Neuroimage Clin. 2025 Feb 18;45:103752. doi: 10.1016/j.nicl.2025.103752. Online ahead of print.

ABSTRACT

Histopathology provides critical insights into the neurological processes inducing neurodegenerative diseases and their impact on the brain, but brain banks combining histology and neuroimaging data are difficult to create. As part of an ongoing global effort to establish new brain banks providing both high-quality neuroimaging scans and detailed histopathology examinations, the South Texas Alzheimer's Disease Re- search Center postmortem repository was recently created with the specific purpose of studying comorbid dementias. As the repository is reaching a milestone of two hundred brain donations and a hundred curated MRI sessions are ready for processing, robust statistical analyses can now be conducted. In this work, we report the very first morphometry analysis conducted with this new data set. We describe the processing pipelines that were specifically developed to exploit the available MRI sequences, and we explain how we addressed several postmortem neuroimaging challenges, such as the separation of brain tissues from fixative fluids, the need for updated brain atlases, and the tissue contrast changes induced by brain fixation. In general, our results establish that a combination of structural MRI sequences can provide enough informa- tion for state-of-the-art Deep Learning algorithms to almost perfectly separate brain tissues from a formalin buffered solution. Regional brain volumes are challenging to measure in postmortem scans, but robust estimates sensitive to sex differences and age trends, reflecting clinical diagnosis, neuropathology findings, and the shrinkage induced by tissue fixation can be obtained. We hope that the new processing methods developed in this work, such as the lightweight Deep Networks we used to identify the formalin signal in multimodal MRI scans and the MRI synthesis tools we used to fix our anisotropic resolution brain scans, will inspire other research teams working with postmortem MRI scans.

PMID:39987858 | DOI:10.1016/j.nicl.2025.103752

Mucosal Immunol. 2025 Feb 21:S1933-0219(25)00023-6. doi: 10.1016/j.mucimm.2025.02.005. Online ahead of print.

ABSTRACT

Interleukin-10 (IL-10) is an immunomodulatory cytokine critical for intestinal immune homeostasis. IL-10 is produced by various immune cells but IL-10 receptor signaling in intestinal CX3CR1+ mononuclear phagocytes is necessary to prevent spontaneous colitis in mice. Here, we utilized fluorescent protein reporters and cell-specific targeting and found that Rorc-expressing innate lymphoid cells (ILCs) produce IL-10 in response to anti-CD40-mediated intestinal inflammation. Deletion of Il10 specifically in Rorc-expressing ILCs led to phenotypic changes in intestinal macrophages and exacerbated both innate and adaptive immune-mediated models of experimental colitis. The population of IL-10+ producing ILCs shared markers with both ILC2 and ILC3 with nearly all ILC3s being of NCR+ subtype. Interestingly, Ccl26 was enriched in IL-10+ ILCs and markedly reduced in IL-10-deficient ILC3s. Since CCL26 is a ligand for CX3CR1, we employed RNA in situ hybridization and observed increased numbers of ILCs in close proximity to Cx3cr1-expressing cells under inflammatory conditions. Finally, we generated a transgenic RorctdTomato reporter mouse that faithfully marked RORγt+ cells that could rescue disease pathology and aberrant macrophage phenotype following adoptive transfer into mice with selective Il10 deficiency in ILC3s. These results demonstrate that IL-10 production by a population of ILCs functions to promote immune homeostasis in the intestine possibly via direct effects on intestinal macrophages.

PMID:39988202 | DOI:10.1016/j.mucimm.2025.02.005

Metab Eng. 2025 Feb 21:S1096-7176(25)00024-2. doi: 10.1016/j.ymben.2025.02.009. Online ahead of print.

ABSTRACT

Polyhydroxyalkanoates (PHA) with high 3-hydroxyvalerate (3HV) monomer ratios lead to their accelerated biodegradation and improved thermal and mechanical properties. In this study, poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) with a broad range of 3HV ratios were produced and characterized using the next generation industrial biotechnology (NGIB) chassis Halomonas bluephagenesis (H. bluephagenesis). Wild type H. bluephagenesis was found to produce P(3HB-co-66.31mol% 3HV) when cultured in the presence of valerate. Deletion on the functional enoyl-CoA hydratase (fadB1) increased to 93.11 mol% 3HV in the PHBV copolymers. Through tuning the glucose and valerate co-feeding, PHBV with controllable 3HV ratios were adjusted to range from 0-to-93.6 mol% in shake-flask studies. Metabolic weakening of the β-oxidation pathway paired with flux limitation to the native 3HB synthesis pathway were used to reach the highest reported 98.3 mol% 3HV by H. bluephagenesis strain G34B grown in shake flasks. H. bluephagenesis strain G34B was grown to 71.42 g/L cell dry weight (CDW) containing 74.12 wt% P(3HB-co-17.97 mol% 3HV) in 7 L fermentors. Mechanical properties of PHBV with 0, 22.81, 42.76, 73.49 and 92.17 mol% 3HV were characterized to find not linearly related to increased 3HV ratios. Engineered H. bluephagenesis has demonstrated as a platform for producing PHBV of various properties.

PMID:39988026 | DOI:10.1016/j.ymben.2025.02.009

ISME J. 2025 Feb 23:wraf034. doi: 10.1093/ismejo/wraf034. Online ahead of print.

ABSTRACT

Numerous studies and clinical applications have underscored the therapeutic potential of the indigenous gut bacterium Akkermansia muciniphila in various diseases. However, our understanding of how A. muciniphila senses and responds to host gastrointestinal signals remains limited. Here, we demonstrate that A. muciniphila exhibits rapid growth, facilitated by its self-produced carbon dioxide, with key enzymes such as glutamate decarboxylase, carbonic anhydrase, and pyruvate ferredoxin oxidoreductase playing pivotal roles. Additionally, we design a novel delivery system, comprising calcium carbonate, inulin, A. muciniphila, and sodium alginate, which enhances A. muciniphila growth and facilitates the expression of part probiotic genes in mice intestinal milieu. Notably, the administration of this delivery system induces weight loss in mice fed high-fat diets. Furthermore, we elucidate the significant impact of carbon dioxide on the composition and functional genes of the human gut microbiota, with genes encoding carbonic anhydrase and amino acid metabolism enzymes exhibiting heightened responsiveness. These findings reveal a novel mechanism by which gut commensal bacteria sense and respond to gaseous molecules, thereby promoting growth. Moreover, they suggest the potential for designing rational therapeutic strategies utilizing live bacterial delivery systems to enhance probiotic growth and ameliorate gut microbiota-related diseases.

PMID:39987558 | DOI:10.1093/ismejo/wraf034

Br J Clin Pharmacol. 2025 Feb 23. doi: 10.1002/bcp.70022. Online ahead of print.

ABSTRACT

AIMS: The primary objective of this study was to determine the pharmacokinetics, mass balance and biotransformation of [14C]GT0918 in humans after the drug was administered to healthy Chinese male subjects.

METHODS: The absorption, metabolism, and excretion (AME) of GT0918 was characterized via isotope labelling technology in six healthy Chinese male subjects after receiving a single 200 mg oral dose of [14C]GT0918 (80 μCi), and the phenotype, together with the metabolic mechanism of GT0918, was confirmed in vitro.

RESULTS: The medium Tmax of total radioactivity was 6.00 h (4.00-8.00 h) post-dose, and the mean Cmax was 10.5 μg eq./mL (8.7-12.3 μg eq./mL) in plasma. Drug-related components in the plasma were eliminated slowly, with a mean t1/2 of 67.7 h (54.4-90.7 h), and the radioactivity of the plasma samples from some subjects was above the below the quantization limit (BQL) until 17 days post-dose. After 19 days of dosing, the mean cumulative excreted radioactivity was 82.81% (79.07-86.07%) of the dose, including 29.47% (26.71-32.02%) in urine and 53.34% (52.01-55.62%) in faeces, indicating that the drug-related components of GT0918 were mainly excreted by faeces. Metabolite profiling revealed that the parent drug was detected in plasma, as well as in faeces and not in urine. In plasma, the most abundant metabolite was GT0955, a mono-oxidative metabolite of GT0918; in urine, the primary metabolite was GT0795, a metabolite of oxazole ring-opening followed by N-dealkylation; in faeces, the two main metabolites were M551 and the glucuronidation of GT0955. The majority of the metabolites were formed via an important aldehyde intermediate derived from the oxazole ring-opening, and the intermediate was trapped by methoxyamine hydrochloride in the in-vitro study. CYP3A4 is the main enzyme involved in the metabolism of GT0918.

CONCLUSIONS: Overall, all the dosed subjects completed the study, and GT0918 was found to be safe, with no grade II or above adverse events reported. A total dose of 82.81% was quantified in the urine (29.47%) and faeces (53.34%) of healthy adult male subjects after a single oral administration of 200 mg (80 μCi) GT0918 ([14C]GT0918). The metabolism of GT0918 is catalysed predominantly by CYP3A4, and an uncommon pathway of oxazole ring-opening to an aldehyde intermediate has also been proposed.

PMID:39987943 | DOI:10.1002/bcp.70022

Cell Genom. 2025 Feb 15:100774. doi: 10.1016/j.xgen.2025.100774. Online ahead of print.

ABSTRACT

DNA methylation (DNAm) is a key epigenetic mark with essential roles in gene regulation, mammalian development, and human diseases. Single-cell technologies enable profiling DNAm at cytosines in individual cells, but they often suffer from low coverage for CpG sites. We introduce scMeFormer, a transformer-based deep learning model for imputing DNAm states at each CpG site in single cells. Comprehensive evaluations across five single-nucleus DNAm datasets from human and mouse demonstrate scMeFormer's superior performance over alternative models, achieving high-fidelity imputation even with coverage reduced to 10% of original CpG sites. Applying scMeFormer to a single-nucleus DNAm dataset from the prefrontal cortex of patients with schizophrenia and controls identified thousands of schizophrenia-associated differentially methylated regions that would have remained undetectable without imputation and added granularity to our understanding of epigenetic alterations in schizophrenia. We anticipate that scMeFormer will be a valuable tool for advancing single-cell DNAm studies.

PMID:39986279 | DOI:10.1016/j.xgen.2025.100774

Am J Hum Genet. 2025 Feb 19:S0002-9297(25)00019-9. doi: 10.1016/j.ajhg.2025.01.019. Online ahead of print.

ABSTRACT

The case-control study is a widely used method for investigating the genetic underpinnings of binary traits. However, long-term, prospective cohort studies often grapple with absent or evolving health-related outcomes. Here, we propose two methods, liability and meta, for conducting genome-wide association studies (GWASs) that leverage disease liabilities calculated from deep patient phenotyping. Analyzing 38 common traits in ∼300,000 UK Biobank participants, we identified an increased number of loci in comparison to the number identified by the conventional case-control approach, and there were high replication rates in larger external GWASs. Further analyses confirmed the disease specificity of the genetic architecture; the meta method demonstrated higher robustness when phenotypes were imputed with low accuracy. Additionally, polygenic risk scores based on disease liabilities more effectively predicted newly diagnosed cases in the 2022 dataset, which were controls in the earlier 2019 dataset. Our findings demonstrate that integrating high-dimensional phenotypic data into deep neural networks enhances genetic association studies while capturing disease-relevant genetic architecture.

PMID:39986278 | DOI:10.1016/j.ajhg.2025.01.019

Int J Med Inform. 2025 Feb 19;197:105843. doi: 10.1016/j.ijmedinf.2025.105843. Online ahead of print.

ABSTRACT

BACKGROUND: Despite the proliferation of heart failure (HF) mortality prediction models, their practical utility is limited. Addressing this, we utilized a significant dataset to develop and validate a deep learning artificial intelligence (AI) model for predicting one-year mortality in heart failure with reduced ejection fraction (HFrEF) patients. The study's focus was to assess the effectiveness of an AI algorithm, trained on an extensive collection of ECG data, in predicting one-year mortality in HFrEF patients.

METHODS: We selected HFrEF patients who had high-quality baseline ECGs from two hospital visits between September 2016 and May 2021. A total of 3,894 HFrEF patients (64% male, mean age 64.3, mean ejection fraction 29.8%) were included. Using this ECG data, we developed a deep learning model and evaluated its performance using the area under the receiver operating characteristic curve (AUROC).

RESULTS: The model, validated against 16,228 independent ECGs from the original cohort, achieved an AUROC of 0.826 (95 % CI, 0.794-0.859). It displayed a high sensitivity of 99.0 %, positive predictive value of 16.6 %, and negative predictive value of 98.4 %. Importantly, the deep learning algorithm emerged as an independent predictor of 1-yr mortality of HFrEF patients with an adjusted hazards ratio of 4.12 (95 % CI 2.32-7.33, p < 0.001).

CONCLUSIONS: The depth and quality of our dataset and our AI-driven ECG analysis model significantly enhance the prediction of one-year mortality in HFrEF patients. This promises a more personalized, future-focused approach in HF patient management.

PMID:39986123 | DOI:10.1016/j.ijmedinf.2025.105843

Food Chem. 2025 Feb 12;476:143293. doi: 10.1016/j.foodchem.2025.143293. Online ahead of print.

ABSTRACT

Herein, we report a sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) method featuring dual imaging and signal-fusion deep learning for specific identification and analysis of glycomacropeptide (GMP) in milk sample. Conventional SDS-PAGE methods lack specificity because of the signle staining of protein bands, and the overlap between GMP and β-lactoglobulin (βLg). Our dual imaging method generated a pair of complementary detection signals by recruiting intrinsic fluorescence imaging (IFI) and silver staining. Comparing the IFI image with the staining image highlighted the presence of GMP and differentiated it from βLg. Additionally, we trained a signal-fusion deep learning model to improve the quantitative performance of our method. The model fused the features extracted from the paired detection signals (IFI and staining) and accurately classified them into different mixing ratios (proportion of GMP-containing whey in the sample), indicating the potential for quantitative analysis on the mixing ratios of GMP added into whey sample. The developed method has the merits of specificity, sensitivity and simplilcity, and has potential to analysis of protein/peptides with unique IFI properties in food safety, basic research and biopharming etc.

PMID:39986063 | DOI:10.1016/j.foodchem.2025.143293

Microb Pathog. 2025 Feb 20:107398. doi: 10.1016/j.micpath.2025.107398. Online ahead of print.

ABSTRACT

The COVID-19 disease has spread rapidly across the world within just six months, affecting 169 million people and causing 3.5 million deaths globally (2021). The most affected countries include the USA, Brazil, India, and several European countries such as the UK and Russia. Healthcare professionals face new challenges in finding better ways to manage patients and save lives. In this regard, more comprehensive research is needed, including genomic and proteomic studies, personalized medicines and the design of suitable treatments. However, finding novel molecular entities (NME) using a standard or de novo strategy to drug development is a time-consuming and costly process. Another alternate strategy is discovering new therapeutic uses for old/existing/available medications, known as drug repurposing. There are a variety of computational repurposing methodologies, and some of them have been used to counter the coronavirus disease pandemic of 2019 (COVID-19). This review article compiles recently published data on the origin, transmission, pathogenesis, diagnosis, and management of the coronavirus by drug repurposing and vaccine development approach. We have attempted to screen probable drugs in clinical trials by using literature survey. This systematic review aims to create priorities for future research of drugs repurposed and vaccine development for COVID-19.

PMID:39986548 | DOI:10.1016/j.micpath.2025.107398

Ann Rheum Dis. 2025 Feb 19:S0003-4967(25)00184-0. doi: 10.1016/j.ard.2025.01.034. Online ahead of print.

ABSTRACT

OBJECTIVES: Mosaic chromosomal alterations (mCAs) increase with age and are associated with age-related diseases. The association between mCAs and rheumatoid arthritis (RA), particularly late-onset RA (LORA), has not been explored.

METHODS: mCAs were detected in peripheral blood samples from 2 independent Japanese datasets (Set 1: 2107 RA cases and 86,998 controls; Set 2: 2359 RA cases and 86,998 controls). The associations between mCAs and RA were evaluated in each dataset using logistic regression models and meta-analysis. In each dataset, the effect sizes of mosaic loss of Y (mLOY) and polygenic risk score (PRS) of RA in males was evaluated, and a meta-analysis was subsequently performed. The interaction between mLOY and PRS was assessed. These models were applied separately to RA, LORA, and young-onset RA (YORA).

RESULTS: mLOY increased significantly in LORA (odds ratio [OR] = 1.43, P = .0070). We observed a negative association between mLOY and YORA (OR = 0.66, P = .0034). On the other hand, we found consistently negative associations of autosomal mCAs or mosaic loss of X with RA, LORA, and YORA. The PRS effect sizes were lower for LORA than for YORA. mLOY with a high cell fraction strengthened the association between PRS and LORA (P = .0036), whereas the association with YORA was independent of mLOY.

CONCLUSIONS: LORA was characterised by the presence of a high burden of mLOY. The observed interaction between mLOY and PRS in LORA, but not in YORA, supports different gene-environment interactions between the subsets. These data suggest that distinct pathophysiological mechanisms underlie the development of LORA and YORA.

PMID:39986957 | DOI:10.1016/j.ard.2025.01.034

Sci Rep. 2025 Feb 22;15(1):6520. doi: 10.1038/s41598-025-91368-3.

ABSTRACT

The study of key Pseudomonas aeruginosa (PA) virulence factors, the molecular basis of pathogenicity, as well as their correlation with the immune response during exacerbations in patients with non-cystic fibrosis bronchiectasis can help to identify novel targets and biomarkers for clinical management. The objective was to compare P. aeruginosa virulence and the patient's immune response during stable phases and exacerbations of bronchiectasis. We used polymerase chain reaction (PCR) and real-time quantitative PCR (qRT-PCR) to perform molecular characterization of the genomic islands and virulence genes present in 42 P. aeruginosa strains obtained from the sputum of patients with bronchiectasis during stability and exacerbations. Immunoglobulin (Ig) and interleukin (IL) levels in 32 serum samples were analyze by ELISA and Luminex assay. A greater presence of the conjugative element pKLC102, specific virulence genes (exoS, exoY) and pyoverdine production characterize the P. aeruginosa strains obtained during exacerbations. The expression levels of type III secretion system (exoS, exoY) showed an important role in the humoral immune response during exacerbations. Exacerbations were associated with high levels of IL-6. The presence of specific genomic islands, virulence genes, and increased IL-6 levels provide an accurate characterization on bronchiectasis exacerbations. These targets could be useful in the prevention, management and treatment of these exacerbations.

PMID:39987197 | DOI:10.1038/s41598-025-91368-3

J Cyst Fibros. 2025 Feb 21:S1569-1993(25)00066-9. doi: 10.1016/j.jcf.2025.02.016. Online ahead of print.

ABSTRACT

BACKGROUND: Symptoms of depression and anxiety can contribute to lower medical treatment adherence. Given that people with cystic fibrosis (PWCF) have higher rates of depressive and anxiety symptoms than those without cystic fibrosis (CF), this study examined factors that mediated the association between mental health and adherence.

METHODS: Participants were 294 adults (M age=25 years) with CF who were enrolled in the Daily Care Check-in Validation Study. Participants completed in-clinic questionnaires that assessed depressive and anxiety symptoms, perceived barriers to self-management, and medication self-efficacy. Medication adherence was measured by pharmacy refill data. Parallel mediation models assessed perceived barriers and medication self-efficacy as mediators between depressive symptoms and adherence, and between anxiety symptoms and adherence.

RESULTS: Perceived interference of barriers to self-management significantly mediated the association between depressive symptoms and adherence (β =-0.005, SE=0.002, 95 % CI [-0.009, -0.001]), and between anxiety symptoms and adherence (β=-0.005, SE=0.003, 95 % CI [-0.008, -0.001]). Additionally, self-efficacy significantly mediated the association between depressive symptoms and adherence (β=-0.004, SE=0.001, 95 % CI [-0.007, -0.002]), and between anxiety symptoms and adherence (β=-0.004, SE=0.001, 95 % CI [-0.007, -0.001]).

CONCLUSIONS: This study found that when PWCF experienced mental health symptoms (either anxiety or depression), they were likely to report more interference from barriers to disease management or experience less medication self-efficacy, which was related to worse adherence. Building self-efficacy around taking medications may reduce the impact that mental health symptoms have on adherence. Care teams should also work with PWCF to minimize the impact of barriers on daily therapies.

PMID:39986976 | DOI:10.1016/j.jcf.2025.02.016

J Cyst Fibros. 2025 Feb 21:S1569-1993(25)00057-8. doi: 10.1016/j.jcf.2025.02.010. Online ahead of print.

ABSTRACT

CFTR modulators represent the international standard of care for the treatment of cystic fibrosis (CF). Yet due to prices of over $250,000 per year they are functionally inaccessible for people with CF (pwCF) unless reimbursed by healthcare systems. Current prices are unaffordable for payors in almost all low- and middle-income countries (LMICs) worldwide, and resulting disparities in access are widening existing global health inequities. In comparable situations in other therapeutic areas, patients have successfully developed strategies to bypass national reimbursement systems and gain access to treatment. We therefore undertook an international survey of CF clinicians in 15 countries where CFTR modulators are not reimbursed, to characterise alternative means of accessing modulator therapy. Successful methods were identified in 11 countries, and could broadly be categorised into legal challenges to access originator modulators, use of generic formulations, and access via donations. Aside from domestically produced generics used in Argentina and an originator-led donation program in Ukraine, these methods were only able to provide treatment to limited proportions of the local CF population due to significant associated financial costs. Accordingly, they are generally not sustainable or widely applicable, and fail to address the underlying structural issues driving international disparities in access. Twelve years after the initial marketing of CFTR modulators, pwCF in LMICs are being forced to take extraordinary measures to access disease-modifying treatment. Corrective measures are urgently required to overcome barriers posed by restrictive patents and prohibitively high prices, and to promote global health equity for pwCF.

PMID:39986975 | DOI:10.1016/j.jcf.2025.02.010

Sci Rep. 2025 Feb 22;15(1):6499. doi: 10.1038/s41598-025-91206-6.

ABSTRACT

Building rooftop extraction has been applied in various fields, such as cartography, urban planning, automatic driving, and intelligent city construction. Automatic building detection and extraction algorithms using high spatial resolution aerial images can provide precise location and geometry information, significantly reducing time, costs, and labor. Recently, deep learning algorithms, especially convolution neural networks (CNNs) and Transformer, have robust local or global feature extraction ability, achieving advanced performance in intelligent interpretation compared with conventional methods. However, buildings often exhibit scale variation, spectral heterogeneity, and similarity with complex geometric shapes. Hence, the building rooftop extraction results exist fragmentation and lack spatial details using these methods. To address these issues, this study developed a multi-scale global perceptron network based on Transformer and CNN using novel encoder-decoders for enhancing contextual representation of buildings. Specifically, an improved multi-head-attention encoder is employed by constructing multi-scale tokens to enhance global semantic correlations. Meanwhile, the context refinement decoder is developed and synergistically uses high-level semantic representation and shallow features to restore spatial details. Overall, quantitative analysis and visual experiments confirmed that the proposed model is more efficient and superior to other state-of-the-art methods, with a 95.18% F1 score on the WHU dataset and a 93.29% F1 score on the Massub dataset.

PMID:39987354 | DOI:10.1038/s41598-025-91206-6

Sci Rep. 2025 Feb 22;15(1):6501. doi: 10.1038/s41598-025-91364-7.

ABSTRACT

Accurate counting of Amorphophallus konjac (Konjac) plants can offer valuable insights for agricultural management and yield prediction. While current studies have primarily focused on detecting and counting crop plants during the early stages of low coverage, there is limited investigation into the later stages of high coverage, which could impact the accuracy of forecasting yield. High canopy coverage and severe occlusion in later stages pose significant challenges for plant detection and counting. Therefore, this study evaluated the performance of the Count Crops tool and a deep learning (DL) model derived from early-stage unmanned aerial vehicle (UAV) imagery in detecting and counting Konjac plants during the high-coverage growth stage. Additionally, the study proposed an approach that integrates the DL model with Konjac location information from both early-stage and high canopy coverage stage imagery to improve the accuracy of recognizing Konjac plants during the high canopy coverage stage. The results indicated that the Count Crops tool outperformed the DL model constructed solely from early-stage imagery in detecting and counting Konjac plants during the high-coverage period. However, given the single stem and erect growth characteristics of Konjac, incorporating the DL model with the location information of the Konjac plants achieved the highest accuracy (Precision = 98.7%, Recall = 86.7%, F1-score = 92.3%). Our findings indicate that combining DL detection results from the early growth stages of Konjac, along with plant positional information from both growth stages, not only significantly improved the accuracy of detecting and counting plants but also saved time on annotating and training DL samples in the later stages. This study introduces an innovative approach for detecting and counting Konjac plants during high-coverage periods, providing a new perspective for recognizing and counting other crop plants at high-overlapping growth stages.

PMID:39987316 | DOI:10.1038/s41598-025-91364-7

NPJ Digit Med. 2025 Feb 22;8(1):120. doi: 10.1038/s41746-025-01491-8.

ABSTRACT

Hypertension is a major risk factor for cardiovascular disease (CVD), yet blood pressure is measured intermittently and under suboptimal conditions. We developed a deep learning model to identify hypertension and stratify risk of CVD using 12-lead electrocardiogram waveforms. HTN-AI was trained to detect hypertension using 752,415 electrocardiograms from 103,405 adults at Massachusetts General Hospital. We externally validated HTN-AI and demonstrated associations between HTN-AI risk and incident CVD in 56,760 adults at Brigham and Women's Hospital. HTN-AI accurately discriminated hypertension (internal and external validation AUROC 0.803 and 0.771, respectively). In Fine-Gray regression analyses model-predicted probability of hypertension was associated with mortality (hazard ratio per standard deviation: 1.47 [1.36-1.60], p < 0.001), HF (2.26 [1.90-2.69], p < 0.001), MI (1.87 [1.69-2.07], p < 0.001), stroke (1.30 [1.18-1.44], p < 0.001), and aortic dissection or rupture (1.69 [1.22-2.35], p < 0.001) after adjustment for demographics and risk factors. HTN-AI may facilitate diagnosis of hypertension and serve as a digital biomarker of hypertension-associated CVD.

PMID:39987256 | DOI:10.1038/s41746-025-01491-8

Sci Rep. 2025 Feb 22;15(1):6506. doi: 10.1038/s41598-025-90221-x.

ABSTRACT

Pathologists have depended on their visual experience to assess tissue structures in smear images, which was time-consuming, error-prone, and inconsistent. Deep learning, particularly Convolutional Neural Networks (CNNs), offers the ability to automate this procedure by recognizing patterns in tissue images. However, training these models necessitates huge amounts of labeled data, which can be difficult to come by due to the skill required for annotation and the unavailability of data, particularly for rare diseases. This work introduces a new semi-supervised method for tissue structure semantic segmentation in histopathological images. The study presents a CNN based teacher model that generates pseudo-labels to train a student model, aiming to overcome the drawbacks of conventional supervised learning approaches. Self-supervised training is used to improve the teacher model's performance on smaller datasets. Consistency regularization is integrated to efficiently train the student model on labeled data. Further, the study uses Monte Carlo dropout to estimate the uncertainty of proposed model. The proposed model demonstrated promising results by achieving an mIoU score of 0.64 on a public dataset, highlighting its potential to improve segmentation accuracy in histopathological image analysis.

PMID:39987243 | DOI:10.1038/s41598-025-90221-x

Sci Rep. 2025 Feb 22;15(1):6456. doi: 10.1038/s41598-025-90924-1.

ABSTRACT

To address the issue of serious inefficiency in the traditional manual evaluation methods of rock core integrity, a deep learning-based algorithm named IDA-RCF (Intelligent detection algorithm for Rock Core Fissure) is proposed in this paper, which realizes the automatic evaluation of rock core integrity in accordance with the fissure identification results. In IDA-RCF, a two-branch feature extraction network is firstly proposed, in which branch one is used to fully extract the complex and variable local detail fissure features by Deformable convolution, and branch two is used to capture the global context information of the rock core images by EfficientViT network based on the self-attention. Then a multi-level feature fusion network is proposed for adaptively fusing local and global features from the same level and the fused feature information from the previous level, thereby capturing more valid information and eliminating redundancies. Then the fused feature layer is decoded by the feature decoder to output the detection results of rock core fissure. Finally, the fissure rate is automatically calculated based on the detection results to predict the degree of rock core integrity. The experimental results show that the accuracy indexes F1, mAP@0.5 and mAP@0.5:0.95 of IDA-RCF are 93.09%, 94.44% and 84.61%, respectively. The relative error between the prediction results and the manual statistical results of the fissure rate is only 4.38%, and the prediction accuracy for the degree of rock core integrity is 93.8%, indicating that the proposed method in this paper is able to accomplish the intelligent evaluation task of rock core integrity with high precision.

PMID:39987183 | DOI:10.1038/s41598-025-90924-1

Sci Rep. 2025 Feb 22;15(1):6490. doi: 10.1038/s41598-025-91322-3.

ABSTRACT

Chest computed tomography (CT) scans are essential for accurately assessing the severity of the novel Coronavirus (COVID-19), facilitating appropriate therapeutic interventions and monitoring disease progression. However, determining COVID-19 severity requires a radiologist with significant expertise. This study introduces a pioneering utilization of deep learning (DL) for evaluate COVID-19 severity using lung CT images, presenting a novel and effective method for assessing the severity of pulmonary manifestations in COVID-19 patients. Inception-Residual networks (Inception-ResNet), advanced hybrid models known for their compactness and effectiveness, were used to extract relevant features from CT scans. Inception-ResNet incorporates the dilated mechanism into its ResNet component, enhancing its ability to accurately classify lung involvement stages. This study demonstrates that dilated residual networks (dResNet) outperform their non-dilated counterparts in image classification tasks, as their architectural designs allow the systems to acquire comprehensive global data by expanding their receptive fields. Our study utilized an initial dataset of 1548 human thoracic CT scans, meticulously annotated by two experienced specialists. Lung involvement was determined by calculating a percentage based on observations made at each scan. The hybrid methodology successfully distinguished the ten distinct severity levels associated with COVID-19, achieving a maximum accuracy of 96.40%. This system demonstrates its effectiveness as a diagnostic framework for assessing lung involvement in COVID-19-affected individuals, facilitating disease progression tracking.

PMID:39987169 | DOI:10.1038/s41598-025-91322-3