Neuro Oncol. 2025 Feb 1:noaf026. doi: 10.1093/neuonc/noaf026. Online ahead of print.

ABSTRACT

BACKGROUND: Glioblastoma (GBM) presents a significant clinical challenge due to its aggressive nature and extensive heterogeneity. Tumour purity, the proportion of malignant cells within a tumour, is an important covariate for understanding the disease, having direct clinical relevance or obscuring signal of the malignant portion in molecular analyses of bulk samples. However, current methods for estimating tumour purity are non-specific and technically demanding. Therefore, we aimed to build a reliable and accessible purity estimator for GBM.

METHODS: We developed GBMPurity, a deep-learning model specifically designed to estimate the purity of IDH-wildtype primary GBM from bulk RNA-seq data. The model was trained using simulated pseudobulk tumours of known purity from labelled single-cell data acquired from the GBmap resource. The performance of GBMPurity was evaluated and compared to several existing tools using independent datasets.

RESULTS: GBMPurity outperformed existing tools, achieving a mean absolute error of 0.15 and a concordance correlation coefficient of 0.88 on validation datasets. We demonstrate the utility of GBMPurity through inference on bulk RNA-seq samples and observe reduced purity of the Proneural molecular subtype relative to the Classical, attributed to the increased presence of healthy brain cells.

CONCLUSIONS: GBMPurity provides a reliable and accessible tool for estimating tumour purity from bulk RNA-seq data, enhancing the interpretation of bulk RNA-seq data and offering valuable insights into GBM biology. To facilitate the use of this model by the wider research community, GBMPurity is available as a web-based tool at: https://gbmdeconvoluter.leeds.ac.uk/.

PMID:39891579 | DOI:10.1093/neuonc/noaf026

Bioinformatics. 2025 Jan 31:btaf051. doi: 10.1093/bioinformatics/btaf051. Online ahead of print.

ABSTRACT

MOTIVATION: As the biological roles and disease implications of non-coding RNAs continue to emerge, the need to thoroughly characterize previously unexplored non-coding RNAs becomes increasingly urgent. These molecules hold potential as biomarkers and therapeutic targets. However, the vast and complex nature of non-coding RNAs data presents a challenge. We introduce MMnc, an interpretable deep learning approach designed to classify non-coding RNAs into functional groups. MMnc leverages multiple data sources-such as the sequence, secondary structure, and expression-using attention-based multi-modal data integration. This ensures learning of meaningful representations while accounting for missing sources in some samples.

RESULTS: Our findings demonstrate that MMnc achieves high classification accuracy across diverse non-coding RNA classes. The method's modular architecture allows for the consideration of multiple types of modalities, whereas other tools only consider one or two at most. MMnc is resilient to missing data, ensuring that all available information is effectively utilized. Importantly, the generated attention scores offer interpretable insights into the underlying patterns of the different non-coding RNA classes, potentially driving future non-coding RNA research and applications.

AVAILABILITY: Data and source code can be found at EvryRNA.ibisc.univ-evry.fr/EvryRNA/MMnc.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:39891346 | DOI:10.1093/bioinformatics/btaf051

J Thromb Thrombolysis. 2025 Feb 1. doi: 10.1007/s11239-025-03072-8. Online ahead of print.

ABSTRACT

Chronic thromboembolic pulmonary hypertension (CTEPH) and COVID-19 share molecular pathways yet remain poorly understood in their interrelation. Using RNA-seq datasets (GSE130391 and GSE169687), we identified 645, 206, and 1,543 differentially expressed genes (DEGs) for long-COVID (16 and 24 weeks post-infection) and CTEPH, respectively. Weighted Gene Co-Expression Network Analysis (WGCNA) pinpointed 234 intersecting key module genes. Three hub genes-DNAJA1, NDUFA5, and SLC2A14-were identified with robust discriminatory capabilities (AUC ≥ 0.7). Enrichment analyses revealed shared pathways linked to immune modulation, oxidative stress, and metabolic dysfunction. Immune analysis highlighted activated CD8 T cells as critical regulators. Regulatory networks implicated TFs and miRNAs, including STAT1 and hsa-mir-23a-3p. Drug prediction identified potential therapeutic compounds with strong molecular docking interactions. These findings unravel critical molecular linkages, emphasizing shared pathogeneses and guiding experimental validations for improved diagnostic and therapeutic strategies in COVID-19 and CTEPH.

PMID:39891865 | DOI:10.1007/s11239-025-03072-8

Eur J Nutr. 2025 Feb 1;64(2):75. doi: 10.1007/s00394-025-03589-x.

ABSTRACT

PURPOSE: During pregnancy, extracellular vesicle and particle microRNAs (EVP miRNA) in maternal circulation have the capacity to cross the placenta and facilitate maternal-fetal communication. Both dysregulation of circulating EVP miRNA during pregnancy and maternal diet quality have been previously associated with pregnancy complications and adverse birth outcomes. However, little is known about how maternal diet influences circulating EVP miRNA during pregnancy. This study assesses associations between maternal diet quality, as measured by the Alternative Healthy Eating Index (2010; AHEI-2010), and EVP miRNA levels in maternal circulation during pregnancy.

METHODS: In a pilot study of 53 pregnant participants in the New Hampshire Birth Cohort Study, maternal diet quality was assessed using AHEI-2010 and plasma (mean gestational age at blood collection: 28.8 weeks) EVP miRNA were profiled using the NanoString nCounter platform which interrogates 798 miRNA transcripts.

RESULTS: In covariate-adjusted models, the AHEI-2010 adherence score was negatively associated (P < 0.05) with the number of unique miRNA transcripts detectable in each sample. In post hoc analyses, greater consumption of red and processed meats was positively associated with levels of 7 miRNA (Q < 0.05), including hsa-miR-512-5p (PBonf < 0.01), a member of the placenta-specific chromosome 19 miRNA cluster.

CONCLUSION: We identified associations between the consumption of red and processed meat and levels of circulating select EVP miRNA during pregnancy, including placenta-specific miRNA and miRNA with target genes overrepresented in pathways involved in placental development. Additional research is needed to assess whether alterations in maternal circulating EVP miRNA may mediate maternal diet quality's impacts on pregnancy and birth outcomes.

PMID:39891736 | DOI:10.1007/s00394-025-03589-x

Expert Opin Drug Saf. 2025 Feb 1. doi: 10.1080/14740338.2025.2460448. Online ahead of print.

ABSTRACT

BACKGROUND: Nasal septum perforation represents a significant clinical concern, with limited investigations into the role of medications in its etiology. This study utilizes the FDA Adverse Event Reporting System (FAERS) database to identify the drugs associated with nasal septum perforation and assess their risk.

RESEARCH DESIGN AND METHODS: This retrospective pharmacovigilance study analyzed drug-induced nasal septum perforation data from January 2004 to December 2023. Disproportionality analysis using reporting odds ratio (ROR) assessed drug associations with nasal septum perforation.

RESULTS: For 552 identified cases, the most commonly reported drugs were bevacizumab (n = 56), fluticasone propionate (n = 50), methotrexate (n = 34), hydrocodone and acetaminophen (n = 22), and paclitaxel (n = 17). Twenty-six drugs showed positive risk signals, with the top five being azelastine hydrochloride and fluticasone propionate (ROR = 173.82), beclomethasone dipropionate (ROR = 90.91), oxymetazoline (ROR = 53.77), desmopressin (ROR = 51.43), and leucovorin (ROR = 42.83). Intriguingly, 18 of these drugs did not list nasal septum perforation as a known side effect.

CONCLUSION: This study provides a comprehensive overview of drug-induced nasal septum perforation from a pharmacovigilance perspective, highlighting the need for further research to clarify these associations and update drug safety information to reduce patient risk.

PMID:39891519 | DOI:10.1080/14740338.2025.2460448

JAMA Netw Open. 2025 Jan 2;8(1):e2457341. doi: 10.1001/jamanetworkopen.2024.57341.

ABSTRACT

IMPORTANCE: Ultraprocessed foods (UPF), characterized as shelf-stable but nutritionally imbalanced foods, pose a public health crisis worldwide. In adults, UPF consumption is associated with increased obesity risk, but findings among children are inconsistent.

OBJECTIVES: To examine the associations among UPF intake, anthropometric adiposity indicators, and obesity status in Canadian children.

DESIGN, SETTING, AND PARTICIPANTS: In the CHILD Cohort Study, one of the largest prospective, multicenter, population-based pregnancy cohorts in Canada, diet was assessed during the 3-year visit (September 2011 to June 2016), and anthropometric measurements were assessed at the 5-year visit (December 2013 to April 2018). Data analysis was performed between July 1, 2023, and June 30, 2024.

EXPOSURE: Diet intake was assessed using a semiquantitative food frequency questionnaire at 3 years of age. UPFs were identified using the NOVA classification system.

MAIN OUTCOMES AND MEASURES: Anthropometric adiposity indicators were measured at 5 years of age and used to calculate age- and sex-standardized z scores for body mass index (BMI), waist to height ratio, and subscapular and triceps skinfold thicknesses, and obesity, which was defined using BMI z score cutoffs. Multivariable-adjusted regression analyses were used to examine the associations of UPF with adiposity and obesity development, accounting for parental, birth, and early-childhood factors.

RESULTS: Among 2217 participants included in this study, median age at the outcome assessment was 5.0 (IQR, 5.0-5.1) years, and 1175 (53.0%) were males. At 3 years of age, UPF contributed 45.0% of total daily energy intake. UPF energy contribution was higher in males vs females (46.0% vs 43.9%; P < .001). Among all participants, higher UPF intake at 3 years of age was associated with higher anthropometric adiposity indicators at 5 years of age, primarily driven by males. In males, every 10% UPF energy increase was associated with higher adiposity indicator z scores for BMI (β, 0.08; 95% CI, 0.03-0.14), waist to height ratio (β, 0.07; 95% CI, 0.01-0.12), and subscapular (β, 0.12; 95% CI, 0.06-0.18) and triceps (β, 0.09; 95% CI, 0.03-0.15) skinfold thickness and higher odds of living with overweight or obesity (odds ratio, 1.19; 95% CI, 1.03-1.36). No significant associations were observed among females.

CONCLUSIONS AND RELEVANCE: In this cohort study of Canadian children, high UPF consumption during early childhood was associated with obesity development, primarily in males. These findings can inform targeted public health initiatives for early childhood centers and caregiver education programs to reduce UPF intake and prevent obesity.

PMID:39888617 | PMC:PMC11786234 | DOI:10.1001/jamanetworkopen.2024.57341

J Transl Med. 2025 Feb 1;23(1):57. doi: 10.1186/s12967-024-05938-6.

ABSTRACT

BACKGROUND: The traditional process of developing new drugs is time-consuming and often unsuccessful, making drug repurposing an appealing alternative due to its speed and safety. Graph neural networks (GCNs) have emerged as a leading approach for predicting drug-disease associations by integrating drug and disease-related networks with advanced deep learning algorithms. However, GCNs generally infer association probabilities only for existing drugs and diseases, requiring network re-establishment and retraining for novel entities. Additionally, these methods often struggle with sparse networks and fail to elucidate the biological mechanisms underlying newly predicted drugs.

METHODS: To address the limitations of traditional methods, we developed HEDDI-Net, a heterogeneous embedding architecture designed to accurately detect drug-disease associations while preserving the interpretability of biological mechanisms. HEDDI-Net integrates graph and shallow learning techniques to extract representative diseases and proteins, respectively. These representative diseases and proteins are used to embed the input features, which are then utilized in a multilayer perceptron for predicting drug-disease associations.

RESULTS: In experiments, HEDDI-Net achieves areas under the receiver operating characteristic curve of over 0.98, outperforming state-of-the-art methods. Rigorous recovery analyses reveal a median recovery rate of 73% for the top 100 diseases, demonstrating its efficacy in identifying novel target diseases for existing drugs, known as drug repurposing. A case study on Alzheimer's disease highlighted the model's practical applicability and interpretability, identifying potential drug candidates like Baclofen, Fluoxetine, Pentoxifylline and Phenytoin. Notably, over 40% of the predicted candidates in the clusters of commonly prescribed clinical drugs Donepezil and Galantamine had been tested in clinical trials, validating the model's predictive accuracy and practical relevance.

CONCLUSIONS: HEDDI-NET represents a significant advancement by allowing direct application to new diseases and drugs without the need for retraining, a limitation of most GCN-based methods. Furthermore, HEDDI-Net provides detailed affinity patterns with representative proteins for predicted candidate drugs, facilitating an understanding of their physiological effects. This capability also supports the design and testing of alternative drugs that are similar to existing medications, enhancing the reliability and interpretability of potential repurposed drugs. The case study on Alzheimer's disease further underscores HEDDI-Net's ability to predict promising drugs and its applicability in drug repurposing.

PMID:39891114 | DOI:10.1186/s12967-024-05938-6

J Transl Med. 2025 Feb 1;23(1):57. doi: 10.1186/s12967-024-05938-6.

ABSTRACT

BACKGROUND: The traditional process of developing new drugs is time-consuming and often unsuccessful, making drug repurposing an appealing alternative due to its speed and safety. Graph neural networks (GCNs) have emerged as a leading approach for predicting drug-disease associations by integrating drug and disease-related networks with advanced deep learning algorithms. However, GCNs generally infer association probabilities only for existing drugs and diseases, requiring network re-establishment and retraining for novel entities. Additionally, these methods often struggle with sparse networks and fail to elucidate the biological mechanisms underlying newly predicted drugs.

METHODS: To address the limitations of traditional methods, we developed HEDDI-Net, a heterogeneous embedding architecture designed to accurately detect drug-disease associations while preserving the interpretability of biological mechanisms. HEDDI-Net integrates graph and shallow learning techniques to extract representative diseases and proteins, respectively. These representative diseases and proteins are used to embed the input features, which are then utilized in a multilayer perceptron for predicting drug-disease associations.

RESULTS: In experiments, HEDDI-Net achieves areas under the receiver operating characteristic curve of over 0.98, outperforming state-of-the-art methods. Rigorous recovery analyses reveal a median recovery rate of 73% for the top 100 diseases, demonstrating its efficacy in identifying novel target diseases for existing drugs, known as drug repurposing. A case study on Alzheimer's disease highlighted the model's practical applicability and interpretability, identifying potential drug candidates like Baclofen, Fluoxetine, Pentoxifylline and Phenytoin. Notably, over 40% of the predicted candidates in the clusters of commonly prescribed clinical drugs Donepezil and Galantamine had been tested in clinical trials, validating the model's predictive accuracy and practical relevance.

CONCLUSIONS: HEDDI-NET represents a significant advancement by allowing direct application to new diseases and drugs without the need for retraining, a limitation of most GCN-based methods. Furthermore, HEDDI-Net provides detailed affinity patterns with representative proteins for predicted candidate drugs, facilitating an understanding of their physiological effects. This capability also supports the design and testing of alternative drugs that are similar to existing medications, enhancing the reliability and interpretability of potential repurposed drugs. The case study on Alzheimer's disease further underscores HEDDI-Net's ability to predict promising drugs and its applicability in drug repurposing.

PMID:39891114 | DOI:10.1186/s12967-024-05938-6

Lancet Digit Health. 2025 Feb;7(2):e145-e156. doi: 10.1016/S2589-7500(24)00253-X.

ABSTRACT

BACKGROUND: The Global Burden of Disease Study has provided key evidence to inform clinicians, researchers, and policy makers across common diseases, but no similar effort with a single-study design exists for hundreds of rare diseases. Consequently, for many rare conditions there is little population-level evidence, including prevalence and clinical vulnerability, resulting in an absence of evidence-based care that was prominent during the COVID-19 pandemic. We aimed to inform rare disease care by providing key descriptors from national data and explore the impact of rare diseases during the COVID-19 pandemic.

METHODS: In this nationwide retrospective observational cohort study, we used the electronic health records (EHRs) of more than 58 million people in England, linking nine National Health Service datasets spanning health-care settings for people who were alive on Jan 23, 2020. Starting with all rare diseases listed in Orphanet (an extensive online resource for rare diseases), we quality assured and filtered down to analyse 331 conditions mapped to ICD-10 or Systemized Nomenclature of Medicine-Clinical Terms that were clinically validated in our dataset. For all 331 rare diseases, we calculated population prevalences, analysed patients' clinical and demographic details, and investigated mortality with SARS-CoV-2. We assessed COVID-19-related mortality by comparing cohorts of patients for each rare disease and rare disease category with controls matched for age group, sex, ethnicity, and vaccination status, at a ratio of two controls per individual with a rare disease.

FINDINGS: Of 58 162 316 individuals, we identified 894 396 with at least one rare disease and assessed COVID-19-related mortality between Sept 1, 2020, and Nov 30, 2021. We calculated reproducible estimates, adjusted for age and sex, for all 331 rare diseases, including for 186 (56·2%) conditions without existing prevalence estimates in Orphanet. 49 rare diseases were significantly more frequent in female individuals than in male individuals, and 62 were significantly more frequent in male individuals than in female individuals; 47 were significantly more frequent in Asian or British Asian individuals than in White individuals; and 22 were significantly more frequent in Black or Black British individuals than in White individuals. 37 rare diseases were significantly more frequent in the White population compared with either the Black or Asian population. 7965 (0·9%) of 894 396 patients with a rare disease died from COVID-19, compared with 141 287 (0·2%) of 58 162 316 in the full study population. In fully vaccinated individuals, the risk of COVID-19-related mortality was significantly higher for eight rare diseases, with patients with bullous pemphigoid (hazard ratio 8·07, 95% CI 3·01-21·62) being at highest risk.

INTERPRETATION: Our study highlights that national-scale EHRs provide a unique resource to estimate detailed prevalence, clinical, and demographic data for rare diseases. Using COVID-19-related mortality analysis, we showed the power of large-scale EHRs in providing insights to inform public health decision making for these often neglected patient populations.

FUNDING: British Heart Foundation Data Science Centre, led by Health Data Research UK.

PMID:39890245 | DOI:10.1016/S2589-7500(24)00253-X

J Am Med Inform Assoc. 2025 Jan 31:ocaf012. doi: 10.1093/jamia/ocaf012. Online ahead of print.

ABSTRACT

OBJECTIVES: Report the development of the patient-centered myAURA application and suite of methods designed to aid epilepsy patients, caregivers, and clinicians in making decisions about self-management and care.

MATERIALS AND METHODS: myAURA rests on an unprecedented collection of epilepsy-relevant heterogeneous data resources, such as biomedical databases, social media, and electronic health records (EHRs). We use a patient-centered biomedical dictionary to link the collected data in a multilayer knowledge graph (KG) computed with a generalizable, open-source methodology.

RESULTS: Our approach is based on a novel network sparsification method that uses the metric backbone of weighted graphs to discover important edges for inference, recommendation, and visualization. We demonstrate by studying drug-drug interaction from EHRs, extracting epilepsy-focused digital cohorts from social media, and generating a multilayer KG visualization. We also present our patient-centered design and pilot-testing of myAURA, including its user interface.

DISCUSSION: The ability to search and explore myAURA's heterogeneous data sources in a single, sparsified, multilayer KG is highly useful for a range of epilepsy studies and stakeholder support.

CONCLUSION: Our stakeholder-driven, scalable approach to integrating traditional and nontraditional data sources enables both clinical discovery and data-powered patient self-management in epilepsy and can be generalized to other chronic conditions.

PMID:39890454 | DOI:10.1093/jamia/ocaf012

Anesthesiol Clin. 2025 Mar;43(1):127-140. doi: 10.1016/j.anclin.2024.08.001. Epub 2024 Sep 17.

ABSTRACT

Biological sex as a variable in pain perception has been rigorously studied. However, there is little correlation between clinical and experimental studies with regard to this. There has been a surge of interest and research in the correlation of genes and single-nucleotide polymorphisms in relation to pain perception, and opioid pharmacokinetics. However, there have not yet been studies or reports of generalized application of this testing to improve acute postoperative pain outcomes.

PMID:39890315 | DOI:10.1016/j.anclin.2024.08.001

Reprod Biomed Online. 2024 Oct 21;50(3):104494. doi: 10.1016/j.rbmo.2024.104494. Online ahead of print.

ABSTRACT

Persistent Müllerian duct syndrome (PMDS) is a rare autosomal recessive syndrome characterized by the coexistence of Müllerian derivatives in a normally virilized male, caused by mutations in the AMH or AMHR2 gene. This paper reports the case of a 33-year-old man with PMDS, diagnosed late during an infertility check-up. Exploratory laparoscopy revealed two intra-pelvic gonads and Müllerian duct structures. Genetic analysis identified an undescribed homozygous missense mutation in the fifth exon of AMH. Typically, PMDS is diagnosed in the presence of cryptorchidism or inguinal hernia, and rarely in the context of infertility. Early orchidopexy is recommended to mitigate fertility sequelae while preserving endogenous hormone secretion. This late diagnosis of PMDS led to a discussion of the management of infertility, surgical strategies and adult follow-up. In this case, the decision was made with the patient to perform minimally invasive surgery, specifically unilateral orchidectomy for fertility management. The biopsy revealed no spermatozoa, probably due to prolonged untreated pelvic cryptorchidism. Retaining one testicle maintains endogenous testosterone production, thus avoiding imperfect hormonal replacement. Given the risk of tumoural degeneration, albeit a low one, annual imaging follow-up is mandatory and removal of Müllerian structures and gonadectomy may be considered if necessary.

PMID:39889328 | DOI:10.1016/j.rbmo.2024.104494

FEMS Microbiol Lett. 2025 Jan 31:fnaf017. doi: 10.1093/femsle/fnaf017. Online ahead of print.

ABSTRACT

Pseudomonas aeruginosa has increasing clinical relevance and commonly occupies the cystic fibrosis (CF) airways. Its ability to colonize and persist in diverse niches is attributed to its large accessory genome, where prophages represent a common feature and may contribute to its fitness and persistence. We focused on the CF airways niche and used 197 longitudinal isolates from 12 patients persistently infected by P. aeruginosa. We computationally predicted intact prophages for each longitudinal group and scored their long-term persistence. We then confirmed prophage inducibility and mapped their location in the host chromosome with lysate sequencing. Using comparative genomics, we evaluated prophage genomic diversity, long-term persistence and level of genomic maintenance. Our findings support previous findings that most P. aeruginosa genomes harbour prophages some of which can self-induce, and that a common CF-treating antibiotic, ciprofloxacin, can induce prophages. Induced prophage genomes displayed high diversity and even genomic novelty. Finally, all induced prophages persisted long-term with their genomes avoiding gene loss and degradation over four years of host replication in the stressful CF airways niche. This and our detection of phage genes which contribute to host competitiveness and adaptation, lends support to our hypothesis that the vast majority of prophages detected as intact and inducible in this study facilitated their host fitness and persistence.

PMID:39890605 | DOI:10.1093/femsle/fnaf017

J Cyst Fibros. 2025 Jan 30:S1569-1993(25)00001-3. doi: 10.1016/j.jcf.2025.01.001. Online ahead of print.

ABSTRACT

BACKGROUND: CFTR modulator (CFTR-M) therapy has led to improved clinical outcomes amongst people with cystic fibrosis (PwCF) eligible for these therapies. However, there is limited data on their impact on the basic life needs and financial concerns of PwCF.

METHODS: We used data from the Wellness in the Modulator Era (Well-ME) survey, which includes data from 900 PwCF both taking and not taking CFTR-M. We examined self-reported financial well-being over time and changes associated with school or work, financial planning, and costs of living. Descriptive statistics were used to analyze responses.

RESULTS: Most respondents reported no change in financial well-being, but 13 % identified a positive change and 16 % reported a negative change. Positive changes in basic life needs included fewer missed work and school days, while negative changes included medical out-of-pocket costs. Worries about financial problems were reported in 35 % of all respondents and were more common in PwCF who never took CFTR-M or had been taking one and then stopped, in PwCF with lower lung function, and in PwCF with Medicaid insurance.

CONCLUSIONS: These results indicate that for most PwCF, CFTR-M have not affected their basic life needs, and a substantial proportion of PwCF continue to experience financial stress and concerns. Many respondents' financial concerns focused on medical costs and insurance. These data underscore the continued need for CF care teams to address PwCF's financial stress and ability to meet basic life needs, even in the era of improved physical health outcomes due to CFTR-M therapy.

PMID:39890522 | DOI:10.1016/j.jcf.2025.01.001

Eur J Med Res. 2025 Jan 31;30(1):61. doi: 10.1186/s40001-025-02339-3.

ABSTRACT

Malignant gliomas, including glioblastoma, are amongst the most aggressive primary brain tumours, characterised by rapid progression and a poor prognosis. Survival analysis is an essential aspect of glioma management and research, as most studies use time-to-event outcomes to assess overall survival (OS) and progression-free survival (PFS) as key measures to evaluate patients. However, predicting survival using traditional methods such as the Kaplan-Meier estimator and the Cox Proportional Hazards (CPH) model has faced many challenges and inaccuracies. Recently, advances in artificial intelligence (AI), including machine learning (ML) and deep learning (DL), have enabled significant improvements in survival prediction for glioma patients by integrating multimodal data such as imaging, clinical parameters and molecular biomarkers. This study highlights the comparative effectiveness of imaging-based, non-imaging and combined AI models. Imaging models excel at identifying tumour-specific features through radiomics, achieving high predictive accuracy. Non-imaging approaches also excel in utilising clinical and genetic data to provide complementary insights, whilst combined methods integrate multiple data modalities and have the greatest potential for accurate survival prediction. Limitations include data heterogeneity, interpretability challenges and computational demands, particularly in resource-limited settings. Solutions such as federated learning, lightweight AI models and explainable AI frameworks are proposed to overcome these barriers. Ultimately, the integration of advanced AI techniques promises to transform glioma management by enabling personalised treatment strategies and improved prognostic accuracy.

PMID:39891313 | DOI:10.1186/s40001-025-02339-3

BMC Biomed Eng. 2025 Feb 1;7(1):2. doi: 10.1186/s42490-025-00088-2.

ABSTRACT

AIM: The aim of this study is to apply a novel hybrid framework incorporating a Vision Transformer (ViT) and bidirectional long short-term memory (Bi-LSTM) model for classifying physical activity intensity (PAI) in adults using gravity-based acceleration. Additionally, it further investigates how PAI and temporal window (TW) impacts the model' s accuracy.

METHOD: This research used the Capture-24 dataset, consisting of raw accelerometer data from 151 participants aged 18 to 91. Gravity-based acceleration was utilised to generate images encoding various PAIs. These images were subsequently analysed using the ViT-BiLSTM model, with results presented in confusion matrices and compared with baseline models. The model's robustness was evaluated through temporal stability testing and examination of accuracy and loss curves.

RESULT: The ViT-BiLSTM model excelled in PAI classification task, achieving an overall accuracy of 98.5% ± 1.48% across five TWs-98.7% for 1s, 98.1% for 5s, 98.2% for 10s, 99% for 15s, and 98.65% for 30s of TW. The model consistently exhibited superior accuracy in predicting sedentary (98.9% ± 1%) compared to light physical activity (98.2% ± 2%) and moderate-to-vigorous physical activity (98.2% ± 3%). ANOVA showed no significant accuracy variation across PAIs (F = 2.18, p = 0.13) and TW (F = 0.52, p = 0.72). Accuracy and loss curves show the model consistently improves its performance across epochs, demonstrating its excellent robustness.

CONCLUSION: This study demonstrates the ViT-BiLSTM model's efficacy in classifying PAI using gravity-based acceleration, with performance remaining consistent across diverse TWs and intensities. However, PAI and TW could result in slight variations in the model's performance. Future research should concern and investigate the impact of gravity-based acceleration on PAI thresholds, which may influence model's robustness and reliability.

PMID:39891283 | DOI:10.1186/s42490-025-00088-2

Reprod Biol Endocrinol. 2025 Jan 31;23(1):16. doi: 10.1186/s12958-025-01351-w.

ABSTRACT

BACKGROUND: Artificial intelligence (AI) models analyzing embryo time-lapse images have been developed to predict the likelihood of pregnancy following in vitro fertilization (IVF). However, limited research exists on methods ensuring AI consistency and reliability in clinical settings during its development and validation process. We present a methodology for developing and validating an AI model across multiple datasets to demonstrate reliable performance in evaluating blastocyst-stage embryos.

METHODS: This multicenter analysis utilizes time-lapse images, pregnancy outcomes, and morphologic annotations from embryos collected at 10 IVF clinics across 9 countries between 2018 and 2022. The four-step methodology for developing and evaluating the AI model include: (I) curating annotated datasets that represent the intended clinical use case; (II) developing and optimizing the AI model; (III) evaluating the AI's performance by assessing its discriminative power and associations with pregnancy probability across variable data; and (IV) ensuring interpretability and explainability by correlating AI scores with relevant morphologic features of embryo quality. Three datasets were used: the training and validation dataset (n = 16,935 embryos), the blind test dataset (n = 1,708 embryos; 3 clinics), and the independent dataset (n = 7,445 embryos; 7 clinics) derived from previously unseen clinic cohorts.

RESULTS: The AI was designed as a deep learning classifier ranking embryos by score according to their likelihood of clinical pregnancy. Higher AI score brackets were associated with increased fetal heartbeat (FH) likelihood across all evaluated datasets, showing a trend of increasing odds ratios (OR). The highest OR was observed in the top G4 bracket (test dataset G4 score ≥ 7.5: OR 3.84; independent dataset G4 score ≥ 7.5: OR 4.01), while the lowest was in the G1 bracket (test dataset G1 score < 4.0: OR 0.40; independent dataset G1 score < 4.0: OR 0.45). AI score brackets G2, G3, and G4 displayed OR values above 1.0 (P < 0.05), indicating linear associations with FH likelihood. Average AI scores were consistently higher for FH-positive than for FH-negative embryos within each age subgroup. Positive correlations were also observed between AI scores and key morphologic parameters used to predict embryo quality.

CONCLUSIONS: Strong AI performance across multiple datasets demonstrates the value of our four-step methodology in developing and validating the AI as a reliable adjunct to embryo evaluation.

PMID:39891250 | DOI:10.1186/s12958-025-01351-w

BMC Med Inform Decis Mak. 2025 Jan 31;25(1):48. doi: 10.1186/s12911-025-02889-w.

ABSTRACT

This paper introduces SkinWiseNet (SWNet), a deep convolutional neural network designed for the detection and automatic classification of potentially malignant skin cancer conditions. SWNet optimizes feature extraction through multiple pathways, emphasizing network width augmentation to enhance efficiency. The proposed model addresses potential biases associated with skin conditions, particularly in individuals with darker skin tones or excessive hair, by incorporating feature fusion to assimilate insights from diverse datasets. Extensive experiments were conducted using publicly accessible datasets to evaluate SWNet's effectiveness.This study utilized four datasets-Mnist-HAM10000, ISIC2019, ISIC2020, and Melanoma Skin Cancer-comprising skin cancer images categorized into benign and malignant classes. Explainable Artificial Intelligence (XAI) techniques, specifically Grad-CAM, were employed to enhance the interpretability of the model's decisions. Comparative analysis was performed with three pre-existing deep learning networks-EfficientNet, MobileNet, and Darknet. The results demonstrate SWNet's superiority, achieving an accuracy of 99.86% and an F1 score of 99.95%, underscoring its efficacy in gradient propagation and feature capture across various levels. This research highlights the significant potential of SWNet in advancing skin cancer detection and classification, providing a robust tool for accurate and early diagnosis. The integration of feature fusion enhances accuracy and mitigates biases associated with hair and skin tones. The outcomes of this study contribute to improved patient outcomes and healthcare practices, showcasing SWNet's exceptional capabilities in skin cancer detection and classification.

PMID:39891245 | DOI:10.1186/s12911-025-02889-w

BMC Pulm Med. 2025 Jan 31;25(1):52. doi: 10.1186/s12890-025-03522-2.

ABSTRACT

BACKGROUND: Pulmonary fibrosis is characterized by the destruction of normal lung tissue and then replacement by abnormal fibrous tissue, leading to an overall decrease in gas exchange function. The effective treatment for pulmonary fibrosis remains unknown. The upstream pathogenesis of pulmonary fibrosis may involve cellular senescence of the lung tissue. Previously, a new gene therapy technology using Box A of the HMGB1 plasmid (Box A) was used to reverse cellular senescence and cure liver fibrosis in aged rats.

METHODS: Here, we show that Box A is a promising medicine for the treatment of lung fibrosis. In a bleomycin-induced pulmonary fibrosis model in the male Wistar rats, Student's t-test and one-way ANOVA were used to compare groups of samples.

RESULTS: Box A effectively lowered fibrous tissue deposits (from 18.74 ± 0.62 to 3.45 ± 1.19%) and senescent cells (from 3.74 ± 0.40% to 0.89 ± 0.18%) to levels comparable to those of the negative control group. Moreover, after eight weeks, Box A also increased the production of the surfactant protein C (from 3.60 ± 1.68% to 6.82 ± 0.65%).

CONCLUSIONS: Our results demonstrate that Box A is a promising therapeutic approach for pulmonary fibrosis and other senescence-promoted fibrotic lesions.

PMID:39891078 | DOI:10.1186/s12890-025-03522-2

J Biotechnol. 2025 Jan 29:S0168-1656(25)00025-2. doi: 10.1016/j.jbiotec.2025.01.019. Online ahead of print.

ABSTRACT

Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the most common chronic liver disease worldwide, affecting more than 30 percent of adults. The most severe form of MASLD, metabolic dysfunction-associated steatohepatitis (MASH), is characterized by necrotizing inflammation and rapid fibrosis progression, often leading to cirrhosis and hepatocellular carcinoma. Currently, only Resmetirom is approved for the treatment of MASH one of the main reasons is the absence of representative in vivo or in vitro models for MASH. To address this challenge, we developed a high-throughput 3D spheroid model consisting of human hepatocellular carcinoma cells (HepG2) and human hepatic stellate cells (LX-2) on microwell arrays. This model, induced with free fatty acids (FFA) to simulate steatosis and fibrosis, enables the assessment of efficacy and mechanisms for potential anti-MASH drugs. Our findings demonstrate that this in vitro spheroid model replicates key pathological features of human MASLD, including steatosis, oxidative stress, and fibrosis. Upon validation, we selected pirfenidone (PFD) and yinfenidone (AC-003), which are commonly used to treat idiopathic pulmonary fibrosis (IPF), to test their anti-MASH efficacy. Treatment with these drugs showed that they could regulate lipid synthesis and metabolism genes, reduce lipid accumulation, oxidative stress, and fibrosis levels. This 3D spheroid model represents a straightforward and efficient tool for screening anti-MASH drugs and investigating the molecular mechanisms of drug action.

PMID:39889902 | DOI:10.1016/j.jbiotec.2025.01.019