Crit Rev Food Sci Nutr. 2025 Mar 7:1-19. doi: 10.1080/10408398.2025.2474183. Online ahead of print.

ABSTRACT

Tea is one of the most popular drinks due to its distinct flavor and numerous health benefits. The quality of tea is closely related to production processing. Human sensory evaluation is the conventional method for quality monitoring in tea processing. However, this method is subjective and susceptible to environmental influences. Therefore, visible/near-infrared spectroscopy (Vis/NIRS) and hyperspectral imaging (HSI) techniques offer great potential due to their rapid detection speed, nondestructive, low cost, and simple operations. Artificial intelligence (AI) is one of the most promising methodological approaches for spectral analysis and decision-making of automated production. Vis/NIRS and HSI techniques assisted by AI further promote the progress of quality monitoring in tea processing. This paper reviewed the updated applications of Vis/NIRS and HSI techniques assisted by AI for quality monitoring in tea processing from 2019 to 2025. In particular, the tea production process, theories of Vis/NIRS and HSI techniques, and AI algorithms in spectral analysis are briefly introduced. Furthermore, the recent applications of Vis/NIRS and HSI techniques assisted by AI in tea processing quality monitoring are summarized and discussed. Finally, the challenges and future trends of Vis/NIRS and HSI techniques associated with their practical application in the tea industry are presented.

PMID:40053139 | DOI:10.1080/10408398.2025.2474183

Abdom Radiol (NY). 2025 Mar 7. doi: 10.1007/s00261-025-04833-y. Online ahead of print.

ABSTRACT

OBJECTIVE: The objective of this study is to investigate the value of radiomics features and deep learning features based on positron emission tomography/computed tomography (PET/CT) in predicting perineural invasion (PNI) in rectal cancer.

METHODS: We retrospectively collected 120 rectal cancer (56 PNI-positive patients 64 PNI-negative patients) patients with preoperative 18F-FDG PET/CT examination and randomly divided them into training and validation sets at a 7:3 ratio. We also collected 31 rectal cancer patients from two other hospitals as an independent external validation set. χ2 test and binary logistic regression were used to analyze PET metabolic parameters. PET/CT images were utilized to extract radiomics features and deep learning features. The Mann-Whitney U test and LASSO were employed to select valuable features. Metabolic parameter, radiomics, deep learning and combined models were constructed. ROC curves were generated to evaluate the performance of models.

RESULTS: The results indicate that metabolic tumor volume (MTV) is correlated with PNI (P = 0.001). In the training set and validation set, the AUC values of the metabolic parameter model were 0.673 (95%CI: 0.572-0.773), 0.748 (95%CI: 0.599-0.896). We selected 16 radiomics features and 17 deep learning features as valuable factors for predicting PNI. The AUC values of radiomics model and deep learning model were 0.768 (95%CI: 0.667-0.868) and 0.860 (95%CI: 0.780-0.940) in the training set. And the AUC values in the validation set were 0.803 (95%CI: 0.656-0.950) and 0.854 (95% CI 0.721-0.987). Finally, the combined model exhibited AUCs of 0.893 (95%CI: 0.825-0.961) in the training set and 0.883 (95%CI: 0.775-0.990) in the validation set. In the external validation set, the combined model achieved an AUC of 0.829 (95% CI: 0.674-0.984), outperforming each individual model. The decision curve analysis of these models indicated that using the combined model to guide treatment provided a substantial net benefit.

CONCLUSIONS: This combined model established by integrating PET metabolic parameters, radiomics features, and deep learning features can accurately predict the PNI in rectal cancer.

PMID:40053051 | DOI:10.1007/s00261-025-04833-y

Transl Vis Sci Technol. 2025 Mar 3;14(3):6. doi: 10.1167/tvst.14.3.6.

ABSTRACT

Optical coherence tomography angiography (OCTA) is a signal processing and scan acquisition approach that enables OCT devices to clearly identify vascular tissue down to the capillary scale. As originally proposed, OCTA included several important limitations, including small fields of view relative to allied imaging modalities and the presence of confounding artifacts. New approaches, including both hardware and software, are solving these problems and can now produce high-quality angiograms from tissue throughout the retina and choroid. Image analysis tools have also improved, enabling OCTA data to be quantified at high precision and used to diagnose disease using deep learning models. This review highlights these advances and trends in OCTA technology, focusing on work produced since 2020.

PMID:40052848 | DOI:10.1167/tvst.14.3.6

J Proteome Res. 2025 Mar 7. doi: 10.1021/acs.jproteome.4c00972. Online ahead of print.

ABSTRACT

Mass spectrometry (MS)-based proteomics relies heavily on MS/MS (MS2) data, which do not fully exploit the available MS1 information. Traditional peptide identity propagation (PIP) methods, such as match-between-runs (MBR), are limited to similar runs, particularly with the same liquid chromatography (LC) gradients, thus potentially underutilizing available proteomics libraries. We introduce SWAPS, a novel and modular MS1-centric framework incorporating advances in peptide property prediction, extensive proteomics libraries, and deep-learning-based postprocessing to enable and explore PIP across more diverse experimental conditions and LC gradients. SWAPS substantially enhances precursor identification, especially in shorter gradients. On the example of 30, 15, and 7.5 min gradients, SWAPS achieves increases of 46.3, 86.2, and 112.1% on precursor level over MaxQuant's MS2-based identifications. Despite the inherent challenges in controlling false discovery rates (FDR) with MS1-based methods, SWAPS demonstrates strong efficacy in deconvoluting MS1 signals, offering powerful discrimination and deeper sequence exploration, while maintaining quantitative accuracy. By building on and applying peptide property predictions in practical contexts, SWAPS reveals that current models, while advanced, are still not fully comparable to experimental measurements, sparking the need for further research. Additionally, its modular design allows seamless integration of future improvements, positioning SWAPS as a forward-looking tool in proteomics.

PMID:40052690 | DOI:10.1021/acs.jproteome.4c00972

Eur Heart J Cardiovasc Imaging. 2025 Mar 7:jeaf081. doi: 10.1093/ehjci/jeaf081. Online ahead of print.

ABSTRACT

AIMS: Understanding determinants of thoracic aortic morphology is crucial for precise diagnostics and therapeutic approaches. This study aimed to automatically characterize ascending aortic morphology based on 3D non-contrast-enhanced magnetic resonance angiography (NC-MRA) data from the epidemiological cross-sectional German National Cohort (NAKO) and to investigate possible determinants of mid-ascending aortic diameter (mid-AAoD).

METHODS AND RESULTS: Deep learning (DL) automatically segmented the thoracic aorta and ascending aortic length, volume, and diameter was extracted from 25,073 NC-MRAs. Statistical analyses investigated relationships between mid-AAoD and demographic factors, hypertension, diabetes, alcohol, and tobacco consumption. Males exhibited significantly larger mid-AAoD than females (M:35.5±4.8mm, F:33.3±4.5mm). Age and body surface area (BSA) were positively correlated with mid-AAoD (age: male: r²=0.20, p<0.001, female: r²=0.16, p<0.001; BSA: male: r²=0.08, p<0.001, female: r²=0.05, p<0.001). Hypertensive and diabetic subjects showed higher mid-AAoD (ΔHypertension = 2.9 ± 0.5mm; ΔDiabetes = 1.5 ± 0.6mm). Hypertension was linked to higher mid-AAoD regardless of age and BSA, while diabetes and mid-AAoD were uncorrelated across age-stratified subgroups. Daily alcohol consumption (male: 37.4±5.1mm, female: 35.0±4.8mm) and smoking history exceeding 16.5 pack-years (male: 36.6±5.0mm, female: 33.9±4.3mm) exhibited highest mid-AAoD. Causal analysis (Peter-Clark algorithm) suggested that age, BSA, hypertension, and alcohol consumption are possibly causally related to mid-AAoD, while diabetes and smoking are likely spuriously correlated.

CONCLUSIONS: This study demonstrates the potential of DL and causal analysis for understanding ascending aortic morphology. By disentangling observed correlations using causal analysis, this approach identifies possible causal determinants, such as age, BSA, hypertension, and alcohol consumption. These findings can inform targeted diagnostics and preventive strategies, supporting clinical decision-making for cardiovascular health.

PMID:40052574 | DOI:10.1093/ehjci/jeaf081

J Food Sci. 2025 Mar;90(3):e70116. doi: 10.1111/1750-3841.70116.

ABSTRACT

Chronic diseases such as obesity and hypertension due to malnutrition can be prevented by following the appropriate diet, correct diet intake with correct measuring portion size, and developing healthy eating habits. Having a system that can automatically measure food consumption is important to determine whether individual nutritional needs are being met in order to accurately diagnose and solve nutritional problems, act quickly, and minimize the risk of malnutrition due to the cross-cultural diversity of foods. In this study, a deep learning system has been developed and implemented for automatically grouping and classifying foods. Dishes from Turkish cuisine were chosen as a sample for application and testing. The deep learning method used in this system is convolutional neural network (CNN) models based on image recognition. This study developed and implemented a deep learning system using CNNs to classify food groups and estimate portion sizes of Turkish cuisine dishes, achieving accuracy rates of up to 80% for food group classification and 80.47% for portion estimation with the inclusion of data augmentation.

PMID:40052549 | DOI:10.1111/1750-3841.70116

Brief Bioinform. 2025 Mar 4;26(2):bbaf087. doi: 10.1093/bib/bbaf087.

ABSTRACT

Cancer neoantigens are peptides that originate from alterations in the genome, transcriptome, or proteome. These peptides can elicit cancer-specific T-cell recognition, making them potential candidates for cancer vaccines. The rapid advancement of proteomics technology holds tremendous potential for identifying these neoantigens. Here, we provided an up-to-date survey about database-based search methods and de novo peptide sequencing approaches in proteomics, and we also compared these methods to recommend reliable analytical tools for neoantigen identification. Unlike previous surveys on mass spectrometry-based neoantigen discovery, this survey summarizes the key advancements in de novo peptide sequencing approaches that utilize artificial intelligence. From a comparative study on a dataset of the HepG2 cell line and nine mixed hepatocellular carcinoma proteomics samples, we demonstrated the potential of proteomics for the identification of cancer neoantigens and conducted comparisons of the existing methods to illustrate their limits. Understanding these limits, we suggested a novel workflow for neoantigen discovery as perspectives.

PMID:40052441 | DOI:10.1093/bib/bbaf087

Rev Med Chil. 2024 Oct;152(10):1060-1066. doi: 10.4067/s0034-98872024001001060. Epub 2025 Feb 3.

ABSTRACT

Organized pneumonia (OP) is an uncommon disease included in the group of idiopathic interstitial pneumonias. It can be cryptogenic (COP) or secondary to various etiologies. Its diagnosis is complex and not standardized. There are no published Chilean series. We present a cohort of patients with pneumonia in organization treated at the National Thoracic Institute (INT).

AIM: To describe the characteristics of patients with OP in a Chilean center.

METHODS: Pathological registries from the INT were reviewed between 2013 and 2022. Clinical and radiological information was obtained from hospital records. Each case was reviewed by the research team. Data are described by means, absolute and relative frequencies.

RESULTS: From an initial list of 203 biopsies, 69 were obtained with clinical/radiological symptoms compatible with OP. The mean age of these subjects was 62 years, of which 33 (47.8%) were men and 36 (52.2%) women. Biopsies were obtained by transbronchial biopsy in 49 (71%) cases and surgical biopsy in 19 (27.5%) cases. In terms of etiology, 37 (53.6%) of them were considered cryptogenic, 12 (17.4%) secondary to the use of medication / drugs and 11 (15.9%) cases associated with connective tissue disease. Regarding treatment, 36 (52.2%) patients received oral steroids and 10 (14.5%) were treated with a mix of corticosteroids and immunosuppressors. In the long-term follow-up, there were 23 deaths in just over 6 years.

CONCLUSIONS: The reported series has similar characteristics to those reported in the literature. Most of the cases described in this series were classified as COP. The most common underlying etiologies were connective tissue diseases and medications. The most used treatment was corticosteroid alone or mixed with immunosuppressors.

PMID:40052979 | DOI:10.4067/s0034-98872024001001060

Elife. 2025 Mar 7;13:RP99545. doi: 10.7554/eLife.99545.

ABSTRACT

The principle of efficient coding posits that sensory cortical networks are designed to encode maximal sensory information with minimal metabolic cost. Despite the major influence of efficient coding in neuroscience, it has remained unclear whether fundamental empirical properties of neural network activity can be explained solely based on this normative principle. Here, we derive the structural, coding, and biophysical properties of excitatory-inhibitory recurrent networks of spiking neurons that emerge directly from imposing that the network minimizes an instantaneous loss function and a time-averaged performance measure enacting efficient coding. We assumed that the network encodes a number of independent stimulus features varying with a time scale equal to the membrane time constant of excitatory and inhibitory neurons. The optimal network has biologically plausible biophysical features, including realistic integrate-and-fire spiking dynamics, spike-triggered adaptation, and a non-specific excitatory external input. The excitatory-inhibitory recurrent connectivity between neurons with similar stimulus tuning implements feature-specific competition, similar to that recently found in visual cortex. Networks with unstructured connectivity cannot reach comparable levels of coding efficiency. The optimal ratio of excitatory vs inhibitory neurons and the ratio of mean inhibitory-to-inhibitory vs excitatory-to-inhibitory connectivity are comparable to those of cortical sensory networks. The efficient network solution exhibits an instantaneous balance between excitation and inhibition. The network can perform efficient coding even when external stimuli vary over multiple time scales. Together, these results suggest that key properties of biological neural networks may be accounted for by efficient coding.

PMID:40053385 | DOI:10.7554/eLife.99545

FASEB J. 2025 Mar 15;39(5):e70411. doi: 10.1096/fj.202402726R.

ABSTRACT

Viral infections can cause cellular dysregulation of metabolic reactions. Viruses alter host metabolism to meet their replication needs. The impact of viruses on specific metabolic pathways is not well understood, even in well-studied viruses, such as human adenovirus. Adenoviral infection is known to influence cellular glycolysis and respiration; however, global effects on overall cellular metabolism in response to infection are unclear. Furthermore, few studies have employed an untargeted approach, combining emphasis on viral dosage and infection. To address this, we employed untargeted metabolomics to quantify the dynamic metabolic shifts in fibroblasts infected with human adenovirus serotype 5 (HAdV-5) at three dosages (0.5, 1.0, and 2.0 multiplicity of infection [MOI]) and across 4 time points (6-, 12-, 24-, and 36-h post-infection [HPI]). The greatest differences in individual metabolites were observed at 6- and 12-h post-infection, correlating with the early phase of the HAdV-5 infection cycle. In addition to its effects on glycolysis and respiration, adenoviral infection downregulates cysteine and unsaturated fatty acid metabolism while upregulating aspects of purine metabolism. These results reveal specific metabolic pathways dysregulated by adenoviral infection and the associated dynamic shifts in metabolism, suggesting that viral infections alter energetics via profound changes in lipid, nucleic acid, and protein metabolism. The results revealed previously unconsidered metabolic pathways disrupted by HAdV-5 that can alter cellular metabolism, thereby prompting further investigation into HAdV mechanisms and antiviral targeting.

PMID:40052831 | DOI:10.1096/fj.202402726R

Brief Bioinform. 2025 Mar 4;26(2):bbaf082. doi: 10.1093/bib/bbaf082.

ABSTRACT

Cancer cells acquire necessary functional capabilities for malignancy through the influence of the nervous system. We evaluate the extent of neural infiltration within the tumor microenvironment (TME) across multiple cancer types, highlighting its role as a cancer hallmark. We identify cancer-related neural genes using 40 bulk RNA-seq datasets across 10 cancer types, developing a predictive score for cancer-related neural infiltration (C-Neural score). Cancer samples with elevated C-Neural scores exhibit perineural invasion, recurrence, metastasis, higher stage or grade, or poor prognosis. Epithelial cells show the highest C-Neural scores among all cell types in 55 single-cell RNA sequencing datasets. The epithelial cells with high C-Neural scores (epi-highCNs) characterized by increased copy number variation, reduced cell differentiation, higher epithelial-mesenchymal transition scores, and elevated metabolic level. Epi-highCNs frequently communicate with Schwann cells by FN1 signaling pathway. The co-culture experiment indicates that Schwann cells may facilitate cancer progression through upregulation of VDAC1. Moreover, C-Neural scores positively correlate with the infiltration of antitumor immune cells, indicating potential response for immunotherapy. Melanoma patients with high C-Neural scores may benefit from trametinib. These analyses illuminate the extent of neural influence within TME, suggesting potential role as a cancer hallmark and offering implications for effective therapeutic strategies against cancer.

PMID:40052442 | DOI:10.1093/bib/bbaf082

Front Cell Dev Biol. 2025 Feb 20;13:1552716. doi: 10.3389/fcell.2025.1552716. eCollection 2025.

ABSTRACT

INTRODUCTION: In vertebrate limb morphogenesis, wingless-related integration site (Wnt) proteins and fibroblast growth factors (Fgfs) secreted from the apical ectodermal ridge (AER) coordinate proximodistal outgrowth. Fgfs also sustain sonic hedgehog (Shh) in the zone of polarizing activity (ZPA). Shh directs anteroposterior patterning and expansion and regulates AER-Fgfs, establishing a positive regulatory feedback loop that is vital in sustaining limb outgrowth. The transcription factor LIM homeodomain 2 (Lhx2) is expressed in the distal mesoderm and coordinates AER and ZPA signals that control cellular proliferation, differentiation, and shaping of the developing limb. Yet how Lhx2 is transcriptionally regulated to support such functions has only been partially characterized.

METHODS/RESULTS: We have identified two limb-specific cis-regulatory modules (CRMs) active within the Lhx2 expression domain in the limb. Chromatin conformation analysis of the Lhx2 locus in mouse embryonic limb bud cells predicted CRMs-Lhx2 promoter interactions. Single-cell RNA-sequencing analysis of limb bud cells revealed co-expression of several Fgf-related Ets and Wnt-related Tcf/Lef transcripts in Lhx2-expressing cells. Additionally, disruption of Ets and Tcf/Lef binding sites resulted in loss of reporter-driven CRM activity. Finally, binding of β-catenin to both Lhx2-associated CRMs supports the associated binding of Tcf/Lef transcription factors.

DISCUSSION: These results suggest a role for Ets and Tcf/Lef transcription factors in the regulation of Lhx2 expression through these limb-specific Lhx2-associated CRMs. Moreover, these CRMs provide a mechanism for Fgf and Wnt signaling to localize and maintain distal Lhx2 expression during vertebrate limb development.

PMID:40052149 | PMC:PMC11882541 | DOI:10.3389/fcell.2025.1552716

In Silico Pharmacol. 2025 Mar 4;13(1):36. doi: 10.1007/s40203-025-00325-5. eCollection 2025.

ABSTRACT

Tuberculosis (TB), a major global health concern, even after significant advancements in diagnosis and treatment, causing millions of deaths annually and severely impacting the healthcare systems of developing nations. Moreover, the rise of drug-resistant strains further diminishes the efforts made to control the infection and to overcome this scenario, highly effective drugs are required. Identifying new therapeutic uses of existing drugs through drug repurposing can significantly shorten the time and cost. In the current study, using a computational experimental approach, near about 3104 FDA-approved drugs and active pharmaceutical ingredients from Selleckchem database were screened against Enhanced intracellular survival (Eis) protein, responsible for causing drug resistance by inhibiting the aminoglycoside drug activity. Based on the three-level screening and Molecular Mechanics generalized Born surface area (MM/GBSA) scores, five drugs including Isavuconazonium sulfate, Cefotiam Hexetil Hydrochloride, Enzastaurin (LY317615), Salbutamol sulfate (Albuterol), and Osimertinib (AZD9291) were considered as potential Eis inhibitors. The 500 ns MD simulation results revealed that all these Eis-drug complexes are stable, with minor structural arrangements and stable binding patterns. The PCA and FEL analysis also confirmed the structural stability of the complexes. Overall, these drugs displayed promising results as Eis inhibitors, that can be regarded as suitable candidates for experimental validation.

PMID:40051485 | PMC:PMC11880469 | DOI:10.1007/s40203-025-00325-5

Front Artif Intell. 2025 Feb 20;8:1557894. doi: 10.3389/frai.2025.1557894. eCollection 2025.

ABSTRACT

INTRODUCTION: Chronic pain affects approximately 30% of the global population, posing a significant public health challenge. Despite their widespread use, traditional pharmacological treatments, such as opioids and NSAIDs, often fail to deliver adequate, long-term relief while exposing patients to risks of addiction and adverse side effects. Given these limitations, medical cannabis has emerged as a promising therapeutic alternative with both analgesic and anti-inflammatory properties. However, its clinical efficacy is hindered by high interindividual variability in treatment response and elevated dropout rates.

METHODS: A comprehensive dataset integrating genetic, clinical, and pharmacological information was compiled from 542 Caucasian patients undergoing cannabis-based treatment for chronic pain. A machine learning (ML) model was developed and validated to predict therapy dropout. To identify the most influential factors driving dropout, SHapley Additive exPlanations (SHAP) analysis was performed.

RESULTS: The random forest classifier demonstrated robust performance, achieving a mean accuracy of 80% and a maximum of 86%, with an AUC of 0.86. SHAP analysis revealed that high final VAS scores and elevated THC dosages were the most significant predictors of dropout, both strongly correlated with an increased likelihood of discontinuation. In contrast, baseline therapeutic benefits, CBD dosages, and the CC genotype of the rs1049353 polymorphism in the CNR1 gene were associated with improved adherence.

DISCUSSION: Our findings highlight the potential of ML and pharmacogenetics to personalize cannabis-based therapies, improving adherence and enabling more precise management of chronic pain. This research paves the way for the development of tailored therapeutic strategies that maximize the benefits of medical cannabis while minimizing its side effects.

PMID:40051572 | PMC:PMC11882547 | DOI:10.3389/frai.2025.1557894

Physiol Rep. 2025 Mar;13(5):e70264. doi: 10.14814/phy2.70264.

ABSTRACT

Intestinal tuft cells have recently been the focus of many studies due to their function in chemosensation and type 2 immunity in human gastrointestinal diseases. This study investigated the impact of acute tuft cell loss on intestinal physiological function. Tuft cell deletion was induced in DCLK1-IRES-GFP-CreERT2/+;Rosa-DTA (DCLK1-DTA) mice by a single tamoxifen injection, concomitant with littermate controls. Transient deletion of intestinal and biliary tuft cells was maximal on day 4 and recovered by day 7 post tamoxifen. DCLK1-DTA mice presented with significantly shortened small intestinal length and greater body weight loss by day 4. The activity of Na+-dependent glucose transporter 1 (SGLT1) and cystic fibrosis transmembrane regulator (CFTR) was reduced. Correlated with tuft cell reduction, the frequency of cholecystokinin (CCK)+ enteroendocrine and intermediate secretory cells, which co-express Paneth and goblet cell markers, was increased. In the lamina propria, fewer mast cells and leukocytes were found in the Day 4 DCLK1-DTA mice compared to controls. Ablation of tuft cells may induce nutrient malabsorption through alterations in epithelial cell proliferation and differentiation, along with changes in the mucosal defense response. These observations identify a new role for tuft cells in regulating intestinal absorption and mucosal regeneration.

PMID:40051209 | DOI:10.14814/phy2.70264

Comput Methods Biomech Biomed Engin. 2025 Mar 7:1-14. doi: 10.1080/10255842.2025.2470809. Online ahead of print.

ABSTRACT

Acute kidney injury (AKI) is a prevalent clinical syndrome that causes over one-fifth of hospitalized patients worldwide to suffer from AKI. We proposed the GCAT, which aims to identify high-risk AKI patients in the hospital settings using the MIMIC-III dataset. Firstly, it fully explores the similarity of attribute features among a large number of patients and calculates the attribute similarity values between patients to generate a node similarity matrix. Then, it selects nodes with high similarity to construct a patient feature similarity network (PFSN). Experiments demonstrate that the GCAT achieves an accuracy of 88.57%, its effectiveness is superior to state-of-the-art methods.

PMID:40052403 | DOI:10.1080/10255842.2025.2470809

Glob Chang Biol. 2025 Mar;31(3):e70118. doi: 10.1111/gcb.70118.

ABSTRACT

Understanding the dynamics of soil respiration (Rs) and its environmental drivers is crucial for accurately modeling terrestrial carbon fluxes. However, current methodologies often lead to divergent estimates and rely on annual predictions that may overlook critical interactions occurring at seasonal scales. A critical knowledge gap lies in understanding how temporal resolution affects both Rs predictions and their environmental drivers. Here, we employ deep learning models to predict global Rs at monthly (MRM) and annual (ARM) scales from 1982 to 2018. We then consider three main drivers potentially affecting Rs, including temperature, precipitation, and a vegetation proxy (leaf area index; LAI). Our models demonstrate strong predictive capabilities with global Rs estimation of 79.4 ± 5.7 Pg C year-1 for the MRM and 78.3 ± 7.5 Pg C year-1 for ARM (mean ± SD). While the difference in global estimations between both models is small, there are notable disparities in the spatial contribution of dominant drivers. The MRM highlights an influence of both temperature and LAI, while the ARM emphasizes a dominant role of precipitation. These findings underscore the critical role of temporal resolution in capturing seasonal variations and identifying key Rs-environment relationships that annual models may obscure. High temporal resolution Rs predictions, such as those provided by the MRM, are essential for capturing nuanced seasonal interactions between Rs and its drivers, refining carbon flux models, detecting critical seasonal thresholds, and enhancing the reliability of future Earth system predictions. This work highlights the need for further research into monthly and seasonal Rs variations, as well as higher timescale resolutions, to advance our understanding of ecosystem carbon dynamics in a rapidly changing climate.

PMID:40052202 | DOI:10.1111/gcb.70118

Front Big Data. 2025 Feb 20;8:1515341. doi: 10.3389/fdata.2025.1515341. eCollection 2025.

ABSTRACT

Neurodegenerative diseases are chronic, progressive conditions that cause irreversible damage to the nervous system, particularly in aging populations. Early diagnosis is a critical challenge, as these diseases often develop slowly and without clear symptoms until significant damage has occurred. Recent advances in radiomics and genomics have provided valuable insights into the mechanisms of these diseases by identifying specific imaging features and genomic patterns. Radiogenomics enhances diagnostic capabilities by linking genomics with imaging phenotypes, offering a more comprehensive understanding of disease progression. The growing field of artificial intelligence (AI), including machine learning and deep learning, opens new opportunities for improving the accuracy and timeliness of these diagnoses. This review examines the application of AI-based radiogenomics in neurodegenerative diseases, summarizing key model designs, performance metrics, publicly available data resources, significant findings, and future research directions. It provides a starting point and guidance for those seeking to explore this emerging area of study.

PMID:40052173 | PMC:PMC11882605 | DOI:10.3389/fdata.2025.1515341

Front Oncol. 2025 Feb 20;15:1536039. doi: 10.3389/fonc.2025.1536039. eCollection 2025.

ABSTRACT

With the rapid development of the "Internet + Medical" model, artificial intelligence technology has been widely used in the analysis of medical images. Among them, the technology of using deep learning algorithms to identify features of ultrasound and pathological images and realize intelligent diagnosis of diseases has entered the clinical verification stage. This study is based on the application research of artificial intelligence technology in medical diagnosis and reviews the early screening and diagnosis of thyroid diseases. The cure rate of thyroid disease is high in the early stage, but once it deteriorates into thyroid cancer, the risk of death and treatment costs of the patient increase. At present, the early diagnosis of the disease still depends on the examination equipment and the clinical experience of doctors, and there is a certain misdiagnosis rate. Based on the above background, it is particularly important to explore a technology that can achieve objective screening of thyroid lesions in the early stages. This paper provides a comprehensive review of recent research on the early diagnosis of thyroid diseases using artificial intelligence technology. It integrates the findings of multiple studies and that traditional machine learning algorithms are widely used as research objects. The convolutional neural network model has a high recognition accuracy for thyroid nodules and thyroid pathological cell lesions. U-Net network model can significantly improve the recognition accuracy of thyroid nodule ultrasound images when used as a segmentation algorithm. This article focuses on reviewing the intelligent recognition technology of thyroid ultrasound images and pathological sections, hoping to provide researchers with research ideas and help clinicians achieve intelligent early screening of thyroid cancer.

PMID:40052126 | PMC:PMC11882420 | DOI:10.3389/fonc.2025.1536039

Front Sports Act Living. 2025 Feb 20;7:1570591. doi: 10.3389/fspor.2025.1570591. eCollection 2025.

ABSTRACT

[This corrects the article DOI: 10.3389/fspor.2024.1512010.].

PMID:40051920 | PMC:PMC11882509 | DOI:10.3389/fspor.2025.1570591