LLMs) have shown promise in identifying and distinguishing misleading health information, although their effectiveness in this area remains underexplored.

OBJECTIVE: This study aimed to evaluate the performance of 4 mainstream LLMs (ChatGPT-3.5, ChatGPT-4, Ernie Bot, and iFLYTEK Spark) in the identification of online health information, providing empirical evidence for their practical application in this field.

METHODS: Web scraping was used to collect data from rumor-refuting websites, resulting in 2708 samples of online health information, including both true and false claims. The 4 LLMs' application programming interfaces were used for authenticity verification, with expert results as benchmarks. Model performance was evaluated using semantic similarity, accuracy, recall, F1-score, content analysis, and credibility.

RESULTS: This study found that the 4 models performed well in identifying online health information. Among them, ChatGPT-4 achieved the highest accuracy at 87.27%, followed by Ernie Bot at 87.25%, iFLYTEK Spark at 87%, and ChatGPT-3.5 at 81.82%. Furthermore, text length and semantic similarity analysis showed that Ernie Bot had the highest similarity to expert texts, whereas ChatGPT-4 showed good overall consistency in its explanations. In addition, the credibility assessment results indicated that ChatGPT-4 provided the most reliable evaluations. Further analysis suggested that the highest misjudgment probabilities with respect to the LLMs occurred within the topics of food and maternal-infant nutrition management and nutritional science and food controversies. Overall, the research suggests that LLMs have potential in online health information identification; however, their understanding of certain specialized health topics may require further improvement.

CONCLUSIONS: The results demonstrate that, while these models show potential in providing assistance, their performance varies significantly in terms of accuracy, semantic understanding, and cultural adaptability. The principal findings highlight the models' ability to generate accessible and context-aware explanations; however, they fall short in areas requiring specialized medical knowledge or updated data, particularly for emerging health issues and context-sensitive scenarios. Significant discrepancies were observed in the models' ability to distinguish scientifically verified knowledge from popular misconceptions and in their stability when processing complex linguistic and cultural contexts. These challenges reveal the importance of refining training methodologies to improve the models' reliability and adaptability. Future research should focus on enhancing the models' capability to manage nuanced health topics and diverse cultural and linguistic nuances, thereby facilitating their broader adoption as reliable tools for online health information identification.

PMID:40367512 | DOI:10.2196/70733

Einstein (Sao Paulo). 2025 May 12;23:eAO1312. doi: 10.31744/einstein_journal/2025AO1312. eCollection 2025.

ABSTRACT

BACKGROUND: Santos et al. analyzed the clinical characteristics and pulmonary function of children with cystic fibrosis infected with SARS-CoV-2. Infected children showed higher rates of dyspnea, coughing, hospitalization, and pulmonary exacerbations. Despite a temporary decline in pulmonary function, the recovery rates matched those of the uninfected children during follow-up. ■ SARS-CoV-2 infection leads to mild-to-moderate disease in children with cystic fibrosis. ■ No worsening of cystic fibrosis was observed months after infection.

OBJECTIVE: This study aimed to evaluate the clinical manifestations of SARS-CoV-2 in children and adolescents with cystic fibrosis.

METHODS: This was a case-control analysis of clinical variables and pulmonary function test results in 43 children with cystic fibrosis, 17 (39.5%) of whom tested positive for SARS-CoV-2.

RESULTS: The infected children exhibited a higher frequency of dyspnea and cough and a greater need for hospitalization. One infected child died. Pulmonary exacerbations were more frequent among the infected children. Additional data indicated a subsequent reduction in pulmonary function in the infected children, although this was not significantly different from that in the uninfected children.

CONCLUSION: Children with cystic fibrosis who contracted SARS-CoV-2 experienced mild to moderate symptoms and required hospitalization but generally had high recovery rates.

PMID:40367008 | DOI:10.31744/einstein_journal/2025AO1312

Comput Biol Med. 2025 May 13;192(Pt B):110304. doi: 10.1016/j.compbiomed.2025.110304. Online ahead of print.

ABSTRACT

Percutaneous screw fixation technique in pelvic trauma surgery is an extremely challenging operation that typically requires a trial-and-error insertion process under the guidance of continuous intraoperative X-ray. This process can be simplified by utilizing surgical navigation systems. Understanding the complexity of the intraosseous pelvis corridor is essential for establishing the optimal screw corridor, which further facilitates preoperative planning and intraoperative application. Traditional screw corridor search algorithms necessitate traversing the entrance and exit areas of the screw and calculating the distance from the corridor axis to the bone surface to ascertain the location of the screw. This process is computationally complex, and manual measurement by the physician is time consuming, labor intensive, and empirically dependent. In this study, we propose an automated planning algorithm for pelvic screw corridors based on deep learning deformable registration technology, which can efficiently and accurately identify the optimal screw corridors. Compared to traditional methods, the innovations of this study include: (1) the introduction of corridor safety range constraints on screw positioning, which enhances search efficiency; (2) the application of deep learning deformable registration to facilitate the automatic annotation of the screw entrance and exit areas, as well as the safety range of the corridor; and (3) the development of a highly efficient algorithm for optimal corridor searching, quickly determining the corridor without traversing the entrance and exit areas and enhancing efficiency via a vector-based diameter calculation method. The whole framework of the algorithm consists of three key components: atlas generation module, deformable registration and optimal corridor searching strategy. In the experiments, we test the performance of the proposed algorithm on 198 intact pelvises for calculating the optimal corridor of anterior column corridor and S1 sacroiliac screws. The results show that the new algorithm can increase the corridor diameter by 2.1%-3.3% compared to manual measurements, while significantly reducing the average time from 1038s and 3398s to 18.9s and 26.7s on anterior column corridor and S1 sacroiliac corridor, respectively, compared to the traditional screw searching algorithm. This demonstrates the advantages of the algorithm in terms of efficiency and accuracy. However, the current method is validated only on intact pelvises; further research is required for pelvic fracture scenarios.

PMID:40367630 | DOI:10.1016/j.compbiomed.2025.110304

Comput Biol Med. 2025 May 13;192(Pt B):110266. doi: 10.1016/j.compbiomed.2025.110266. Online ahead of print.

ABSTRACT

Accurate and secure classifying informal documents related to mental disorders is challenging due to factors such as informal language, noisy data, cultural differences, personal information and mixed emotions. Conventional deep learning models often struggle to capture patterns in informal text, as they miss long-range dependencies, explain words and phrases literally, and have difficulty processing non-standard inputs like emojis. To address these limitations, we expand data dimensionality, transforming and fusing textual data and signs from a 1D to a 2D space. This enables the use of pre-trained 2D CNN models, such as AlexNet, Restnet-50, and VGG-16 removing the need to design and train new models from scratch. We apply this approach to a dataset of social media posts to classify informal documents as either related to suicide or non-suicide content. Our results demonstrate high classification accuracy, exceeding 99%. In addition, our 2D visual data representation conceals individual private information and helps explainability.

PMID:40367624 | DOI:10.1016/j.compbiomed.2025.110266