Nature. 2025 Jun 4. doi: 10.1038/s41586-025-09093-w. Online ahead of print.

ABSTRACT

RNA polymerase III (Pol III) transcribes highly demanded RNAs grouped into three types of classical promoters, including type 1 (5S rRNA), type 2 (tRNA) and type 3 (short non-coding RNAs, such as U6, 7SK and RNase H1) promoters1-7. While structures of the Pol III preinitiation complex (PIC)8-11 and elongation complex (EC)12-16 have been determined, the mechanism underlying the transition from initiation to elongation remains unclear. Here we reconstituted seven human Pol III transcribing complexes (TC4, TC5, TC6, TC8, TC10, TC12 and TC13) halted on U6 promoters with nascent RNAs of 4-13 nucleotides. Cryo-electron microscopy structures captured initially transcribing complexes (ITCs; TC4 and TC5) and ECs (TC6-13). Together with KMnO4 footprinting, the data reveal extensive modular rearrangements: the transcription bubble expands from PIC to TC5, followed by general transcription factor (GTF) dissociation and abrupt bubble collapse from TC5 to TC6, marking the ITC-EC transition. In TC5, SNAPc and TFIIIB remain bound to the promoter and Pol III, while the RNA-DNA hybrid adopts a tilted conformation with template DNA blocked by BRF2, a TFIIIB subunit. Hybrid forward translocation during ITC-EC transition triggers BRF2-finger retraction, GTF release and transcription-bubble collapse. Pol III then escapes the promoter while GTFs stay bound upstream, potentially enabling reinitiation. These findings reveal molecular insights into Pol III dynamics and reinitiation mechanisms on type 3 promoters of highly demanded small RNAs, with the earliest documented initiation-elongation transition for an RNA polymerase.

PMID:40468065 | DOI:10.1038/s41586-025-09093-w

Cell Res. 2025 Jun 4. doi: 10.1038/s41422-025-01132-5. Online ahead of print.

NO ABSTRACT

PMID:40467773 | DOI:10.1038/s41422-025-01132-5

Sci Rep. 2025 Jun 4;15(1):19528. doi: 10.1038/s41598-025-04273-0.

ABSTRACT

Fatty acids, particularly those derived from plant and animal oils, play roles in physiological functions and metabolic regulation in pigs. Imbalances in the ratio of polyunsaturated fatty acids (PUFAs) may influence inflammatory responses, including the modulation of pro-inflammatory cytokines. This study aimed to investigate the gene co-expression profiles in the skeletal muscle of pigs fed diets supplemented with 3.0% soybean, canola, or fish oils, correlating these profiles with cytokine abundance and identifying hub genes associated with immune-related pathways using a systems biology approach. Skeletal muscle samples from 27 pigs were subjected to RNA sequencing and weighted gene co-expression network analysis (WGCNA) to construct gene co-expression networks. The concentrations of six cytokines (IL-10, IFN-γ, IL-1β, IL-6, IL-18, and TNF-α) were quantified in muscle tissue using ELISA. Functional enrichment analysis and hub gene identification revealed several key genes involved in immune function and fatty acid metabolism. WGCNA uncovered distinct co-expression modules associated with specific dietary oil treatments. These findings provide new insights into the immunomodulatory effects of soybean, canola, and fish oils, highlighting the relevance of nutrigenomics in understanding gene-diet interactions in pigs.

PMID:40467700 | DOI:10.1038/s41598-025-04273-0

Trends Plant Sci. 2025 Jun 3:S1360-1385(25)00137-2. doi: 10.1016/j.tplants.2025.05.009. Online ahead of print.

ABSTRACT

Two recent papers provide new insights into plant immunity. Li et al. identified an evolutionary pattern in which reduced pathogen pressure leads to a convergent reduction of immune receptors. While Guo et al. uncovered recurrent losses of nucleotide-binding site-leucine-rich repeat receptors (NLRs), through intergenomic synteny analysis, revealing the molecular mechanism for immune receptor reduction.

PMID:40467438 | DOI:10.1016/j.tplants.2025.05.009

Dent Mater. 2025 Jun 3:S0109-5641(25)00641-4. doi: 10.1016/j.dental.2025.05.006. Online ahead of print.

ABSTRACT

OBJECTIVES: To assess the antiproteolytic effect of EGCG-methacrylate monomers and its inhibitory effect on gelatinolytic activity in the hybrid layer. Also, to investigate the effect of an adhesive material functionalized with EGCG-methacrylate monomers on immediate and long-term dentin-resin bond strength.

METHODS: Neat EGCG (E0) was reacted with three different ratios of methacryloyl ester and dissolved in ethyl acetate to obtain EGCG-methacrylates with hydroxyl functionalization at 33 % (M-E33), 67 % (M-E67) and 100 % (M-E100) levels. Resin composite blocks were built on human dentin surfaces using self-etching adhesive containing E0, M-E33, M-E67, and M-E100 at 1 wt%. Demineralized human dentin disks were immersed in deionized water (DW) or lactic acid (LA) and subsequently treated with DW, acetone (as controls), E0, M-E33, M-E67 and M-E100 diluted in acetone. Concentrations of solubilized type I collagen C-terminal (CTX and ICTP) and N-terminal (NTX) telopeptides were determined from 7-day extracts of dentin matrix specimens by ELISA assays. In situ zymography of adhesive-dentin interface slices from restored teeth was performed by confocal microscope after 24 h dentin treatment. Microtensile bond strength (µTBS) and failure pattern were evaluated after 24 h and 6 months. Data were analyzed using two-way ANOVA and Tukey post hoc test (p < 0.05).

RESULTS: All experimental groups statistically reduced the release of solubilized telopeptides from dentin samples in DW and LA. E0 and M-E100 incorporated into the adhesive system reduced the gelatinolytic activity within the hybrid layer. The lowest µTBS values for restored teeth were observed for E0 and M-E100 groups, after 24 h and 6 months, respectively. The most prevalent failure observed was classified as type 4, except for M-E100.

SIGNIFICANCE: EGCG-methacrylate monomers effectively protected collagen from degradation. When incorporated into adhesive systems, EGCG-methacrylates reduced gelatinolytic activity within the hybrid layer, and did not affect immediate and long-term bond strength values of restorations.

PMID:40467428 | DOI:10.1016/j.dental.2025.05.006

Mech Ageing Dev. 2025 Jun 2:112081. doi: 10.1016/j.mad.2025.112081. Online ahead of print.

ABSTRACT

Sex is a key determinant of human phenotype, with males and females exhibiting distinct anthropometric and metabolic features that influence disease susceptibility. This study investigated sex-specific metabolic differences in late middle-aged adults without diagnosed metabolic diseases, both in the fasting state and during an oral glucose tolerance test (OGTT). Using data from the NutriTech project, we analyzed plasma metabolomic responses during the OGTT, along with detailed assessments of body composition and fasting clinical parameters. Females exhibited 28% greater total adipose tissue, mainly subcutaneous, whereas males had more intra-abdominal fat and higher energy expenditure. Females showed elevated fasting levels of fatty acids-particularly very-long-chain fatty acids- leptin, and adiponectin. Males had slightly higher fasting glycemia (~ 5%) and a more pronounced glycemic increase during the OGTT (17%), along with elevated insulin levels. In both fasting and postprandial states, males showed higher circulating levels (p<0.05) of aromatic and branched-chain amino acids (BCAA) and their catabolites. Conversely, females had higher sphingomyelins levels during fasting and throughout the OGTT, and increased postprandial levels of secondary bile acids (p<0.05). These sex-specific metabolic features in late middle-aged adults may enhance our understanding of metabolic disease risk and support the development of more targeted prevention strategies. CLINICAL TRIAL REGISTRATION NUMBER: NCT01684917.

PMID:40467008 | DOI:10.1016/j.mad.2025.112081

Dev Biol. 2025 Jun 2:S0012-1606(25)00145-9. doi: 10.1016/j.ydbio.2025.05.022. Online ahead of print.

ABSTRACT

Vertebrate development from an egg to a complex multi-cell organism is accompanied by multiple phases of genome-scale changes in the repertoire of proteins and their post-translational modifications. While much has been learned at the RNA level, we know less about changes at the protein level. In this paper, we present a deep analysis of changes of ∼15,000 proteins and ∼11,500 phospho-sites at 11 developmental time points in Xenopus laevis embryos ranging from the stage VI oocyte to juvenile tadpole. We find that the most dramatic changes to the proteome occur during the transition to functional organ systems, which occurs as the embryo becomes a tadpole. At that time, the absolute amount of non-yolk protein increases two-fold, and there is a shift in the balance of expression from proteins regulating gene expression to receptors, ligands, and proteins involved in cell-cell and cell-environment interactions. Between the early and late tadpole, the median increase for membrane and secreted proteins is substantially higher than that of nuclear proteins. To begin to appreciate changes at the post-translational level, we have measured quantitative phospho-proteomic data across the same developmental stages. In contrast to the significant protein changes that are concentrated at the end of the time series, the most significant phosphorylation changes are concentrated in the very early stages of development. A clear exception are phosphorylations of proteins involved in gene expression: these increase just after fertilization, with patterns that are highly correlated with the underlying protein changes. To facilitate the interpretation of this unique phospho-proteome data set, we created a pipeline for identifying homologous human phosphorylations from the measured Xenopus phospho-proteome. Collectively, our data reveal multiple coordinated transitions in protein and phosphorylation profiles, reflecting distinct developmental strategies and providing an extensive resource to further explore developmental biology at the proteomic and phospho-proteomic levels.

PMID:40466852 | DOI:10.1016/j.ydbio.2025.05.022

Spectrochim Acta A Mol Biomol Spectrosc. 2025 Jun 2;343:126507. doi: 10.1016/j.saa.2025.126507. Online ahead of print.

ABSTRACT

BACKGROUND: As the number of people with dementia grows, it becomes even more important to improve how it is diagnosed and treated. Metabolomics, the study of small molecule metabolites in biological systems, helps us learn a lot about dementia by revealing changes in the brain's systems through blood plasma research.

OBJECTIVE: The objective of this study is to clarify the changes in metabolism that are linked to dementia. This will be achieved by using metabolic fingerprinting techniques on human plasma to differentiate between patients with dementia and individuals who have normal cognitive function.

METHODS: This study used data from the Birjand Longitudinal Aging Study and high-tech Raman Spectroscopy, as well as multivariate statistical methods such as PCA and OPLS-DA. The study looked at 34 people with dementia and 34 people who did not have any cognitive damage.

RESULTS: Metabolic fingerprints distinguished two groups with extremely distinct metabolic characteristics. Main findings demonstrate that oxidative stress and energy metabolism metabolites have changed significantly. OPLS-DA distinguished healthy and dementia samples with high accuracy and sensitivity. Both the expected high model accuracy and the clear score plot split confirmed this.

CONCLUSION: The metabolic deviations detected offer a deeper understanding of the biochemical processes associated with dementia. These results enhance our understanding of dementia-related biochemical changes and underscore the exploratory potential of Raman-based metabolomic fingerprinting as a complementary, non-invasive approach for identifying broader functional group-level alterations.

PMID:40466487 | DOI:10.1016/j.saa.2025.126507

Talanta. 2025 May 22;296:128326. doi: 10.1016/j.talanta.2025.128326. Online ahead of print.

ABSTRACT

Immunoglobulin G (IgG) subclasses glycosylation reflects the progression of colorectal cancer (CRC). Precise identification of IgG subclass-specific glycopeptides is a critical step. However, it is still hard to achieve a one-step mass spectrometry (MS) since all four subclasses of IgG (IgG1- IgG4) contain several possible N-glycans in the Fc regions with a highly similar amino acid sequence. In this study, we set up a label-free workflow to quantify IgG subclass site-specific N-glycopeptides in a single run MS based on the poly (glycerol methacrylate) @ chitosan (PGMA@CS) nanomaterial enrichment. The nanomaterial was used to purify glycopeptides effectively and the LC-SCE-HCD-MS/MS was used to obtain the peptide and glycan fragment in one single run MS. Through our workflow, all four subtypes of IgG glycopeptides were distinguished. For the first time, a total of 89 biantennary IgG subclass-specific N-glycopeptides were detected for quantification in 50 CRC patients and 66 healthy individuals. We have found that the decrease in galactosylation, fucosylation of sialylated glycans, sialylation of IgG2-Fc was associated with colorectal cancer. The results demonstrated that the glycopeptides of IgG-Fc are associated with CRC and potential to serve as noninvasive biomarkers. And it implies that the workflow can also accommodate the precise high-throughput identification of intact N-glycopeptides at the proteome scale.

PMID:40466447 | DOI:10.1016/j.talanta.2025.128326

PLoS One. 2025 Jun 4;20(6):e0323880. doi: 10.1371/journal.pone.0323880. eCollection 2025.

ABSTRACT

BACKGROUND: Metabolic dysfunction-associated steatotic liver disease (MASLD) is a major risk factor for liver cirrhosis, yet effective prevention or treatment strategies remain limited. To address this, we utilized a signature-based in silico drug repositioning approach to identify potential therapeutics for MASLD that may reduce the risk of cirrhosis.

METHODS: We analyzed gene expression datasets to identify differentially expressed genes (DEGs) in MASLD and matched them to candidate drugs using L1000CDS2. We further validated potential drugs by cross-referencing with prescription data from the Korea National Health Insurance Service (NHIS). Participants who underwent health screenings between 2013 and 2014 were included. MASLD was diagnosed in individuals with hepatic steatosis (fatty liver index ≥60) and at least one cardiometabolic risk factor.

RESULTS: We identified 11 drug candidates and analyzed 49,555 MASLD patients (mean age: 63.0 years, SD: 8.6). Atenolol (SHR: 0.81; 95% CI: 0.72-0.92; P < 0.001), isosorbide dinitrate (SHR: 0.82; 95% CI: 0.73-0.93; P = 0.001), and valsartan (SHR: 0.52; 95% CI: 0.45-0.60; P < 0.001) were associated with a reduced risk of cirrhosis. Conversely, amlodipine-based combinations (SHR: 1.24; 95% CI: 1.11-1.39; P < 0.001), torasemide (SHR: 1.39; 95% CI: 1.24-1.56; P < 0.001), and valsartan-based combinations (SHR: 1.22; 95% CI: 1.09-1.37; P < 0.001) were linked to an increased risk.

CONCLUSIONS: Our findings suggest that antihypertensive drugs such as atenolol and isosorbide dinitrate may protect MASLD patients from cirrhosis, providing valuable insights for clinical applications and treatment strategies.

LIMITATIONS: This study is limited to drugs registered in the Korean NHIS, potentially excluding other relevant candidates. Additionally, the absence of dietary and genetic data in the NHIS database may introduce residual confounding. Lastly, as the study population consists solely of Korean adults, the findings may not be generalizable to other populations.

PMID:40465795 | DOI:10.1371/journal.pone.0323880

Sci Transl Med. 2025 Jun 4;17(801):eadq1810. doi: 10.1126/scitranslmed.adq1810. Epub 2025 Jun 4.

ABSTRACT

Hydrocephalus (HC) is a failure of brain and cerebrospinal fluid (CSF) homeostasis often associated with dilation of the CSF-filled ventricles (ventriculomegaly). Hallmarks of HC include aberrant CSF dynamics, neural stem cell dysfunction resulting in impaired neurogenesis and corticogenesis, biomechanical instability at the brain-CSF interface, and disrupted synaptogenesis and neural circuitry. Pleiotropic mechanisms, including genetic and environmental insults to the brain, contribute to neurodevelopmental comorbidities. Hypothesis generation from genome-wide, single-cell multi-omic analyses coupled to experimental validation using induced pluripotent stem cell-derived cerebral organoids will refine molecular classification of HC subtypes and may lead to precision-based surgical and pharmacologic treatments.

PMID:40465691 | DOI:10.1126/scitranslmed.adq1810

Sci Adv. 2025 Jun 6;11(23):eadt0539. doi: 10.1126/sciadv.adt0539. Epub 2025 Jun 4.

ABSTRACT

We addressed the underrepresentation of non-European populations in genome-wide association studies (GWASs) by building HiGenome, a large-scale genetic resource for the Taiwanese Han population. Using a custom genotyping array, we integrated deidentified electronic medical records (2003 to 2021) with genomic data to enable GWASs, phenome-wide association studies, and polygenic risk score (PRS) analysis. Among 413,000 participants, 323,397 passed ancestry and quality control filtering. GWASs covered 1085 traits, focusing on diseases prevalent in Taiwan such as type 2 diabetes, chronic kidney disease, gout, and alcoholic liver damage. PRSs were calculated for 238 traits, with the strongest associations observed in musculoskeletal disorders. Incorporating PRS into clinical practice supports early risk prediction and personalized prevention. To further expand translational value, we also conducted pharmacogenomic analysis and human leukocyte antigen typing. HiGenome offers a large-scale genetic and clinical dataset from the Taiwanese Han population, supporting population-specific analyses and precision medicine development in East Asia. The hospital-based design enables continuous follow-up and longitudinal data expansion.

PMID:40465716 | DOI:10.1126/sciadv.adt0539

Clin Infect Dis. 2025 Jun 4;80(5):e65-e77. doi: 10.1093/cid/ciaf009.

ABSTRACT

Transformative changes in care for people with cystic fibrosis (CF; pwCF) have occurred, including most recently, the widespread use of CF transmembrane regulator modulator therapy. These novel therapies improve lung function, decrease pulmonary exacerbations, increase life expectancy, and improve quality of life. Changes in the CF population have also occurred. There are now more adults than children living with CF. A growing proportion of pwCF are black and/or Hispanic, many of whom are ineligible for modulator therapy due to their CF transmembrane regulator mutations, which may further exacerbate disparities in healthcare. Management of pulmonary exacerbations-including shared decision making between pwCF and providers, the limitations of antimicrobial susceptibility testing to predict treatment response, and the role of antimicrobial stewardship-is increasingly recognized by the CF community. Collaborations among infectious diseases specialists, antimicrobial stewards, CF care teams, and clinical microbiology laboratories are increasingly needed to optimize these newer care paradigms.

PMID:40465484 | DOI:10.1093/cid/ciaf009

Clin Infect Dis. 2025 Jun 4;80(5):939-941. doi: 10.1093/cid/ciaf010.

NO ABSTRACT

PMID:40465483 | DOI:10.1093/cid/ciaf010

JMIR Res Protoc. 2025 Jun 4;14:e67875. doi: 10.2196/67875.

ABSTRACT

BACKGROUND: Atrial fibrillation (AF) after radiotherapy (RT) in patients with breast cancer (BC) is a relatively new and understudied topic. AF can increase the risk of stroke and other serious cardiovascular complications, compromising patients' quality of life and survival. Screening of AF, both asymptomatic and symptomatic forms, is therefore essential for optimal management.

OBJECTIVE: The aim of the Watch Your Heart After Radiotherapy for Breast Cancer (WATCH) study is to assess the incidence of AF (symptomatic or asymptomatic) occurring throughout a 5-year follow-up after RT and to investigate whether cardiac radiation exposure is associated with the occurrence of such events.

METHODS: WATCH is a cohort study that will include 200 patients over 65 years old, treated with RT for BC 5 years before inclusion and without a history of AF. Cross-sectional screening for AF at the time of the scheduled 5-year post-RT visit will be conducted by recording data from a Withings ScanWatch smartwatch for 1 month, confirmed by an electrocardiogram (ECG), and validated by a physician. In addition, a transthoracic echocardiography (TTE) will be performed, providing a comprehensive assessment of cardiac structures, and allowing us to investigate the underlying etiology and assess the risk of complications. Patients' medical records will provide retrospective information about the timing and risk factors for the occurrence of AF and other arrhythmias and cardiac diseases during the 5 years following RT. The development of deep learning algorithms for autosegmentation analysis of potentially critical substructures for the occurrence of AF, including cardiac chambers, the sinoatrial node, the atrioventricular node, coronary arteries, and pulmonary veins (PVs), will produce dosimetry linked to previous RT treatment for all contoured structures.

RESULTS: Enrollment started in October 2023 and will continue until mid-2026 to include 200 patients, which will ensure an 80% statistical power to detect a significant difference in AF incidence around 15% for the group of patients moderately exposed (<75th percentile of the mean heart radiation dose) and 25% for the group of patients highly exposed (>75th percentile of the mean heart radiation dose). The results are expected by the end of 2026.

CONCLUSIONS: This study will contribute to generating new knowledge on AF after RT for BC and help considering the inclusion of AF screening into routine clinical practice for these patients. Identifying the dose-risk associations would improve RT delivery protocols to limit the occurrence of different forms of AF and, if necessary, initiate appropriate treatment.

TRIAL REGISTRATION: ClinicalTrials.gov NCT06073509; clinicaltrials.gov/study/NCT06073509?id=NCT06073509&rank=1.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/67875.

PMID:40466092 | DOI:10.2196/67875

Crit Care Explor. 2025 Jun 4;7(6):e1276. doi: 10.1097/CCE.0000000000001276. eCollection 2025 Jun 1.

ABSTRACT

IMPORTANCE: Sepsis is a major cause of morbidity and mortality, with early intervention shown to improve outcomes. Predictive modeling and artificial intelligence (AI) can aid in early sepsis recognition, but there remains a gap between algorithm development and clinical use. Despite the importance of user experience in adopting clinical predictive models, few studies have focused on provider acceptance and feedback.

OBJECTIVES: To evaluate healthcare worker perception and acceptance of a deep learning sepsis prediction model in the emergency department (ED).

DESIGN, SETTING, AND PARTICIPANTS: COnformal Multidimensional Prediction Of SEpsis Risk (COMPOSER), a deep learning algorithm, is used at two EDs of a large academic medical center to predict sepsis before clear clinical presentation. An internally developed survey following the Checklist for Reporting Results of Internet E-Surveys was distributed to team members who received a COMPOSER alert.

ANALYSIS: Mann-Whitney U testing was performed on results stratified by provider experience.

RESULTS: A total of 114 responses were received: 76 from doctors of medicine/doctors of osteopathic medicine, 34 from registered nurses, and four from nurse practicioners/physician assistants. Of these, 53% were from providers with fewer than 5 years of experience. Seventy-seven percent of respondents had a positive or neutral perception of the alert's usefulness. Providers with 0-5 years of experience were more likely to expect sepsis after the alert (p = 0.021) and found the alert more useful (p = 0.016) compared with those with 6+ years of experience. Additionally, physicians with 0-5 years of experience were more likely to say the alert changed their patient management (p = 0.048).

CONCLUSIONS: Less experienced providers were more likely to perceive benefit from the alert, which was overall received favorably. Future AI implementations might consider tailored alert patterns and education to enhance reception and reduce fatigue.

PMID:40466050 | DOI:10.1097/CCE.0000000000001276

Clin Transplant. 2025 Jun;39(6):e70201. doi: 10.1111/ctr.70201.

ABSTRACT

INTRODUCTION: Sarcopenia, or the loss of muscle quality and quantity, has been associated with poor clinical outcomes in liver transplantation such as infection, increased length of stay, and increased patient mortality. Abdominal computed tomography (CT) scans are utilized to measure patient core musculature as a measurement of sarcopenia. Methods to extract information on core body musculature can be through either freehand region-of-interest (ROI) or machine learning algorithms to quantitate total body muscle within a given area. This study directly compares these two collection methods leveraging length of stay (LOS) outcomes previously found to be associated with freehand ROI measurements.

METHODS: A total of 50 individuals were included who underwent liver transplantation from our single center between January 1, 2016, and May 30, 2021, and had a non-contrast abdominal CT scan within 6-months of surgery. CT-derived skeletal muscle measures at the third lumbar vertebrae were obtained using freehand ROI and an automated deep learning system.

RESULTS: Correlation analysis of freehand psoas muscle measures, psoas area index (PAI) and mean Hounsfield units (mHU), were significantly correlated to the automated deep learning system's total skeletal muscle measures at the level of the L3, skeletal muscle index (SMI) and skeletal muscle density (SMD), respectively (R2 = 0.4221; p value < 0.0001; R2 = 0.6297; p value < 0.0001). The automated deep learning model's SMI predicted ∼20% of the variability (R2 = 0.2013; hospital length of stay) while the PAI variable only predicted about 10% of the variability (R2 = 0.0919; total healthcare length of stay) of the length of stay variables. In contrast, both the freehand ROI mHU and the automated deep learning model's muscle density variables were associated with ∼20% of the variability in the inpatient length of stay (R2 = 0.2383 and 0.1810, respectively) and total healthcare length of stay variables (R2 = 0.2190 and 0.1947, respectively).

CONCLUSION: Sarcopenia measurements represent an important risk stratification tool for liver transplantation outcomes. For muscle sarcopenia assessment association with LOS, freehand measures of sarcopenia perform similarly to automated deep learning system measurements.

PMID:40465826 | DOI:10.1111/ctr.70201

Br J Dermatol. 2025 Jun 4:ljaf212. doi: 10.1093/bjd/ljaf212. Online ahead of print.

ABSTRACT

BACKGROUND: Mycosis fungoides (MF) is the most common type of cutaneous T-cell lymphoma, and early-stage MF is difficult to differentiate from erythematous inflammatory disease. Except biopsy, non-invasive information such as patient's basic information, clinical images and dermoscopic images is of great significance for early diagnosis of MF. However, there is still a lack of diagnosis models based on convolutional neural network that can utilize the above multimodal information.

OBJECTIVES: We aim to develop an artificial intelligence (AI) deep learning model based on multimodal information, verify its classification efficiency, and construct an AI-aided early diagnostic model of MF and inflammatory skin diseases for dermatologists.

METHODS: This is a single center retrospective study based on multimodal information including clinical information, clinical images, and dermoscopic images. A total of 1157 cases of MF and inflammatory diseases were collected, including 2452 clinical images, 6550 dermoscopic images and corresponding clinical data. RegNetY-400MF was selected as the feature extractors in the study.

RESULTS: AI model demonstrates higher levels of total accuracy, precision, sensitivity, and specificity in classification of MF and other inflammatory skin diseases compared to the participating dermatologists. A significant enhancement was noticed in average accuracy, sensitivity, and specificity for MF and inflammatory diseases within the Doctor+AI group, with values of 82.94%, 86.16%, and 96.45% respectively, compared to 71.52%, 74.56%, and 94.06% within the Doctor-only group. The more accurately diagnosis of each disease was also achieved by the multi-classification model.

CONCLUSIONS: These results indicate that our AI model has a significantly strong discriminative ability to assist doctors in improving diagnostic accuracy of early-stage MF and common inflammatory skin diseases.

PMID:40465821 | DOI:10.1093/bjd/ljaf212

PLoS One. 2025 Jun 4;20(6):e0325491. doi: 10.1371/journal.pone.0325491. eCollection 2025.

ABSTRACT

This study focuses on the impact of learning experience on college students' deep learning of English and the chain-mediated effects of motivation and strategy. In the context of globalization, English is crucial for university students, but traditional teaching models often neglect the role of learning experience in deep learning. Deep learning emphasizes critical understanding, creative application and long-term memory construction, which is particularly important for English learning. Learning experience covers affective, cognitive and behavioral responses, and influences learning attitudes and effects, but there are fewer studies on its impact on college students' deep learning of English and the related mechanisms. In this study, college students of different genders, ages, educational backgrounds and academic achievement levels were selected as samples, and learning experience, motivation, learning strategies and deep learning were comprehensively assessed by well-designed scales and statistically analyzed with the help of SPSS and AMOS software. The results of the study show that learning experience has a significant positive effect on English deep learning, and motivation and learning strategies play an important chain mediating role. Specifically, learning experience enhances motivation, which in turn promotes the use of learning strategies and ultimately improves English deep learning. This study validates for the first time the chain mediation model of "learning experience→learning motivation→learning strategies→deep learning"in the field of English language learning, which provides a new perspective for understanding the intrinsic mechanism of college students' English language learning and enriches related research. In practice, it provides specific guidance for English teaching, and teachers can enhance students' English deep learning by optimizing learning experience, stimulating learning motivation and guiding the use of learning strategies. However, there are some limitations in this study, such as the limited sample scope and the use of a cross-sectional design, etc. Future studies can expand the sample scope, adopt a longitudinal research design, and further explore other potential mediating variables.

PMID:40465789 | DOI:10.1371/journal.pone.0325491

PLoS One. 2025 Jun 4;20(6):e0322166. doi: 10.1371/journal.pone.0322166. eCollection 2025.

ABSTRACT

With the development of smart wearable devices and deep learning (DL) technology, the monitoring and analysis of daily sports activities of teenagers face new opportunities. At present, traditional CNN (Convolutional Neural Network) models are mostly used for recognition in daily sports activities. It is difficult to capture the temporal relationship between action sequences, and the ability to express important features is weak, resulting in poor recognition accuracy. This paper took badminton as the object, based on the VGG16 (Visual Geometry Group 16) model, and adopted the advantages of the bidirectional learning time series information of the BiLSTM (Bidirectional Long Short-Term Memory) model and the channel and regional feature representation of the CBAM (Convolutional Block Attention Module) module to verify and apply the recognition of badminton movements in daily sports for teenagers. The study first built and optimized the baseline model VGG16, removed the last three fully connected layers, and used VGG16 to extract the deep features of each frame of video image and output feature maps. The CBAM module was then embedded after the last convolutional layer of the VGG16 network, and the feature maps optimized by CBAM were flattened into a time series input vector. Finally, the BiLSTM model is introduced, and the CBAM and BiLSTM are connected in a cascade manner to capture the information of the previous and next dependencies in the video frame sequence and output the action classification results of badminton. The experiment is based on the badminton training dataset in the public dataset Roboflow to explore the action recognition performance in badminton in daily sports activities of teenagers. Experimental results show that the recognition accuracy of the VGG16-BiLSTM-CBAM model has reached 0.98, which is 0.08 higher than the benchmark model VGG16, and F1 has reached 0.96. Experimental results show that combined with the DL model VGG19 and the sequential model BiLSTM, the attention CBAM module can significantly improve the performance of action recognition in youth badminton, promote the safe conduct of sports activities, and provide a good reference for incorrect postures.

PMID:40465756 | DOI:10.1371/journal.pone.0322166