J Agric Food Chem. 2025 May 25. doi: 10.1021/acs.jafc.5c02750. Online ahead of print.

ABSTRACT

The rhizomes of Alpinia galanga have a characteristic aroma, the molecular basis of which has not yet been fully clarified. Application of gas chromatography-olfactometry and aroma extract dilution analysis to the volatiles isolated from fresh A. galanga rhizomes by solvent-assisted flavor evaporation led to the detection of 43 odorants with flavor dilution factors of 1-4096. Enantiodifferentiation increased the number to 47. The structures of 36 odorants were elucidated, 18 of which were previously unknown in A. galanga, and 21 were shown to be present in concentrations above the odor threshold concentration (OTC). High odor activity values (OAVs; concentration/OTC) were particularly obtained for 1,8-cineole (130000), (S)-galangal acetate (44000), and myrcene (8000). As proof of success, the characteristic aroma of A. galanga rhizome was reconstructed with 20 major odorants in their natural concentrations. The results will contribute to a deeper molecular understanding of the aroma of Alpinia species and serve as the basis to study aroma changes during processing and culinary use of A. galanga rhizome.

PMID:40413641 | DOI:10.1021/acs.jafc.5c02750

Pharmacogenomics J. 2025 May 24;25(3):15. doi: 10.1038/s41397-025-00369-y.

ABSTRACT

Childhood cancer survivors (CCS) often suffer from cardiac disease (CD) after treatment that included anthracycline and radiotherapy involving the heart. However, the variability in CD occurrence cannot be explained solely by these treatments, suggesting the existence of genetic predisposition. We conducted a systematic review searching on Medline-PubMed and Scopus, to identify studies reporting associations between genetic factors and CD in CCS. We included studies published up to 11 April 2023, with no lower limit, and assessed the quality of genetic associations by the Q-genie tool. As a result, 20 studies were included (15 case-control and five cohorts), revealing several genes and variants associated with cardiomyopathy, among which, SLC28A3-rs7853758, RARG-rs2229774, P2RX7-rs208294 and P2RX7-rs3751143 variants gave the most consistent findings. This review highlights the necessity to establish a set of clinically useful genes and variants to identify patients most at risk of developing cardiomyopathy, and to implement monitoring and prevention strategies.

PMID:40413218 | DOI:10.1038/s41397-025-00369-y

Contraception. 2025 May 22:110969. doi: 10.1016/j.contraception.2025.110969. Online ahead of print.

ABSTRACT

Abysmal sexual and reproductive health (SRH) outcomes in the United States persist due to multiple factors, including diminishing SRH care access and inequities in care for socially or economically marginalized populations. Digital innovations have the potential to address gaps in SRH care as scalable, low-cost, patient-centered solutions that supplement the formal healthcare system. Our multidisciplinary team has developed a suite of patient-facing digital tools to help address suboptimal SRH care delivery for marginalized individuals capable of pregnancy, including those with chronic medical conditions. These tools-MyPath for reproductive preferences, prepregnancy health, and contraception; MyVoice for SRH needs of people with rheumatic/autoimmune disease or cystic fibrosis; MyDecision for tubal sterilization; and MyHealthyPregnancy for tailored pregnancy support-are guided by principles of community engagement, person-centeredness, and health equity. In the wake of the Dobbs v. Jackson Women's Health Organization 2022 Supreme Court decision overturning federal abortion protections, as well as the rapidly shifting policy landscape under the current administration, there are new considerations for use and implementation of digital SRH tools. In this commentary, we draw directly from lessons learned in our work to discuss emerging concerns related to data privacy and pregnancy criminalization, trust in healthcare providers and systems, and research. We then propose recommendations for researchers seeking to create, implement, and evaluate these tools with the goal of safeguarding reproductive autonomy and achieving health equity in this new policy context.

PMID:40412590 | DOI:10.1016/j.contraception.2025.110969

Bull Math Biol. 2025 May 24;87(7):85. doi: 10.1007/s11538-025-01455-9.

ABSTRACT

T Cell Engager (TCE)s are an exciting therapeutic modality in immuno-oncology that acts to bypass antigen presentation and forms a direct link between cancer and immune cells in the Tumor Microenvironment (TME). TCEs are efficacious only when the drug is bound to both immune and cancer cell targets. Therefore, approaches that maximize the formation of the drug-target trimer in the TME are expected to increase the drug's efficacy. In this study, we quantitatively investigate how the concentration of ternary complex and its biodistribution depend on both the targets' specific properties and the design characteristics of the TCE, and specifically on the binding kinetics of the drug to its targets. A simplified mathematical model of drug-target interactions is considered here, with insights from the "three-body" problem applied to the model. Parameter identifiability analysis performed on the model demonstrates that steady state data, which is often available at the early pre-clinical stages, is sufficient to estimate the binding affinity of the TCE molecule to both targets. We used the model to analyze several existing antibodies, both clinically approved and under development, to explore their common kinetic features. The manuscript concludes with an assessment of a full quantitative pharmacology model that accounts for drug disposition into the peripheral compartment.

PMID:40413295 | DOI:10.1007/s11538-025-01455-9

Sci Rep. 2025 May 24;15(1):18085. doi: 10.1038/s41598-025-02584-w.

ABSTRACT

This study aims to develop a novel antibacterial hydrogel scaffold composed of gelatin (Gel), amniotic membrane extract (AME), and aminophylline (AMP) for neural regeneration. We investigate its ability to sustain AMP release, inhibit bacterial growth, and promote neural differentiation of human Wharton's jelly mesenchymal stem cells (hWJMSCs) via the CREB pathway, addressing unmet needs in neural tissue engineering. The composite hydrogels were synthesized and characterized using various methods and techniques, including X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), Fourier transform infrared spectroscopy (FTIR), porosity, contact angle, water uptake, thermogravimetric analysis (TGA), biodegradation, tensile strength, drug release, and antibacterial activity. Biocompatibility tests (MTT assay, AO/EB staining) confirmed > 95% viability of hWJMSCs over six days and their differentiation to the neural cells was analyzed through immunocytochemistry (ICC) staining and real-time reverse transcription-polymerase chain reaction (RT-PCR) at different time points. The results demonstrate the successful synthesis of porous hydrogels with desirable properties, including hydrophilicity, thermal stability, biodegradability, and mechanical strength. The hydrogels support the sustained release of AMP (53.18% over 336 h) and exhibit antibacterial activity against Pseudomonas aeruginosa (90.52 ± 0.26%) and Staphylococcus aureus (93.06 ± 0.34%) due to the presence of penicillin and streptomycin (P-S) antibiotics. The biocompatibility results show that the hydrogels do not have a cytotoxic effect on the viability of human WJMSCs. The neural differentiation of human WJMSCs seeded on surface hydrogels was confirmed by evaluating specific neural markers at both protein and gene levels. In conclusion, the new antibacterial gel-based hydrogel can support the release of AMP and after further evaluation, can be introduced as a new candidate for neural repair applications.

PMID:40413259 | DOI:10.1038/s41598-025-02584-w

Sci Data. 2025 May 24;12(1):858. doi: 10.1038/s41597-025-05218-y.

ABSTRACT

Managed Aquifer Recharge (MAR) systems, which intentionally replenish groundwater aquifers with excess water, are critical for addressing water scarcity exacerbated by demographic shifts and climate variability. To date, little is known about the functional diversity of the soil microbiome at different soil depth inhabiting agricultural soils used for MAR. Knowing the functional diversity is pivotal in regulating nutrient cycling and maintaining soil health. Metagenomics, particularly Metagenome-Assembled Genomes (MAGs), provide a powerful tool to explore the diversity of uncultivated soil microbes, facilitating in-depth investigations into microbial functions. In a field experiment conducted in a California vineyard, we sequenced soil DNA before and after water application of MAR. Through this process, we assembled 146 medium and 14 high-quality MAGs, uncovering a wide array of archaeal and bacterial taxa across different soil depths. These findings advance our understanding of the microbial ecology and functional diversity of soils used for MAR, contributing to the development of more informed and sustainable land management strategies.

PMID:40413198 | DOI:10.1038/s41597-025-05218-y

NPJ Syst Biol Appl. 2025 May 24;11(1):55. doi: 10.1038/s41540-025-00521-1.

ABSTRACT

The genome-scale model of metabolism and gene expression (ME-model) for Pseudomonas putida KT2440, iPpu1676-ME, provides a comprehensive representation of biosynthetic costs and proteome allocation. Compared to a metabolic-only model, iPpu1676-ME significantly expands on gene expression, macromolecular assembly, and cofactor utilization, enabling accurate growth predictions without additional constraints. Multi-omics analysis using RNA sequencing and ribosomal profiling data revealed translational prioritization in P. putida, with core pathways, such as nicotinamide biosynthesis and queuosine metabolism, exhibiting higher translational efficiency, while secondary pathways displayed lower priority. Notably, the ME-model significantly outperformed the M-model in alignment with multi-omics data, thereby validating its predictive capacity. Thus, iPpu1676-ME offers valuable insights into P. putida's proteome allocation and presents a powerful tool for understanding resource allocation in this industrially relevant microorganism.

PMID:40413180 | DOI:10.1038/s41540-025-00521-1

Nat Commun. 2025 May 24;16(1):4838. doi: 10.1038/s41467-025-60067-y.

ABSTRACT

Public untargeted metabolomics data is a growing resource for metabolite and phenotype discovery; however, accessing and utilizing these data across repositories pose significant challenges. Therefore, here we develop pan-repository universal identifiers and harmonized cross-repository metadata. This ecosystem facilitates discovery by integrating diverse data sources from public repositories including MetaboLights, Metabolomics Workbench, and GNPS/MassIVE. Our approach simplified data handling and unlocks previously inaccessible reanalysis workflows, fostering unmatched research opportunities.

PMID:40413169 | DOI:10.1038/s41467-025-60067-y

J Immunother Cancer. 2025 May 24;13(5):e011404. doi: 10.1136/jitc-2024-011404.

ABSTRACT

BACKGROUND: Immune checkpoint blockers (ICBs) have revolutionized cancer therapy, yet they remain largely ineffective in treating pancreatic ductal adenocarcinoma (PDAC). Moreover, ICBs can cause severe immune-related adverse events (irAEs), including fatal cardiac toxicity. Finally, obesity is a risk factor in PDAC that may differentially modulate ICB efficacy in a malignancy-dependent manner.

METHODS: We investigated the mechanisms underlying irAEs induced by dual ICB therapy and sought to identify strategies to mitigate them while improving ICB efficacy in the obese setting. To this end, we used a clinically relevant mouse model that integrated key features of human PDAC: (1) high-fat diet-induced obesity, (2) an orthotopic PDAC, and (3) a therapeutic regimen combining chemotherapy (FOLFIRINOX) with ICBs (α-programmed cell death protein-1 + α-cytotoxic T-lymphocyte associated protein-4 antibodies).

RESULTS: Obese mice developed cardiac irAEs and had elevated serum interleukin (IL)-1β levels after chemoimmunotherapy. IL-1β blockade not only prevented myocarditis and reduced cardiac fibrosis but also enhanced the antitumor efficacy of the combination of chemotherapy plus dual ICB therapy and significantly improved the overall survival of PDAC-bearing obese mice.

CONCLUSIONS: Our findings provide the rationale and compelling data to test a Food and Drug Administration-approved anti-IL-1β antibody in combination with chemotherapy and dual ICB therapy in patients with pancreatic cancer with obesity.

PMID:40413022 | DOI:10.1136/jitc-2024-011404

Neuroscience. 2025 May 22:S0306-4522(25)00394-X. doi: 10.1016/j.neuroscience.2025.05.032. Online ahead of print.

ABSTRACT

The western painted turtle is the most anoxia-tolerant tetrapod known, surviving ∼ 4 months at 3 °C without oxygen. In the mammalian brain, absence of oxygen leads to hyper-excitability and cell death within minutes. A major mechanism by which painted turtles survive anoxia is a large increase of γ-aminobutyric acid (GABA) in the brain leading to a dominating Cl- conductance that clamps membrane potential near the reversal potential of the GABAA receptor. Whole-cell GABAA receptor currents are known to increase with the onset of anoxia because of increased presynaptic GABA release, we hypothesized that GABAA receptor currents may also exhibit a large increase due to increased channel open time. To investigate this, we used cell-attached single-channel patch-clamp electrophysiological techniques to measure GABAA receptor open times (Popen) during a normoxic to anoxic transition in pyramidal neurons in turtle brain cortical sheets. GABAA receptor Popen significantly increased 13-fold with the onset of anoxia and was blocked by the inclusion of the protein kinase C (PKC) activator PMA phorbol-12-myristate-13-acetate. Indicating the receptor was regulated by covalent modification. To investigate the molecular evolutionary mechanisms underlying these adaptations, we used codon-based likelihood models to detect changes in selective pressure amongst the GABAA receptor subunit genes. We found positive selection in GABRB2 and GABRB3 at sites near their ligand binding interface, likely impacting channel kinetics associated with hypoxia-tolerance. The elucidation of the adaptations associated with increased hypoxia tolerance furthers our understanding of physiological adaptations to extreme low-oxygen environments.

PMID:40412545 | DOI:10.1016/j.neuroscience.2025.05.032

Dev Cell. 2025 May 21:S1534-5807(25)00286-2. doi: 10.1016/j.devcel.2025.04.026. Online ahead of print.

ABSTRACT

Disruptions in foregut morphogenesis can result in life-threatening conditions where the trachea and esophagus fail to separate, such as esophageal atresia (EA) and tracheoesophageal fistulas (TEFs). The developmental basis of these congenital anomalies is poorly understood, but recent genome sequencing reveals that de novo variants in intracellular trafficking genes are enriched in EA/TEF patients. Here, we confirm that mutation of orthologous genes in Xenopus disrupts trachea-esophageal separation similar to EA/TEF patients. The Rab11a recycling endosome pathway is required to localize Vangl-Celsr polarity complexes at the luminal cell surface where opposite sides of the foregut tube fuse. Partial loss of endosomal trafficking or Vangl-Celsr complexes disrupts epithelial polarity and cell division orientation. Mutant cells accumulate at the fusion point, fail to relocalize cadherin, and do not separate into distinct trachea and esophagus. These data provide insights into the mechanisms of congenital anomalies and general paradigms of tissue fusion during organogenesis.

PMID:40412385 | DOI:10.1016/j.devcel.2025.04.026

Neuron. 2025 May 23:S0896-6273(25)00309-5. doi: 10.1016/j.neuron.2025.04.026. Online ahead of print.

ABSTRACT

Intracerebral hemorrhage (ICH) is a prevalent disease with high mortality. Despite advances in clinical care, the prognosis of ICH remains poor due to an incomplete understanding of the complex pathological processes. To address this challenge, we generated single-cell-resolution spatiotemporal transcriptomic maps of the mouse brain following ICH. This dataset is the most extensive resource available, providing detailed information about the temporal expression of genes along with a high-resolution cellular profile and preserved cellular organization. We identified 100 distinct cell subclasses, 17 of which were found to play significant roles in the pathophysiology of ICH. We also report similarities and differences between two experimental ICH models and human postmortem ICH brain tissue. This study advances the understanding of the local and global responses of brain cells to ICH. It provides a valuable resource that can facilitate future research and aid the development of novel therapies for this devastating condition.

PMID:40412375 | DOI:10.1016/j.neuron.2025.04.026

Forensic Sci Int Genet. 2025 May 13;78:103301. doi: 10.1016/j.fsigen.2025.103301. Online ahead of print.

ABSTRACT

Short tandem repeats (STRs) are widely used in forensic genetics for individual identification. While traditional STR analysis relies on capillary electrophoresis (CE), next-generation sequencing (NGS) offers advantages such as full allelic sequence resolution, improving sensitivity and discrimination power. However, genetic variations in flanking regions can lead to discordant genotyping results between CE and NGS approaches, as well as among different analysis software. During the GEDNAP Proficiency Test 65, a genotyping discrepancy was observed at the D19S433 locus. The sample was analyzed using the ForenSeq® DNA Signature Prep Kit on the MiSeq FGx® Sequencing System, yielding a genotype 11.2,16 when analyzed with the ForenSeq™ Universal Analysis Software. This result differed from the 11.1,16 genotype reported by GEDNAP's CE-based results. Sequencing data from ForenSeq was further reanalyzed with STRait Razor Online and STRNaming, resulting in a genotype 11.1,16. Additional testing with three different CE kits (AmpFLSTR™ Identifiler™ Plus, NGM SElect™, and GlobalFiler™) produced a 16,16 genotype, leading to three different genotype assignments for the same sample. A 3-bp TCT deletion in the 5' flanking region of D19S433, located within the International Society for Forensic Genetics (ISFG) minimum reporting range was identified as the cause of these genotyping inconsistencies. Long-read sequencing with PacBio Sequel II technology confirmed that no additional variants were present in the primer binding regions, demonstrating that the TCT deletion alone was responsible for the discrepancies. This study highlights the impact of flanking region mutations on allele calling across different STR typing technologies and the lack of consensus in sequence analysis among bioinformatics pipelines, emphasizing the need to incorporate the ISFG minimum range in the regions sequenced and reported by NGS kits to ensure inter-laboratory and inter-kit consistency, ultimately minimizing discrepancies in forensic STR typing.

PMID:40412064 | DOI:10.1016/j.fsigen.2025.103301

Sci Rep. 2025 May 24;15(1):18102. doi: 10.1038/s41598-025-03114-4.

ABSTRACT

The purpose of this study is to analyze FAERS data to identify cases of drug-induced photosensitivity (DIP), examine demographic patterns, determine the drug classes involved, and highlight emerging trends in these reactions. Additionally, we explore potential signal drugs by mining the relevant reported data, aiming to provide insights for safer clinical use of medications. We reviewed the publicly available FAERS database from 2004 to 2023. Using DIP-related search terms such as "photosensitivity reaction," "polymorphic light eruption," or et al., we identified reports of DIP. The frequency and trends of these reports were then analyzed. Between 2004 and 2023, the FDA received 17,384,824 reports of adverse reactions, with 20,236 of these linked to DIP. After excluding cases with incomplete data on age, gender, or country of origin, the median patient age was 52 years (IQR = 66). Females comprised 55.71% of the cases (11,274), and 66.96% (12,459) of the reports originated from the United States. The top 45 drugs were responsible for 9,810 cases (48.48%). The three drug classes most commonly associated with DIP in the FAERS database were immunosuppressants, monoclonal antibodies, and antineoplastic agents. A disproportionality analysis of the top drugs revealed several newly identified drugs with signals for photosensitivity, including adalimumab, adapalene, secukinumab, and fingolimod. By analyzing publicly available FAERS data, we identified key themes and trends in DIP reactions. Immunosuppressants and monoclonal antibodies show mild trends in DIP occurrence. Additionally, adalimumab, adapalene, secukinumab, and fingolimod are novel drug signals of DIP.

PMID:40413262 | DOI:10.1038/s41598-025-03114-4

Australas J Dermatol. 2025 May 24. doi: 10.1111/ajd.14530. Online ahead of print.

NO ABSTRACT

PMID:40411268 | DOI:10.1111/ajd.14530

Small. 2025 May 24:e2503997. doi: 10.1002/smll.202503997. Online ahead of print.

ABSTRACT

With the advancement of intelligent and refined manufacturing, the demand for vibration sensors in smart equipment has surged. Traditional commercial vibration sensors and triboelectric nanogenerator (TENG)-based sensors are limited to basic amplitude and frequency recognition, failing to address both self-powering and diagnostic needs due to inherent design constraints. To overcome these limitations, this study introduces a novel mechanism combining interface dipole energy and vacuum level optimization in triboelectric materials to explain charge generation and separation under vibration. A TENG device with polydimethylsiloxane (PDMS)-encapsulated metal electrode is designed and developed, enabling the precise recognition of equipment operating status through vibration waveform analysis. By optimizing interface contact area and electron transfer capacity, the device achieves enhanced signal clarity and the introduction of subtler characteristics in the signal waveform. Furthermore, the integration of a deep learning algorithm enables high-resolution classification of vibration states with an accuracy of 98.3% approximately, achieving effective monitoring of the operating status of the jaw crusher and vibrating screen. This work not only verifies the feasibility of designing a self-powered vibration sensor but also demonstrates its potential for real-time monitoring and diagnostic applications in smart equipment.

PMID:40411864 | DOI:10.1002/smll.202503997

Med Biol Eng Comput. 2025 May 24. doi: 10.1007/s11517-025-03380-4. Online ahead of print.

ABSTRACT

In order to address the challenges posed by the large number of images acquired during wireless capsule endoscopy examinations and fatigue-induced leakage and misdiagnosis, a deep ensemble framework is proposed, which consists of CA-EfficientNet-B0, ECA-RegNetY, and Swin transformer as base learners. The ensemble model aims to automatically recognize four lesions in capsule endoscopy images, including angioectasia, bleeding, erosions, and polyps. All the three base learners employed transfer learning, with the inclusion of attention modules in EfficientNet-B0 and RegNetY for optimization. The recognition outcomes from the three base learners were subsequently combined and weighted to facilitate automatic recognition of multi-lesion images and normal images of the gastrointestinal (GI) tract. The weights were determined through the Bayesian optimization. The experiment collected a total of 8358 images of 281 cases at Shanghai East Hospital from 2017 to 2021. These images were organized and labeled by clinicians to verify the performance of the algorithm. The experimental results showed that the model achieved an accuracy of 84.31%, m-Precision of 88.60%, m-Recall of 79.36%, and m-F1-score of 81.08%. Compared to mainstream deep learning models, the ensemble model effectively improves the classification performance of GI diseases and can assist clinicians in making initial diagnoses of GI diseases.

PMID:40411689 | DOI:10.1007/s11517-025-03380-4

Eur Radiol. 2025 May 24. doi: 10.1007/s00330-025-11681-3. Online ahead of print.

ABSTRACT

PURPOSE: To investigate whether the deep learning reconstruction (DLR) combined with contrast-enhancement-boost (CE-boost) technique can improve the diagnostic quality of CT pulmonary angiography (CTPA) at low radiation and contrast doses, compared with routine CTPA using hybrid iterative reconstruction (HIR).

MATERIALS AND METHODS: This prospective two-center study included 130 patients who underwent CTPA for suspected pulmonary embolism. Patients were randomly divided into two groups: the routine CTPA group, reconstructed using HIR; and the dual-low dose CTPA group, reconstructed using HIR and DLR, additionally combined with the CE-boost to generate HIR-boost and DLR-boost images. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of pulmonary arteries were quantitatively assessed. Two experienced radiologists independently ordered CT images (5, best; 1, worst) based on overall image noise and vascular contrast. Diagnostic performance for PE detection was calculated for each dataset.

RESULTS: Patient demographics were similar between groups. Compared to HIR images of the routine group, DLR-boost images of the dual-low dose group were significantly better at qualitative scores (p < 0.001). The CT values of pulmonary arteries between the DLR-boost and the HIR images were comparable (p > 0.05), whereas the SNRs and CNRs of pulmonary arteries in the DLR-boost images were the highest among all five datasets (p < 0.001). The AUCs of DLR, HIR-boost, and DLR-boost were 0.933, 0.924, and 0.986, respectively (all p > 0.05).

CONCLUSION: DLR combined with CE-boost technique can significantly improve the image quality of CTPA with reduced radiation and contrast doses, facilitating a more accurate diagnosis of pulmonary embolism.

KEY POINTS: Question The dual-low dose protocol is essential for detecting pulmonary emboli (PE) in follow-up CT pulmonary angiography (PA), yet effective solutions are still lacking. Findings Deep learning reconstruction (DLR)-boost with reduced radiation and contrast doses demonstrated higher quantitative and qualitative image quality than hybrid-iterative reconstruction in the routine CTPA. Clinical relevance DLR-boost based low-radiation and low-contrast-dose CTPA protocol offers a novel strategy to further enhance the image quality and diagnosis accuracy for pulmonary embolism patients.

PMID:40411550 | DOI:10.1007/s00330-025-11681-3

Epilepsia. 2025 May 24. doi: 10.1111/epi.18463. Online ahead of print.

ABSTRACT

OBJECTIVE: This study was undertaken to develop a deep learning framework that can classify and segment interictal epileptiform discharges (IEDs) in multichannel electroencephalographic (EEG) recordings with high accuracy, preserving both spatial information and interchannel interactions.

METHODS: We proposed a novel deep learning framework, U-IEDNet, for detecting IEDs in multichannel EEG. The U-IEDNet framework employs convolutional layers and bidirectional gated recurrent units as a temporal encoder to extract temporal features from single-channel EEG, followed by the use of transformer networks as a spatial encoder to fuse multichannel features and extract interchannel interaction information. Transposed convolutional layers form a temporal decoder, creating a U-shaped architecture with the encoder. This upsamples features to estimate the probability of each EEG sampling point falling within the IED range, enabling segmentation of IEDs from background activity. Two datasets, a public database with 370 patient recordings and our own annotated database with 43 patient recordings, were used for model establishment and validation.

RESULTS: The results showed prominent advantage compared with other methods. U-IEDNet achieved a recall of .916, precision of .911, F1-score of .912, and false positive rate (FPR) of .030 on the public database. The classification performance in our own annotated database achieved a recall of .905, a precision of .902, an F1-score of .903, and an FPR of .072. The segmentation performance had a recall of .903, a precision of .916, and an F1-score of .909. Additionally, this study analyzes attention weights in the transformer network based on brain network theory to elucidate the spatial feature fusion process, enhancing the interpretability of the IED detection model.

SIGNIFICANCE: In this paper, we aim to present an artificial intelligence-based toolbox for IED detection, which may facilitate epilepsy diagnosis at the bedside in the future. U-IEDNet demonstrates great potential to improve the accuracy and efficiency of IED detection in multichannel EEG recordings.

PMID:40411529 | DOI:10.1111/epi.18463

Expert Rev Anticancer Ther. 2025 May 24. doi: 10.1080/14737140.2025.2512040. Online ahead of print.

ABSTRACT

BACKGROUND: Prostate Cancer (PCa) is a severe disease that affects males globally. The Gleason grading system is a widely recognized method for diagnosing the aggressiveness of PCa using histopathological images. This system evaluates prostate tissue to determine the severity of the disease and guide treatment decisions. However, manual analysis of histopathological images requires highly skilled professionals and is time-consuming.

METHODS: To address these challenges, deep learning (DL) is utilized, as it has shown promising results in medical image analysis. Although numerous DL networks have been developed for Gleason grading, many existing methods have limitations such as suboptimal accuracy and high computational complexity. The proposed network integrates MobileNet, an Attention Mechanism (AM), and a capsule network. MobileNet efficiently extracts features from images while addressing computational complexity. The AM focuses on selecting the most relevant features, enhancing the accuracy of Gleason grading. Finally, the capsule network classifies the Gleason grades from histopathological images.

RESULTS: The validation of the proposed network used two datasets, PANDA and Gleason-2019. Ablation studies were conducted and evaluated in the proposed architecture. The results highlight the effectiveness of the proposed network.

CONCLUSIONS: The proposed network outperformed existing approaches, achieving an accuracy of 98.08% on the PANDA dataset and 97.07% on the Gleason-2019 dataset.

PMID:40411485 | DOI:10.1080/14737140.2025.2512040