Ultramicroscopy. 2025 Apr 5;273:114137. doi: 10.1016/j.ultramic.2025.114137. Online ahead of print.

ABSTRACT

Precise alignment of the electron beam is critical for successful application of scanning transmission electron microscopes (STEM) to understanding materials at atomic level. Despite the success of aberration correctors, aberration correction is still a complex process. Here we approach aberration correction from the perspective of accelerator physics and show it is equivalent to minimizing the emittance growth of the beam, the span of the phase space distribution of the probe. We train a deep learning model to predict emittance growth from experimentally accessible Ronchigrams. Both simulation and experimental results show the model can capture the emittance variation with aberration coefficients accurately. We further demonstrate the model can act as a fast-executing function for the global optimization of the lens parameters. Our approach enables new ways to quickly quantify and automate aberration correction that takes advantage of the rapid measurements possible with high-speed electron cameras. In part II of the paper, we demonstrate how the emittance metric enables rapid online tuning of the aberration corrector using Bayesian optimization.

PMID:40222084 | DOI:10.1016/j.ultramic.2025.114137

Comput Biol Chem. 2025 Apr 5;118:108456. doi: 10.1016/j.compbiolchem.2025.108456. Online ahead of print.

ABSTRACT

The prevalent type of blood cancer is called leukemia, which is caused by the irregular production of immature malignant cells in the bone marrow. This dangerous condition weakens the immune system, making the body susceptible to infections, and can lead to death if not treated quickly. Thus, immediate treatments are necessary to detect leukemia at the initial stage to control abnormal cell growth. Leukemia detection from microscopic images of blood smears of malignant leukemia cells is a time-consuming and tedious task. Thus, a Tangent Sand Cat Swarm Optimization-Long Short-Term Memory-LeNet (TSCO-L-LeNet) with incremental learning is designed for the precise classification of acute lymphoblastic leukemia. The proposed model offers cheaper, faster and safer diagnosis service as the use of blood smear images reduces the diagnosis time and improves accuracy. Here, the input image is pre-processed using the adaptive median filter and the Scribble2label is used to segment the image. Later, the augmentation of segmented image is performed and the feature extraction process is employed to extract the necessary features from the augmented image. Finally, the L-LeNet with incremental learning is executed for acute lymphoblastic leukemia classification from the extracted features, where the TSCO approach is used to train the weights of L-LeNet. The experimental results show that TSCO-L-LeNet achieved maximum performance of 0.987 for accuracy, 0.977 for True Negative Rate (TNR), 0.967 for recall, 0.033 for False Negative rate, 0.023 for False Positive rate, and 0.979 for precision.

PMID:40222054 | DOI:10.1016/j.compbiolchem.2025.108456

Eur Radiol. 2025 Apr 13. doi: 10.1007/s00330-025-11574-5. Online ahead of print.

ABSTRACT

OBJECTIVE: This study aims to assess diagnostic performance of high-resolution dynamic contrast-enhanced (DCE) MRI with deep learning-based compressed sensing and super-resolution (DLCS-SR) reconstruction for identifying microadenomas.

MATERIALS AND METHODS: This prospective study included 126 participants with suspected pituitary microadenomas who underwent DCE MRI between June 2023 and January 2024. Four image groups were derived from single-scan DCE MRI, which included 1.5-mm slice thickness images using DLCS-SR (1.5-mm DLCS-SR images), 1.5-mm slice thickness images with deep learning-based compressed sensing reconstruction (1.5-mm DLCS images), 1.5-mm routine images, and 3-mm slice thickness images using DLCS-SR (3-mm DLCS-SR images). Diagnostic criteria were established by incorporating laboratory findings, clinical symptoms, medical histories, previous imaging, and certain pathologic reports. Two readers assessed the diagnostic performance in identifying pituitary abnormalities and microadenomas. Diagnostic agreements were assessed using κ statistics, and intergroup comparisons for microadenoma detection were performed using the DeLong and McNemar tests.

RESULTS: The 1.5-mm DLCS-SR images (κ = 0.746-0.848) exhibited superior diagnostic agreement, outperforming 1.5-mm DLCS (κ = 0.585-0.687), 1.5-mm routine (κ = 0.449-0.487), and 3-mm DLCS-SR images (κ = 0.347-0.369) (p < 0.001 for all). Additionally, the performance of 1.5-mm DLCS-SR images in identifying microadenomas [area under the receiver operating characteristic curve (AUC), 0.89-0.94] surpassed that of 1.5-mm DLCS (AUC, 0.83-0.87; p = 0.042 and 0.011, respectively), 1.5-mm routine (AUC, 0.76-0.78; p < 0.001), and 3-mm DLCS-SR images (AUC, 0.72-0.74; p < 0.001).

CONCLUSION: The findings revealed superior diagnostic performance of 1.5-mm DLCS-SR images in identifying pituitary abnormalities and microadenomas, indicating the clinical-potential of high-resolution DCE MRI.

KEY POINTS: Question What strategies can overcome the resolution limitations of conventional dynamic contrast-enhanced (DCE) MRI, and which contribute to a high false-negative rate in diagnosing pituitary microadenomas? Findings Deep learning-based compressed sensing and super-resolution reconstruction applied to DCE MRI achieved high resolution while improving image quality and diagnostic efficacy. Clinical relevance DCE MRI with a 1.5-mm slice thickness and high in-plane resolution, utilizing deep learning-based compressed sensing and super-resolution reconstruction, significantly enhances diagnostic accuracy for pituitary abnormalities and microadenomas, enabling timely and effective patient management.

PMID:40221940 | DOI:10.1007/s00330-025-11574-5

Dentomaxillofac Radiol. 2025 Apr 12:twaf027. doi: 10.1093/dmfr/twaf027. Online ahead of print.

ABSTRACT

OBJECTIVES: This systematic review aimed to assess the performance of artificial intelligence (AI) in the imaging diagnosis of maxillary sinusitis (MS) compared to human analysis.

METHODS: Studies that presented radiographic images for the diagnosis of paranasal sinus diseases, as well as control groups for AI, were included. Articles that performed tests on animals, presented other conditions, surgical methods, didn't present data on the diagnosis of MS or on the outcomes of interest (area under the curve, sensitivity, specificity, and accuracy), compared the outcome only among different AIs, were excluded. Searches were conducted in five electronic databases and a gray literature. The risk of bias (RB) was assessed using the QUADAS-2 and the certainty of evidence by GRADE.

RESULTS: Six studies were included. The type of study considered was retrospective observational; with serious RB, and a considerable heterogeneity in methodologies. The IA presents similar results to humans, however, imprecision was assessed as serious for the outcomes and the certainty of evidence was classified as very low according to the GRADE approach. Furthermore, a dose-response effect was determined, as specialists demonstrate greater mastery of the diagnosis of MS when compared to resident professionals or general clinicians.

CONCLUSIONS: Considering the outcomes, the AI represents a complementary tool for diagnosing MS, especially considering professionals with less experience. Finally, performance analysis and definition of comparison parameters should be encouraged considering future research perspectives.

ADVANCES IN KNOWLEDGE: AI can be used as a complementary tool for diagnosing MS, however studies are still lacking methodological standardization.

PMID:40221848 | DOI:10.1093/dmfr/twaf027

Nat Commun. 2025 Apr 12;16(1):3496. doi: 10.1038/s41467-025-58437-7.

ABSTRACT

Preeclampsia is a major cause of maternal and perinatal mortality with no known cure. Delivery timing is critical to balancing maternal and fetal risks. We develop and externally validate PEDeliveryTime, a class of clinically informative models which resulted from deep-learning models, to predict the time from PE diagnosis to delivery using electronic health records. We build the models on 1533 PE cases from the University of Michigan and validate it on 2172 preeclampsia cases from the University of Florida. PEDeliveryTime full model contains only 12 features yet achieves high c-index of 0.79 and 0.74 on the Michigan and Florida data set respectively. For the early-onset preeclampsia subset, the full model reaches 0.76 and 0.67 on the Michigan and Florida test sets. Collectively, these models perform an early assessment of delivery urgency and might help to better prioritize medical resources.

PMID:40221413 | DOI:10.1038/s41467-025-58437-7

Nat Commun. 2025 Apr 12;16(1):3491. doi: 10.1038/s41467-025-58481-3.

ABSTRACT

The three-dimensional organization of chromatin is essential for gene regulation and cellular function, with epigenome playing a key role. Hi-C methods have expanded our understanding of chromatin interactions, but their high cost and complexity limit their use. Existing models for predicting chromatin interactions rely on limited ChIP-seq inputs, reducing their accuracy and generalizability. In this work, we present a computational approach, EpiVerse, which leverages imputed epigenetic signals and advanced deep learning techniques. EpiVerse significantly improves the accuracy of cross-cell-type Hi-C prediction, while also enhancing model interpretability by incorporating chromatin state prediction within a multitask learning framework. Moreover, EpiVerse predicts Hi-C contact maps across an array of 39 human tissues, which provides a comprehensive view of the complex relationship between chromatin structure and gene regulation. Furthermore, EpiVerse facilitates unprecedented in silico perturbation experiments at the "epigenome-level" to unveil the chromatin architecture under specific conditions. EpiVerse is available on GitHub: https://github.com/jhhung/EpiVerse .

PMID:40221401 | DOI:10.1038/s41467-025-58481-3

Acad Radiol. 2025 Apr 11:S1076-6332(25)00291-0. doi: 10.1016/j.acra.2025.03.048. Online ahead of print.

ABSTRACT

RATIONALE AND OBJECTIVES: To construct a deep learning model (DL) based on high-resolution T2-weighted images for preoperative differentiation between T2 and T3 stage rectal cancer (RC), and to compare its performance with experienced radiologists.

METHODS: This retrospective study included 281 patients with pathologically confirmed RC from four centers (January 2017-December 2022). A DenseNet model was developed using 255 patients from three centers (training:validation ratio=8:2) and externally tested on 26 patients from a fourth center. Two experienced radiologists independently assessed T staging. Diagnostic performance was evaluated using accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).

RESULTS: The DL model outperformed radiologists in differentiating T2 and T3 stages across all datasets. In the training set, the DL model achieved an AUC of 0.810, compared to 0.578 and 0.625 for radiologists A and B, respectively. In the external test set, the DL model maintained superior diagnostic performance (AUC=0.715) compared to radiologist A (AUC=0.549) and radiologist B (AUC=0.493). The DL model demonstrated higher accuracy for T2 staging (0.625-0.787) and T3 staging (0.611-0.814) compared to radiologists (0.373-0.526 for T2; 0.611-0.783 for T3), who showed a tendency to over-stage T2 tumors. Inter-observer agreement between radiologists was moderate (kappa=0.451).

CONCLUSION: The DenseNet-based DL model demonstrated superior accuracy and diagnostic efficiency than radiologists in preoperative differentiation between T2 and T3 stages RC. This automated approach could potentially improve staging accuracy and support clinical decision-making in RC treatment planning.

PMID:40221285 | DOI:10.1016/j.acra.2025.03.048

Radiol Clin North Am. 2025 May;63(3):477-490. doi: 10.1016/j.rcl.2024.11.005. Epub 2025 Jan 4.

ABSTRACT

While artificial intelligence (AI) has shown considerable progress in many areas of medical imaging, applications in abdominal imaging, particularly for the gastrointestinal (GI) system, have notably lagged behind advancements in other body regions. This article reviews foundational concepts in AI and highlights examples of AI applications in GI tract imaging. The discussion on AI applications includes acute & emergent GI imaging, inflammatory bowel disease, oncology, and other miscellaneous applications. It concludes with a discussion of important considerations for implementing AI tools in clinical practice, and steps we can take to accelerate future developments in the field.

PMID:40221188 | DOI:10.1016/j.rcl.2024.11.005

Neuroimage. 2025 Apr 10:121190. doi: 10.1016/j.neuroimage.2025.121190. Online ahead of print.

ABSTRACT

Diffusion MRI (dMRI) offers unique insights into the microstructure of fetal brain tissue in utero. Longitudinal and cross-sectional studies of fetal dMRI have the potential to reveal subtle but crucial changes associated with normal and abnormal neurodevelopment. However, these studies depend on precise spatial alignment of data across scans and subjects, which is particularly challenging in fetal imaging due to the low data quality, rapid brain development, and limited anatomical landmarks for accurate registration. Existing registration methods, primarily developed for superior-quality adult data, are not well-suited for addressing these complexities. To bridge this gap, we introduce FetDTIAlign, a deep learning approach tailored to fetal brain dMRI, enabling accurate affine and deformable registration. FetDTIAlign integrates a novel dual-encoder architecture and iterative feature-based inference, effectively minimizing the impact of noise and low resolution to achieve accurate alignment. Additionally, it strategically employs different network configurations and domain-specific image features at each registration stage, addressing the unique challenges of affine and deformable registration, enhancing both robustness and accuracy. We validated FetDTIAlign on a dataset covering gestational ages between 23 and 36 weeks, encompassing 60 white matter tracts. For all age groups, FetDTIAlign consistently showed superior anatomical correspondence and the best visual alignment in both affine and deformable registration, outperforming two classical optimization-based methods and a deep learning-based pipeline. Further validation on external data from the Developing Human Connectome Project demonstrated the generalizability of our method to data collected with different acquisition protocols. Our results show the feasibility of using deep learning for fetal brain dMRI registration, providing a more accurate and reliable alternative to classical techniques. By enabling precise cross-subject and tract-specific analyses, FetDTIAlign paves the way for new discoveries in early brain development. The code is available at https://gitlab.com/blibli/fetdtialign.

PMID:40221066 | DOI:10.1016/j.neuroimage.2025.121190

Sci Rep. 2025 Apr 12;15(1):12698. doi: 10.1038/s41598-025-96284-0.

ABSTRACT

The dengue virus, a member of the arbovirus family, can cause a variety of clinical symptoms. However, there are currently no Food and Drug Administration-approved drugs are currently available for its treatment. We have used RNA-dependent RNA polymerase to identify drug candidates against dengue virus 2 or dengue virus 3. The Smina molecular docking program was used to screen natural compounds and FDA-approved drugs. This study used the pkCSM web server for pharmacokinetic profiling, OSIRIS Data Warrior for physicochemical property assessment, Data Warrior software for cytotoxicity profiling, and molecular dynamics simulations to evaluate the stability of ligand-RdRp interactions. Specifically, the drugs and compounds with the highest negative binding energy and most hydrogen bonds are chlorthalidone, valdecoxib, and ZINC14824819, which interact with the RdRp domain of dengue virus 2, and empagliflozin, netarsudil, and ZINC13375652, which interact with the RdRp domain of dengue virus 3. We propose several FDA-approved drugs and natural compounds that can bind to the RdRp of dengue virus serotypes 2 and 3 and prevent the virus from infecting cells. These compounds show a high level of safety and strong skin and intestinal absorption. Further in vitro and in vivo testing is needed to verify these predictions and assess therapeutic potential.

PMID:40221558 | DOI:10.1038/s41598-025-96284-0

Rev Neurol (Paris). 2025 Apr 10:S0035-3787(25)00495-3. doi: 10.1016/j.neurol.2025.04.003. Online ahead of print.

ABSTRACT

In recent years, the increasing availability of antiseizure medications has not reduced the incidence of drug-resistant epilepsy. Precision medicine offers the potential for mechanism-driven treatments for rare pediatric epilepsies. The concept of precision medicine is not new in the field of epilepsy, as demonstrated by the use of pyridoxine for antiquitin deficiency (pyridoxine-dependent epilepsy) and the ketogenic diet for GLUT1 deficiency syndrome. More recently, preclinical evidence has led to phase 3 clinical trials, such as the use of everolimus to inhibit the mTOR pathway in tuberous sclerosis complex. However, preclinical findings do not always translate into effective treatments, as illustrated by the heterogeneous effects of quinidine in KCNT1-related epilepsy. Currently, an exponential increase in compounds identified at the preclinical level will require clinical trial validation. However, it remains uncertain whether these developments will lead to improved efficacy in drug-resistant epilepsy or have any disease-modifying effects. This article does not explicitly address antisense oligonucleotides or gene therapy.

PMID:40221358 | DOI:10.1016/j.neurol.2025.04.003

Neurobiol Dis. 2025 Apr 10:106909. doi: 10.1016/j.nbd.2025.106909. Online ahead of print.

ABSTRACT

BACKGROUND: Previous studies have shown significant sex differences in AD with regarding its epidemiology, pathophysiology, clinical presentation, and treatment response. However, the transcriptome variances associated with sex in AD remain unclear.

METHODS: RNA sequencing (RNA-seq) and transcriptomic analyses were performed on peripheral blood samples from total of 54 patients, including male AD patients (n = 15), female AD patients (n = 10), male MCI patients (n = 7), female MCI patients (n = 11), male healthy controls (n = 6), female healthy controls (n = 5). The snRNA-seq dataset (GSE167494, GSE157827) of prefrontal cortex tissues was obtained from the Gene Expression Omnibus (GEO). We conducted an investigation into differentially expressed genes and pathways in the peripheral blood cells as well as prefrontal cortex tissues of both male and female AD patients with consideration to sex-related factors. Additionally, we analyzed the distribution and characteristics of cells in the cerebral cortex as well as the interaction and communication between cells of male and female AD patients. Connectivity Map (CMap) was utilized for predicting and screening potential sex-specific drugs for AD.

RESULTS: The transcriptome profile and associated biological processes in the peripheral blood of male and female AD and MCI patients exhibit discernible differences, including upregulation of BASP1 in AD male patients and arousing TNS1 in AD female patients. The distribution of various cell types in the prefrontal cortex tissues differs between male and female AD patients, like neuron and oligodendrocyte decreased and endothelial cell and astrocyte increased in female compared with male, while a multitude of genes exhibit significant differential expression. The results of cell communication analysis, such as collagen signaling pathway, suggest that sex disparities impact intercellular interactions within prefrontal cortex tissues among individuals with AD. By drug repositioning, several drugs, including torin-2 and YM-298198, might have the potential to therapeutic value of MCI or AD, while drugs like homoharringtonine and teniposide have potential opposite effects in different sexes.

CONCLUSION: The characteristics of the transcriptome in peripheral blood and single-cell transcriptome in the prefrontal cortex exhibit significant differences between male and female patients with AD, which providing a basis for future sex stratified treatment of AD.

PMID:40220916 | DOI:10.1016/j.nbd.2025.106909

Int Immunopharmacol. 2025 Apr 11;155:114622. doi: 10.1016/j.intimp.2025.114622. Online ahead of print.

ABSTRACT

The lysosomal cysteine protease cathepsin S supports host defence by promoting the maturation of MHC class-II proteins. In contrast, increased cathepsin S activity mediates tissue destructive immune responses in autoimmune and inflammatory diseases. Therefore, cathepsin S is a key target in drug discovery programs. Here, we critically reviewed the specific mechanisms by which cathepsin S mediates autoimmune and hyperinflammatory responses to identify new targets for therapeutic immunomodulation. To this end, we performed literature review utilizing PubMed, drug database of US FDA, European Medicines Agency and the Drug-Gene Interaction Database. Cathepsin S destroys T cell epitopes and reduces endogenous antigen diversity, impairing negative selection of autoreactive T cells that could recognize these epitopes. Moreover, cathepsin S critically regulates inflammatory disease severity by generating proinflammatory molecules (PAR-1, PAR-2, IL-36γ, Fractalkine, Endostatin, Ephrin-B2), inactivating anti-inflammatory mediators (SLPI) and degrading molecules involved in antimicrobial and immunomodulatory responses (surfactant protein-A, LL-37, beta-defensins), inter-endothelial/-epithelial barrier function, gene repair and energy homeostasis. These pathways could be targeted by repositioning of existing drugs. These findings suggest that inhibiting cathepsin S or a specific downstream target of cathepsin S by repositioning of existing drugs could be a promising strategy for treating autoimmune and inflammatory diseases. Current cathepsin S inhibitors in clinical trials face challenges, highlighting the need for innovative inhibitors that function effectively in various cellular compartments with differing pH levels, without targeting the shared catalytic site of cysteine cathepsins.

PMID:40220622 | DOI:10.1016/j.intimp.2025.114622

Hum Genomics. 2025 Apr 12;19(1):40. doi: 10.1186/s40246-025-00751-8.

ABSTRACT

Genomics of athletic performance is an emerging discipline with a high degree of controversy. With the existing level of evidence, it is both premature and highly risky to exploit current human genomics knowledge to predict exercise and sports performance or enhance existing training methodologies. Until more solid evidence on the influence of genomic variants in athletic performance becomes available, accompanied by regulatory approved genome-guided recommendations, all genetic associations should be restricted from general public access as commercial services, since genomic markers cannot per se predict athletic performance for talent identification, resistance to injuries or the ability to recover from them. Evidently, the complex interplay of genetics with other physical, physiological and even psychological and mental characteristics to produce a world-class athlete is still not understood.

PMID:40221803 | DOI:10.1186/s40246-025-00751-8

Orphanet J Rare Dis. 2025 Apr 12;20(1):176. doi: 10.1186/s13023-025-03676-6.

ABSTRACT

BACKGROUND: Treatment satisfaction can be described as the patient's experience in patients with cystic fibrosis (CF). It can be influenced using modulators and clinical characteristics. The aims of this study were to evaluate and compare adult CF patients with and without modulators regarding treatment satisfaction, related factors and to manage their drug related problems (DRPs).

METHODS: A single-center prospective cohort study was conducted between June 2023 and January 2024. Treatment Satisfaction Questionnaire for Medication (TSQM 1.4), including effectiveness, side effects, convenience, global satisfaction domains, CF Questionnaire-Revised (CFQ-R), and Medication Adherence Report Scale were applied and assessed with and without modulator therapy groups. The relationship between clinical characteristics and TSQM was analyzed by correlations and regression analysis. Recommendations on DRPs identified by clinical pharmacists were made to the physicians and patients and classified according to Pharmaceutical Care Network Europe (PCNE v9.1).

RESULTS: A total of 110 patients with 51 modulator therapy and 59 without modulator therapy were included. The mean global satisfaction score of modulator users was found to be 19.733 (p < 0.001) points higher than non-users. When the CFQ-R treatment burden score improved by 1point, global satisfaction score increased by 0.233 points (p < 0.001). When the number of hospitalizations increased by 1 day, the global satisfaction score decreased by 4.751 points (p < 0.001). A total of 84 DRPs were identified, and 69 (82.1%) of them were resolved.

CONCLUSIONS: Treatment satisfaction in adult CF patients is influenced by modulators, treatment burden, and clinical factors, so access to modulators is important. This is the first study to classify DRPs according to PCNE in CF. Clinical pharmacists contribute to the management of CF.

PMID:40221761 | DOI:10.1186/s13023-025-03676-6

NPJ Antimicrob Resist. 2025 Apr 12;3(1):27. doi: 10.1038/s44259-025-00093-4.

ABSTRACT

Two of the three most commonly used classes of antifungal drugs target the fungal membrane through perturbation of sterol biosynthesis or function. In addition to these triazole and polyene antifungals, recent research is identifying new antifungal molecules that perturb lipid biosynthesis and function. Here, we review fungal lipid biosynthesis pathways and their potential as targets for antifungal drug development. An emerging goal is discovering new molecules that potentiate contemporary antifungal drugs in part through perturbation of lipid form and function.

PMID:40221522 | DOI:10.1038/s44259-025-00093-4

BMC Genomics. 2025 Apr 12;26(1):367. doi: 10.1186/s12864-025-11534-9.

ABSTRACT

BACKGROUND: The recent AI breakthrough of AlphaFold2 has revolutionized 3D protein structural modeling, proving crucial for protein design and variant effects prediction. However, intrinsically disordered regions-known for their lack of well-defined structure and lower sequence conservation-often yield low-confidence models. The latest Variant Effect Predictor (VEP), AlphaMissense, leverages AlphaFold2 models, achieving over 90% sensitivity and specificity in predicting variant effects. However, the effectiveness of tools for variants in disordered regions, which account for 30% of the human proteome, remains unclear.

RESULTS: In this study, we found that predicting pathogenicity for variants in disordered regions is less accurate than in ordered regions, particularly for mutations at the first N-Methionine site. Investigations into the efficacy of variant effect predictors on intrinsically disordered regions (IDRs) indicated that mutations in IDRs are predicted with lower sensitivity and the gap between sensitivity and specificity is largest in disordered regions, especially for AlphaMissense and VARITY.

CONCLUSIONS: The prevalence of IDRs within the human proteome, coupled with the increasing repertoire of biological functions they are known to perform, necessitated an investigation into the efficacy of state-of-the-art VEPs on such regions. This analysis revealed their consistently reduced sensitivity and differing prediction performance profile to ordered regions, indicating that new IDR-specific features and paradigms are needed to accurately classify disease mutations within those regions.

PMID:40221640 | DOI:10.1186/s12864-025-11534-9

Sci Rep. 2025 Apr 12;15(1):12552. doi: 10.1038/s41598-025-97074-4.

ABSTRACT

Landslides are among the geological disasters that frequently occur worldwide and significantly restrict the sustainable development of society. Therefore, it is of great practical significance to perform landslide susceptibility assessment. In addressing issues such as limited training samples, inadequate utilization of spatially effective features, and high computational costs associated with existing methods, we propose a landslide susceptibility assessment method (DS-DRN), which uses a lightweight dense residual network with emphasis on deep spatial features. To minimize computational costs, we design a depthwise separable residual module that optimizes traditional convolution on residual branches into depthwise separable convolution. Additionally, to prevent vanishing gradient and improve the reuse rate of landslide feature information, dense connections are employed to construct a deep feature extraction module. Finally, the output of the model is fed into the Softmax classifier for landslide susceptibility prediction. Taking Ya'an City in Sichuan Province as the study area, we compare the proposed DS-DRN method with three widely used deep learning methods: CNN, CPCNN-RF, and U-net. Evaluating model accuracy and performance, the DS-DRN method exhibits the highest prediction accuracy while also saving computational costs. Therefore, the proposed model can better fit the complex nonlinear relationship in landslide susceptibility, effectively mine deep spatial features, and address the high computational costs associated with complex networks.

PMID:40221608 | DOI:10.1038/s41598-025-97074-4

Sci Rep. 2025 Apr 12;15(1):12661. doi: 10.1038/s41598-025-97331-6.

ABSTRACT

Early screening for cancer has proven to improve the survival rate and spare patients from intensive and costly treatments due to late diagnosis. Cancer screening in the healthy population involves an initial risk stratification step to determine the screening method and frequency, primarily to optimize resource allocation by targeting screening towards individuals who draw most benefit. For most screening programs, age and clinical risk factors such as family history are part of the initial risk stratification algorithm. In this paper, we focus on developing a blood marker-based risk stratification approach, which could be used to identify patients with elevated cancer risk to be encouraged for taking a diagnostic test or participate in a screening program. We demonstrate that the combination of simple, widely available blood tests, such as complete blood count and complete metabolic panel, could potentially be used to identify patients at risk for colorectal, liver, and lung cancers with areas under the ROC curve of 0.76, 0.85, 0.78, respectively. Furthermore, we hypothesize that such an approach could not only be used as pre-screening risk assessment for individuals but also as population health management tool, for example to better interrogate the cancer risk in certain sub-populations.

PMID:40221571 | DOI:10.1038/s41598-025-97331-6

Sci Rep. 2025 Apr 12;15(1):12576. doi: 10.1038/s41598-025-97946-9.

ABSTRACT

The Green View Index (GVI) is utilized to evaluate urban street value and ecosystem services and to gauge public perceptions of street greening. This study investigates the spatial heterogeneity of the GVI and its influencing factors in Yuzhong District, Chongqing, a mountainous city in China. Deep learning algorithms were employed to calculate the green visibility of street view images, and Geographic Weighted Regression (GWR) and the Optimal Parameter-Based Geodetector (OPGD) were utilized to analyze the relationships between GVI and factors such as road physical attributes, the Normalized Difference Vegetation Index (NDVI), and topographic features. The results indicate that: (1) In Yuzhong District, 58.9% of streets have a GVI within a low to moderate range, suggesting room for improvement. Higher GVI levels are generally associated with elevated Digital Elevation Models (DEM), while slope, aspect, and terrain undulation have relatively minor overall impacts on GVI. (2) The GVI is highest in the western regions and lowest in the eastern regions, with streets along the riversides exhibiting lower GVI levels. (3) GWR analysis reveals that road type and NDVI significantly influence the GVI. Higher DEM values promote increased GVI, whereas high road density suppresses it. (4) The interaction between influencing factors drives the differentiated distribution of GVI within the study area. The interaction effects between Road type, NDVI, and DEM are particularly notable among these.

PMID:40221555 | DOI:10.1038/s41598-025-97946-9