MethodsX. 2024 Dec 21;14:103131. doi: 10.1016/j.mex.2024.103131. eCollection 2025 Jun.

ABSTRACT

Recent advancements in artificial intelligence (AI) have increased interest in intelligent transportation systems, particularly autonomous vehicles. Safe navigation in traffic-heavy environments requires accurate road scene segmentation, yet traditional computer vision methods struggle with complex scenarios. This study emphasizes the role of deep learning in improving semantic segmentation using datasets like the Indian Driving Dataset (IDD), which presents unique challenges in chaotic road conditions. We propose a modified CANet that incorporates U-Net and LinkNet elements, focusing on accuracy, efficiency, and resilience. The CANet features an encoder-decoder architecture and a Multiscale Context Module (MCM) with three parallel branches to capture contextual information at multiple scales. Our experiments show that the proposed model achieves a mean Intersection over Union (mIoU) value of 0.7053, surpassing state-of-the-art models in efficiency and performance. Here we demonstrate:•Traditional computer vision methods struggle with complex driving scenarios, but deep learning based semantic segmentation methods show promising results.•Modified CANet, incorporating U-Net and LinkNet elements is proposed for semantic segmentation of unstructured driving scenarios.•The CANet structure consists of an encoder-decoder architecture and a Multiscale Context Module (MCM) with three parallel branches to capture contextual information at multiple scales.

PMID:39846010 | PMC:PMC11751566 | DOI:10.1016/j.mex.2024.103131

Eur J Obstet Gynecol Reprod Biol X. 2024 Dec 24;25:100362. doi: 10.1016/j.eurox.2024.100362. eCollection 2025 Mar.

ABSTRACT

This review examines the emerging applications of machine learning (ML) and radiomics in the diagnosis and prediction of placenta accreta spectrum (PAS) disorders, addressing a significant challenge in obstetric care. It highlights recent advancements in ML algorithms and radiomic techniques that utilize medical imaging modalities like magnetic resonance imaging (MRI) and ultrasound for effective classification and risk stratification of PAS. The review discusses the efficacy of various deep learning models, such as nnU-Net and DenseNet-PAS, which have demonstrated superior performance over traditional diagnostic methods through high AUC scores. Furthermore, it underscores the importance of integrating quantitative imaging features with clinical data to enhance diagnostic accuracy and optimize surgical planning. The potential of ML to predict surgical morbidity by analyzing demographic and obstetric factors is also explored. Emphasizing the need for standardized methodologies to ensure consistent feature extraction and model performance, this review advocates for the integration of radiomics and ML into clinical workflows, aiming to improve patient outcomes and foster a multidisciplinary approach in high-risk pregnancies. Future research should focus on larger datasets and validation of biomarkers to refine predictive models in obstetric care.

PMID:39845985 | PMC:PMC11751428 | DOI:10.1016/j.eurox.2024.100362

Genome Integr. 2025 Jan 22;15:e20240002. doi: 10.14293/genint.15.1.002. eCollection 2024.

ABSTRACT

Artificial intelligence (AI) offers a broad range of enhancements in medicine. Machine learning and deep learning techniques have shown significant potential in improving diagnosis and treatment outcomes, from assisting clinicians in diagnosing medical images to ascertaining effective drugs for a specific disease. Despite the prospective benefits, adopting AI in clinical settings requires careful consideration, particularly concerning data generalisation and model explainability. This commentary aims to discuss two potential use cases for AI in the field of medicine and the overarching challenges involved in their implementation.

PMID:39845982 | PMC:PMC11752870 | DOI:10.14293/genint.15.1.002

Front Immunol. 2025 Jan 8;15:1436491. doi: 10.3389/fimmu.2024.1436491. eCollection 2024.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a progressive and fatal disease, characterized by impaired wound repair, tissue remodeling and fibrosis. Immune system may participate in the development and progression of the disease as indicated by altered activity in IPF sufferers. This study investigates the immune response to the BNT162b2 COVID-19 vaccine in patients with IPF compared to healthy controls, with a particular focus on evaluation of antibody responses, interferon-gamma release, cytokine profiling and a broad panel of immune cell subpopulations. IPF patients without prior exposure to SARS-CoV-2 had undetectable levels of anti-N IgG antibodies, highlighting their lack of previous infection. After vaccination, IPF patients showed a significant increase in anti-S1 IgG and IgA antibodies, though their levels were lower compared to healthy controls and convalescent IPF patients. Additionally, IPF patients exhibited altered proportions of regulatory T cells (Tregs) and effector T lymphocytes (Teffs) before and after vaccination. Specifically, IPF patients had higher percentages of Tregs with a Th2 phenotype and Th17 Tregs, along with reduced proportions of Th1/17 Tregs. Teffs in IPF patients showed a decrease in Th1-like and Th2-like populations after vaccination. Moreover, IPF patients demonstrated elevated populations of cytotoxic T lymphocytes (Tc) before vaccination and increased levels of γδ Tc cells throughout the study. Alterations in cytokine profiles were also observed, IPF patients showed higher levels of IL-6 and IL-22 compared to healthy controls. These findings suggest a distinct immune response in IPF patients to the COVID-19 vaccine, characterized by differences in antibody production, T cell differentiation and cytokine secretion compared to healthy individuals.

PMID:39845961 | PMC:PMC11750670 | DOI:10.3389/fimmu.2024.1436491

Respir Res. 2025 Jan 22;26(1):34. doi: 10.1186/s12931-025-03107-x.

ABSTRACT

Progressive forms of interstitial lung diseases, including idiopathic pulmonary fibrosis (IPF), are deadly disorders lacking non-invasive biomarkers for assessment of early disease activity, which presents a major obstacle in disease management. Excessive extracellular matrix (ECM) deposition is a hallmark of these disorders, with fibronectin being an abundant ECM glycoprotein that is highly upregulated in early fibrosis and serves as a scaffold for the deposition of other matrix proteins. Due to its role in active fibrosis, we are targeting fibronectin as a biomarker of early lung fibrosis disease activity via the PEGylated fibronectin-binding polypeptide (PEG-FUD). In this work, we demonstrate the binding of PEG-FUD to the fibrotic lung throughout the course of bleomycin-induced murine model of pulmonary fibrosis. We first analyzed the binding of radiolabeled PEG-FUD following direct incubation to precision cut lung slices from mice at different stages of experimental lung fibrosis. Then, we administered fluorescently labeled PEG-FUD subcutaneously to mice over the course of bleomycin-induced pulmonary fibrosis and assessed peptide uptake 24 h later through ex vivo tissue imaging. Using both methods, we found that peptide targeting to the fibrotic lung is increased during the fibrogenic phase of the single dose bleomycin lung fibrosis model (days 7 and 14 post-bleomycin). At these timepoints we found a correlative relationship between peptide uptake and fibrotic burden. These data suggest that PEG-FUD targets fibronectin associated with active fibrogenesis in this model, making it a promising candidate for a clinically translatable molecular imaging probe to non-invasively determine pulmonary fibrosis disease activity, enabling accelerated therapeutic decision-making.

PMID:39844185 | DOI:10.1186/s12931-025-03107-x

Circ Genom Precis Med. 2025 Jan 23:e004594. doi: 10.1161/CIRCGEN.124.004594. Online ahead of print.

ABSTRACT

BACKGROUND: Transcriptional dysregulation, possibly affected by genetic variation, contributes to disease development. Due to dissimilarities in development, function, and remodeling during disease progression, transcriptional differences between the left atrial (LA) and right atrial (RA) may provide insight into diseases such as atrial fibrillation.

METHODS: Lateral differences in atrial transcription were evaluated in CATCH ME (Characterizing Atrial fibrillation by Translating its Causes into Health Modifiers in the Elderly) using a 2-stage discovery and replication design. The design took advantage of the availability of 32 paired samples, for which both LA and RA tissue were obtained, as a discovery cohort, and 98 LA and 69 RA unpaired samples utilized as a replication cohort.

RESULTS: A total of 714 transcripts were identified and replicated as differentially expressed (DE) between LA and RA, as well as 98 exons in 55 genes. Approximately 50% of DE transcripts were colocated with another frequently correlated DE transcript (PFDR ≤0.05 for 579 regions). These transcription disequilibrium blocks contained examples including side-specific differential exon usage, such as the PITX2 locus, where ENPEP showed evidence of differential exon usage. Analysis of this region in conjunction with BMP10 identified rs9790621 as associated with ENPEP transcription in LA, while rs7687878 was associated with BMP10 expression in RA. In RA, BMP10 and ENPEP were strongly correlated in noncarriers, which was attenuated in risk-allele carriers, where BMP10 and PITX2 expression were strongly correlated.

CONCLUSIONS: These results significantly expand knowledge of the intricate, tissue-specific transcriptional landscape in human atria, including DE transcripts and side-specific isoform expression. Furthermore, they suggest the existence of blocks of transcription disequilibrium influenced by genetics.

PMID:39846178 | DOI:10.1161/CIRCGEN.124.004594

Front Pharmacol. 2025 Jan 8;15:1462193. doi: 10.3389/fphar.2024.1462193. eCollection 2024.

ABSTRACT

INTRODUCTION: Tuberculosis (TB) poses a significant threat to global health, with millions of new infections and approximately one million deaths annually. Various modeling efforts have emerged, offering tailored data-driven and physiologically-based solutions for novel and historical compounds. However, this diverse modeling panorama may lack consistency, limiting result comparability. Drug-specific models are often tied to commercial software and developed on various platforms and languages, potentially hindering access and complicating the comparison of different compounds.

METHODS: This work introduces stormTB: SimulaTOr of a muRine Minimal-pbpk model for anti-TB drugs. It is a web-based interface for our minimal physiologically based pharmacokinetic (mPBPK) platform, designed to simulate custom treatment scenarios for tuberculosis in murine models. The app facilitates visual comparisons of pharmacokinetic profiles, aiding in assessing drug-dose combinations.

RESULTS: The mPBPK model, supporting 11 anti-TB drugs, offers a unified perspective, overcoming the potential inconsistencies arising from diverse modeling efforts. The app, publicly accessible, provides a user-friendly environment for researchers to conduct what-if analyses and contribute to collective TB eradication efforts. The tool generates comprehensive visualizations of drug concentration profiles and pharmacokinetic/pharmacodynamic indices for TB-relevant tissues, empowering researchers in the quest for more effective TB treatments. stormTB is freely available at the link: https://apps.cosbi.eu/stormTB.

PMID:39845781 | PMC:PMC11750688 | DOI:10.3389/fphar.2024.1462193

Front Plant Sci. 2025 Jan 8;15:1446383. doi: 10.3389/fpls.2024.1446383. eCollection 2024.

ABSTRACT

INTRODUCTION: Rice samples exposed to the space environment have generated diverse phenotypic variations. Miniature-inverted-repeat transposable elements (MITEs), often found adjacent to genes, play a significant role in regulating the plant genome. Herein, the contribution of MITEs in regulating space-mutagenic phenotypes was explored.

METHODS: The space-mutagenic phenotype changes in the F3 to F5 generations of three space-mutagenic lines from the rice varieties Dongnong423 (DN423) and Dongnong (DN416) were meticulously traced. Rice leaves samples at the heading stage from three space-mutagenic lines were subjected to high coverage whole-genome bisulfite sequencing and whole-genome sequencing. These analyses were conducted to investigate the effects of MITEs related epigenetic and genetic variations on space-mutagenic phenotypes.

RESULTS AND DISCUSSION: Studies have indicated that MITEs within gene regulatory regions might contribute to the formation and differentiation of space-mutagenic phenotypes. The space environment has been shown to induce the transposable elements insertion polymorphisms of MITEs (MITEs-TIPs), with a notable preference for insertion near genes involved in stress response and phenotype regulation. The space-induced MITEs-TIPs contributed to the formation of space-mutagenic phenotype by modulating the expression of gene near the insertion site. This study underscored the pivotal role of MITEs in modulating plant phenotypic variation induced by the space environment, as well as the transgenerational stability of these phenotypic variants.

PMID:39845491 | PMC:PMC11751223 | DOI:10.3389/fpls.2024.1446383

Front Plant Sci. 2025 Jan 8;15:1457713. doi: 10.3389/fpls.2024.1457713. eCollection 2024.

NO ABSTRACT

PMID:39845489 | PMC:PMC11750989 | DOI:10.3389/fpls.2024.1457713

Microb Biotechnol. 2025 Jan;18(1):e70068. doi: 10.1111/1751-7915.70068.

NO ABSTRACT

PMID:39844583 | DOI:10.1111/1751-7915.70068

BMC Bioinformatics. 2025 Jan 22;26(1):26. doi: 10.1186/s12859-024-06026-8.

ABSTRACT

BACKGROUND: Interpretability is a topical question in recommender systems, especially in healthcare applications. An interpretable classifier quantifies the importance of each input feature for the predicted item-user association in a non-ambiguous fashion.

RESULTS: We introduce the novel Joint Embedding Learning-classifier for improved Interpretability (JELI). By combining the training of a structured collaborative-filtering classifier and an embedding learning task, JELI predicts new user-item associations based on jointly learned item and user embeddings while providing feature-wise importance scores. Therefore, JELI flexibly allows the introduction of priors on the connections between users, items, and features. In particular, JELI simultaneously (a) learns feature, item, and user embeddings; (b) predicts new item-user associations; (c) provides importance scores for each feature. Moreover, JELI instantiates a generic approach to training recommender systems by encoding generic graph-regularization constraints.

CONCLUSIONS: First, we show that the joint training approach yields a gain in the predictive power of the downstream classifier. Second, JELI can recover feature-association dependencies. Finally, JELI induces a restriction in the number of parameters compared to baselines in synthetic and drug-repurposing data sets.

PMID:39844056 | DOI:10.1186/s12859-024-06026-8

Acta Endocrinol (Buchar). 2024 Apr-Jun;20(2):249-255. doi: 10.4183/aeb.2024.249. Epub 2025 Jan 18.

ABSTRACT

INTRODUCTION: Diabetes mellitus, a chronic metabolic disorder stemming from pancreatic dysfunction, is surging in India, notably among those aged 60 and above. The escalating disease prevalence in this demographic necessitates heightened medication use, escalating the risk of Adverse Drug Reactions (ADRs). This underscores the vital role of ADR monitoring to curtail potential harm.

METHOD: A 12-month cross-sectional, prospective, observational study engaged 200 participants from the geriatric Outpatient Department (OPD). Diabetic patients in the geriatric OPD, willing to participate, underwent face-to-face evaluations using a structured questionnaire focused on adverse reactions to anti-diabetic medications. The study also included a Knowledge, Attitude, and Practice (KAP) assessment.

RESULTS: Of the 200 patients, 57% were male, 43% female. Thirteen participants (7 male, 6 female) reported ADR encounters during therapy, predominantly categorized as mild in causality and severity. KAP assessments unveiled a robust understanding of ADRs, primarily shaped by physicians and reinforced by pharmacists. Anticipation of ADR occurrence was noted in 70% of respondents, linked to non-compliance and lifestyle factors.

CONCLUSION: Educating caregivers about the critical importance of monitoring medication adherence among the elderly is imperative. Cultivating an attitude of reporting even minor ADRs to appropriate authorities is essential for harm prevention.

PMID:39845748 | PMC:PMC11750232 | DOI:10.4183/aeb.2024.249

Clin Transl Sci. 2025 Jan;18(1):e70134. doi: 10.1111/cts.70134.

ABSTRACT

The Timor-Leste Pharmacovigilance (PV) became an associate member of the WHO Programme for International Drug Monitoring in 2019; however, the adverse drug reaction (ADR) reporting rate remains low, with only nine reports per 1342 million inhabitants over 5 years. This study aimed to evaluate the knowledge, attitude, practice, and barriers related to ADRs, pharmacovigilance, and ADR reporting among healthcare professionals (HCPs) in Timor-Leste. A cross-sectional survey with a validated, self-administered questionnaire was conducted among 600 HCPs, including clinical doctors, nurses, and pharmacy employees from one national referral and five referral hospitals. Of the 461 HCPs who responded (76.8% response rate), 98 were clinical doctors (21.3%), 311 nurses (67.4%), and 52 pharmacy employees (11.3%). The knowledge score on ADRs was 3.81 ± 0.36 out of 8, with clinical doctors, nurses, and pharmacy employees scoring 4.49 ± 0.51, 3.47 ± 0.24, and 4.56 ± 0.26, respectively. On pharmacovigilance and ADR reporting, the score was 3.00 ± 0.16 out of 8, with clinical doctors, nurses, and pharmacy employees scoring 3.36 ± 0.26, 2.81 ± 0.08, and 3.50 ± 0.24, respectively. All scores referred to the number of correctly answered questions. Positive attitudes were observed, with 53.4% agreeing that ADR reporting is crucial for drug safety, although only 22.0% reported observed ADRs. Key barriers included unavailability of reporting forms (81.0%), insufficient financial support (71.9%), and lack of reporting by colleagues (71.4%). These findings highlight the need for increased awareness, training, and resources to improve ADR reporting in Timor-Leste.

PMID:39844473 | DOI:10.1111/cts.70134

BMC Immunol. 2025 Jan 23;26(1):4. doi: 10.1186/s12865-025-00682-y.

ABSTRACT

Vanishing bile duct syndrome (VBDS) is a serious drug induced liver injury characterized by chronic cholestasis and loss of intrahepatic bile ducts. VBDS has been reported also following checkpoint inhibitor treatment. We compared CD3 + , CD4 + , CD8 + , CD20 + , CD57 + , PD-1 + and PD-L1 + lymphocyte infiltrates in liver biopsies of patients that encountered VBDS (n = 2) or hepatotoxicity (n = 3) after pembrolizumab (n = 4) or nivolumab (n = 1) treatment with samples from normal liver (n = 10), non-alcohol steatohepatitis (NASH, n = 10), primary biliary cholangitis (PBC, n = 10) or pembrolizumab-treated patients without adverse events (n = 2). Notably, none of the cancer patients had primary nor metastatic liver tumors. We also studied direct growth effects of pembrolizumab on primary human intrahepatic biliary epithelial cells (HIBEpiC) in vitro. Liver sections of all checkpoint inhibitor- treated patients exhibited significantly higher CD3 + infiltration than normal livers, and significantly higher PD-L1 + , CD4 + and CD8 + infiltration, than other groups. PD-1 + infiltration was significantly increased in livers of patients with severe hepatic adverse event. CD57 + infiltration was similar in normal livers, NASH- and PBC groups, but highly increased in the checkpoint inhibitor-treated patients. Immune cell infiltrates were similar between NASH and normal livers. PBC samples had significantly higher CD3 + , CD4 + , CD8 + and CD20 + infiltrates than normal livers. HIBEpiC express PD-L1 but pembrolizumab did not affect their viability in vitro. Our findings suggest that VBDS is not due to direct cytotoxicity of checkpoint inhibitors and that the immunological attack against livers induced by these drugs is different from other cholestatic liver conditions.Biological insight: Checkpoint inhibitors upregulate PD-1 and PD-L1, as well as cytotoxic CD57 + cells in the non-cancerous liver tissues and this may be associated with checkpoint inhibitor-induced hepatotoxicity.

PMID:39844069 | DOI:10.1186/s12865-025-00682-y

Funding Opportunity PAR-25-378 from the NIH Guide for Grants and Contracts. The purpose of this notice of funding opportunity (NOFO) is to support research on interventions to improve health in Native American (NA) populations. This includes 1) etiologic research, where there is a significant gap in knowledge, that will directly inform intervention development or adaptations, 2) research that develops, adapts, or tests the efficacy or effectiveness of health promotion and disease prevention interventions, 3) research that tests culturally informed treatment or recovery interventions and 4) where a sufficient body of knowledge on intervention efficacy exists, research on dissemination and implementation that develops and tests strategies to overcome barriers to the adoption, integration, scale-up, and sustainability of effective interventions. Existing data suggest that significant acute and chronic disease inequities exist for NA populations. Concurrently, NA populations experience unique sociopolitical, historical, and environmental stressors and risks that may exacerbate health conditions and/or impact the effectiveness of existing solutions to address the conditions. They also possess unique strengths and resiliencies that can mitigate stressors or inform intervention strategies. Through this initiative, intervention and related research is sought to build upon community knowledge, resources, and resilience to test science-based, culturally appropriate solutions to reduce morbidity and mortality through identification and remediation of precursors to diseases and disorders and through culturally informed treatment. Interventions should be designed with a consideration for sustainability within the communities where they are tested, and have the flexibility to be readily adapted, disseminated, and scaled up to other communities where culturally appropriate. For the purposes of this NOFO, NA includes the following populations: Alaska Natives, American Indians (whose ancestral lands fall at least partially within the U.S. main land).

Notice NOT-HS-25-015 from the NIH Guide for Grants and Contracts

Notice NOT-EY-25-006 from the NIH Guide for Grants and Contracts

Notice NOT-TW-25-001 from the NIH Guide for Grants and Contracts

Funding Opportunity RFA-DK-26-009 from the NIH Guide for Grants and Contracts. The New Investigator Gateway Award in T1D Research is designed to supporta robust pipeline of innovative projects and talented new investigators in T1D research. In addition to providing support for preliminary research, the Gateway program provides an opportunity for new Program Directors/Principal Investigators (PD/PIs) to pursue their studies within the intellectual environment of a select number of large, ongoing collaborative research programs. Embedding awardees within an established scientific framework in each of these consortia will provide unique opportunities for New andEarly Stage Investigators to increase their understanding of key questions in the field, to network, and to establish unique and potentially long-lasting collaborations that will propel their careers forward. Bringing New and Early Stage Investigators into existing collaborative research networks will also benefit the networks by providing new ideas and perspectives.

Notice NOT-HD-25-001 from the NIH Guide for Grants and Contracts