Antioxidants (Basel). 2025 Feb 3;14(2):178. doi: 10.3390/antiox14020178.

ABSTRACT

Chronic kidney disease (CKD) is associated with a high incidence of cardiovascular disease (CVD) due to the accumulation of uremic toxins, altered redox state, and chronic systemic inflammation. This study aimed to analyze the relationship between the redox status of patients with CKD and the phenotype of microvesicles (MVs) subtypes, and cardiovascular events. The oxidative stress level of each participant was determined using an individualized OXY-SCORE. The relationship between pro-oxidant and antioxidant parameters and the expression of membrane markers in endothelial-derived microvesicles (EMVs) and platelet-derived microvesicles (PMVs) was established. Patients with advanced CKD (ACKD) and hemodialysis (HD) had a higher OXY-SCORE than healthy subjects (HS), whereas peritoneal dialysis (PD) patients had similar scores to HS. PD patients showed elevated PMVs and CD41 expression, whereas HD patients had higher EMVs and CD31 expression. Patients with ACKD had higher tissue factor (TF) expression in the PMVs and EMVs. TF expression was correlated with xanthine oxidase (XO) activity and was negatively correlated with antioxidant parameters. Patients with cardiovascular events show elevated TF. In conclusion, microvesicles and oxidative stress may serve as markers of cardiovascular risk in CKD, with TF expression in PMVs and EMVs being potential predictive and prognostic biomarkers of CVD.

PMID:40002365 | DOI:10.3390/antiox14020178

Foods. 2025 Feb 17;14(4):673. doi: 10.3390/foods14040673.

ABSTRACT

Microbial foods include microbial biomass, naturally fermented foods, and heterologously synthesized food ingredients derived from microbial fermentation. Terpenoids, using isoprene as the basic structure, possess various skeletons and functional groups. They exhibit diverse physicochemical properties and physiological activities, such as unique flavor, anti-bacterial, anti-oxidant, anti-cancer, and hypolipemic, making them extensively used in the food industry, such as flavor, fragrance, preservatives, dietary supplements, and medicinal health food. Compared to traditional strategies like direct extraction from natural species and chemical synthesis, microbial cell factories for edible terpenoids have higher titers and yields. They can utilize low-cost raw materials and are easily scaling-up, representing a novel green and sustainable production mode. In this review, we briefly introduce the synthetic pathway of terpenoids and the applications of microbial cell factories producing edible terpenoids. Secondly, we highlight several typical and non-typical microbial chassis in edible terpenoid-producing cell factories. In addition, we reviewed the recent advances of representative terpenoid microbial cell factories with a gram-scale titer in food flavor, food preservation, nutritional enhancers, and medicinal health foods. Finally, we predict the future directions of microbial cell factories for edible terpenoids and their commercialization process.

PMID:40002116 | DOI:10.3390/foods14040673

Biomolecules. 2025 Feb 19;15(2):304. doi: 10.3390/biom15020304.

ABSTRACT

The dynamic nature of mitochondria makes live cell imaging an important tool in mitochondrial research. Although imaging using fluorescent probes is the golden standard in studying mitochondrial morphology, these probes might introduce aspecific features. In this study, live cell fluorescent imaging was applied to investigate a pearl-necklace-shaped mitochondrial phenotype that arises when mitochondrial fission is restricted. In this fibroblast-specific pearl-necklace phenotype, constricted and expanded mitochondrial regions alternate. Imaging studies revealed that the formation time of this pearl-necklace phenotype differs between laser scanning confocal, widefield and spinning disk confocal microscopy. We found that the phenotype formation correlates with the excitation of the fluorescent probe and is the result of phototoxicity. Interestingly, the phenotype only arises in cells stained with red mitochondrial dyes. Serial section electron tomography of the pearl-necklace mitochondria revealed that the mitochondrial membranes remained intact, while the cristae structure was altered. Furthermore, filaments and ER were present at the constricted sites. This study illustrates the importance of considering experimental conditions for live cell imaging to prevent imaging artifacts that can have a major impact on the obtained results.

PMID:40001607 | DOI:10.3390/biom15020304

Biomolecules. 2025 Feb 5;15(2):229. doi: 10.3390/biom15020229.

ABSTRACT

Mitochondria play a crucial role in human biology, affecting cellular processes at the smallest spatial scale as well as those involved in the functionality of the whole system. Imaging is the most important research tool for studying the fundamental role of mitochondria across these diverse spatial scales. A wide array of available imaging techniques have enabled us to visualize mitochondrial structure and behavior, as well as their effect on cells and tissues in a range from micrometers to centimeters. Each of the various imaging techniques that are available offers unique advantages tailored to specific research needs. Selecting an appropriate technique suitable for the scale and application of interest is therefore crucial, but can be challenging due to the large range of possibilities. The aim of this review is two-fold. First, we provide an overview of the available imaging techniques and discuss their strengths and limitations for applications across the sub-mitochondrial, cellular, tissue and organ levels for the imaging of mitochondria. Second, we identify opportunities for novel applications and advancement in the field. We emphasize the importance of integration across scales in mitochondrial imaging studies, particularly to bridge the gap between microscopic and non-invasive techniques. While integrating these diverse scales is challenging, primarily because such multi-scale approaches require expertise that spans different imaging modalities, we argue that integration has the potential to provide groundbreaking insights into mitochondrial biology. By providing a comprehensive overview of imaging techniques, this review paves the way for multi-scale imaging initiatives in mitochondrial research.

PMID:40001532 | DOI:10.3390/biom15020229

Antibiotics (Basel). 2025 Feb 13;14(2):194. doi: 10.3390/antibiotics14020194.

ABSTRACT

Background/Objectives: Cm-p5 and its cyclic monomeric and dimeric analogues are known for their antifungal, antibacterial, antiviral, and antibiofilm activities. Previously, our cyclization method produced a mixture of peptides that were difficult to separate, which was then improved by a selective synthesis of the parallel dimer and its differentiation from the antiparallel by comparison of the retention times in RP-HPLC. Methods: Here, we developed a more reliable identification method for the Cm-p5 dimer identification, which included chymotrypsin proteolytic digestion and sequencing of the different fragments by ESI-MSMS. We also improved our cyclization methods to specifically produce higher amounts of the desired cyclic variant, either cyclic monomer or dimer. Results: We show that liquid phase oxidation with 20% DMSO or iodine oxidation yields only the cyclic analogue. However, the on-resin oxidation with iodine showed greater efficacy and efficiency. Additionally, liquid phase cyclization yields the antiparallel dimer in high EtOH or peptide concentration, indicating a kinetic control. On the other hand, the parallel dimer was preferentially produced in 5% of TFE and low peptide concentration without the formation of the cyclic analogue indicating a thermodynamic control. Conclusions: In conclusion, we report that chymotryptic digestion combined with ESI-MS and MS/MS allows an unambiguous differentiation of Cm-p5 dimers. Here, we develop more selective and efficient methods for the synthesis of cyclic and dimeric analogues of Cm-p5.

PMID:40001437 | DOI:10.3390/antibiotics14020194

Antibiotics (Basel). 2025 Jan 22;14(2):114. doi: 10.3390/antibiotics14020114.

ABSTRACT

Background: Bacteriophages (phages) are viruses that infect and kill bacteria. The antimicrobial resistance crisis has driven renewed interest in phage therapy, including the use of phages to treat chronic orthopaedic infections. Methods: Here, we present the results of the first three orthopaedic patients treated with phage therapy in the United Kingdom. Results: The first patient was treated in May 2023 and received phages active against Staphylococcus aureus. At nine months follow-up, the patient's wound remained healed, the C-reactive protein normal and the patient was walking independently. The second patient received phages active against Klebsiella pneumoniae and S. aureus; the infection remained unresolved. The third patient received phages active against Staphylococcus epidermidis; at six months follow-up, the patient was free of infection. Endotoxin was considered at least partially responsible for mild self-limiting adverse effects in two cases. Conclusions: These promising results hint at the potential for phage therapy to transform the care of chronic orthopaedic infections.

PMID:40001358 | DOI:10.3390/antibiotics14020114

Hum Genomics. 2025 Feb 25;19(1):18. doi: 10.1186/s40246-025-00731-y.

ABSTRACT

BACKGROUND: The protective effects of higher educational attainment (EA) and intelligence on COVID-19 outcomes are not yet understood with regard to their dependency on income. The objective of our study was to examine the overall as well as independent effects of the three psychosocial factors on the susceptibility to and severity of COVID-19. To accomplish this, we utilized genetic correlation, Mendelian randomization (MR), and multivariable MR (MVMR) analyses to evaluate genetic associations between EA, intelligence, household income, and three specific COVID-19 outcomes: SARS-CoV-2 infection, hospitalized COVID-19, and critical COVID-19.

RESULTS: The genetic correlation analysis revealed that COVID-19 outcomes were negatively correlated with the three psychosocial factors (rg: -0.19‒-0.36). The MR analysis indicated that genetic liability to EA, intelligence, and income exerted overall protective effects against SARS-CoV-2 infection (OR: 0.86‒0.92), hospitalized COVID-19 (OR: 0.70‒0.80), and critical COVID-19 (OR: 0.65‒0.85). MVMR analysis revealed that elevated levels of EA conferred independent protective effects against SARS-CoV-2 infection (OR: 0.85), hospitalization due to COVID-19 (OR: 0.79), and critical COVID-19 (OR: 0.63). Furthermore, intelligence exhibited a negative association with the risk of SARS-CoV-2 infection (OR: 0.91), whereas a higher income was linked to an elevated risk of SARS-CoV-2 infection (OR: 1.13).

CONCLUSIONS: Our findings indicated that EA could significantly reduce the risk and severity of COVID-19, regardless of intelligence and income. However, the impact of intelligence or income on COVID-19 severity was not supported by our research.

PMID:40001252 | DOI:10.1186/s40246-025-00731-y

Brain Commun. 2025 Feb 8;7(1):fcaf063. doi: 10.1093/braincomms/fcaf063. eCollection 2025.

ABSTRACT

Metabolic imbalance is associated with amyotrophic lateral sclerosis progression. Impaired glucose oxidation and increased reliance on fatty acid oxidation contribute to reduced metabolic flexibility and faster disease progression in amyotrophic lateral sclerosis. We sought to evaluate the safety and tolerability, and explore the pharmacodynamic response of trimetazidine, a partial fatty acid oxidation inhibitor, on oxidative stress markers and energy expenditure in amyotrophic lateral sclerosis. The study was conducted between June 29, 2021 and May 24, 2023. People living with amyotrophic lateral sclerosis, recruited in Australia and the Netherlands, received open-label oral trimetazidine for 12 weeks after an initial 4-week lead-in period. The primary outcome measures were safety and tolerability, as well as the change from baseline in oxidative stress markers malondialdehyde (MDA) and 8-hydroxy-2'-deoxyguanosine (8-OHdG). Secondary outcome measures were change from baseline in energy expenditure, amyotrophic lateral sclerosis functional rating scale-revised, and slow vital capacity (SVC). Linear mixed effects were used to estimate the mean difference in MDA and 8-OHdG between the on- and off-treatment periods. This trial is registered under ClinicalTrial.gov National Clinical Trial (NCT) number NCT04788745 and European Union Drug Regulating Authorities Clinical Trials (EudraCT) number 2020-005018-17. Twenty-one participants received trimetazidine; 19 (90%) completed the treatment period. Trimetazidine was well tolerated; there were 57 adverse events reported, of which 7 (11%) were deemed potentially drug-related, including hot flushes (2), nausea (2), paraesthesia (2) and fatigue (1). MDA was numerically lower during treatment [-0.29 uM; 95% confidence interval (CI) -0.90 to 0.33, P = 0.36]; 8-OHdG was significantly lower during treatment (-0.12 nM; 95% CI -0.23 to -0.01, P = 0.0245). The decrease in oxidative stress markers was accompanied by a reduction in resting energy expenditure (95 kcal, 95% CI 36.8-154, P = 0.0014). The absence of a placebo group prevented the interpretation of the clinical parameters. Oral trimetazidine was safe and well tolerated among patients with amyotrophic lateral sclerosis. This, combined with the significant reduction in markers of oxidative stress and resting energy expenditure, warrants a larger double-blind placebo-controlled efficacy study.

PMID:40008327 | PMC:PMC11851067 | DOI:10.1093/braincomms/fcaf063

Expert Opin Drug Saf. 2025 Feb 25. doi: 10.1080/14740338.2025.2472918. Online ahead of print.

ABSTRACT

BACKGROUND: To evaluate the incidence and risk of drug-induced interstitial lung disease (DIILD) associated with anti-neoplastic drugs among patients with cancer in Korea. Research design and methods: This nested case-control study included 457,685 patients diagnosed with cancer and treated with anti-neoplastic drugs from a retrospective nationwide population-based cohort between 2017 and 2021. The incidence rate of DIILD and the risks of DIILD by anti-neoplastic drug categories were analyzed.

RESULTS: Among 270,595 patients, 2,634 developed ILD, resulting in an incidence rate of 4.12 per 1,000 person-years (95% confidence interval (CI): 3.97-4.28). DIILD was more prevalent in men, older patients, and those with a history of pulmonary disease or lung cancer. In a multivariable conditional logistic regression analysis, immune checkpoint inhibitors (odds ratio (OR): 2.37; 95%CI: 1.48-3.78), mammalian target of rapamycin inhibitors (OR: 9.79; 95%CI: 5.20-18.45), antibody-drug conjugates (OR: 7.99; 95%CI: 3.24-19.74), cyclin-dependent kinase 4/6 inhibitors (OR: 2.28; 95%CI: 1.26-4.12), and any combination of different drug categories (OR: 1.93; 95%CI: 1.21-3.09) were associated with an increased risk of DIILD.

CONCLUSION: Our findings suggest that the risk of incident DIILD depends on the category of anti-neoplastic drugs. Patients with identified risk factors and treated with these drugs should be monitored closely.

PMID:40007198 | DOI:10.1080/14740338.2025.2472918

Vaccines (Basel). 2025 Feb 7;13(2):161. doi: 10.3390/vaccines13020161.

ABSTRACT

Background: Despite adverse events following immunization (AEFI) being well described in vaccine trials, there is a need to produce more real-world data on events supposedly attributed to vaccination against COVID-19. This study aims to estimate the prevalence of AEFI in the first dose of COVID-19 vaccines in the state of Brazil and to verify whether such events differ among the types of vaccines offered in this country. Methods: A population-based study using linked administrative data on vaccine registry and adverse events following immunization in 2021 and 2022. The study included 10,169,378 individuals aged 18 or over who lived in Bahia and received the first dose of COVID-19 vaccines. We calculated AEFI prevalence and verified differences among vaccines by logistic regression to estimate crude and adjusted by sex and age group prevalence ratio (PR). Results: The prevalence of AEFI was 74.3 per 100,000 doses applied, with a higher rate of nonserious events, mainly following the ChAdOx1-S. More than two-thirds of these adverse effects occurred in women, and almost half were between 30 and 49 years old. The individuals who received ChAdOx1-S had a 125% higher prevalence than those who received CoronaVac. Those who received BNT162b2 and Ad26.COV2.S had a 71% and 58%, respectively, lower prevalence of AEFI than those who received CoronaVac. Conclusions: The use of vaccines against COVID-19 has proven to be positive and effective in combating SARS-CoV-2, significantly reducing morbidity and mortality from the disease. We cannot deny the presence of adverse events in the context of vaccination. However, the vaccines have proven to be safe and reliable. The results of this study offer relevant data that can contribute to the qualification of AEFI pharmacovigilance in Brazil and worldwide.

PMID:40006708 | DOI:10.3390/vaccines13020161

Pharmaceuticals (Basel). 2025 Jan 29;18(2):180. doi: 10.3390/ph18020180.

ABSTRACT

The emergence of antibody-drug conjugates (ADCs) has transformed the treatment landscape of a variety of cancers. ADCs typically consist of three main components: monoclonal antibody, chemical linker, and cytotoxic payload. These integrated therapeutic modalities harness the benefits of each component to provide a therapeutic response that cannot be achieved by conventional chemotherapy. Antibodies play roles in determining tumor specificity through target-mediated uptake, prolonging the circulation half-life of cytotoxic payloads, and providing additional mechanisms of action inherent to the original antibody, thus significantly contributing to the overall performance of ADCs. However, ADCs have unique safety concerns, such as drug-induced adverse events related to the target-mediated uptake of the ADC in normal tissues (so-called "on-target, off-tumor toxicity") and platform toxicity, which are partially derived from limited tumor uptake of antibodies. Identifying suitable target antigens thus impacts the clinical success of ADCs and requires careful consideration, given the multifaceted aspects of this unique treatment modality. This review briefly summarizes the representative roles that antibodies play in determining the efficacy and safety of ADCs. Key considerations for selecting suitable cell surface target antigens for ADC therapy are also highlighted.

PMID:40005994 | DOI:10.3390/ph18020180

Pathogens. 2025 Feb 15;14(2):195. doi: 10.3390/pathogens14020195.

ABSTRACT

Adverse events following immunizations (AEFIs) with recombinant zoster vaccine (RZV) are underexplored in fragile populations. This study aims to assess incidence, duration, and characteristics of AEFIs, focusing on the impact of sex, age, and prior Herpes Zoster (HZ) infection in a frail population, including solid organ transplant recipients. We conducted an observational study on patients receiving RZV, and AEFIs were classified as local or systemic and analyzed for incidence, duration, and patterns across groups. We showed that females had a higher incidence of AEFIs (p = 0.02), both local and systemic symptoms, such as swelling +/- redness at the site of injection and fatigue, after the first and second doses. Younger adults experienced more systemic reactions, while older adults reported more local events (e.g., redness and swelling, p = 0.01). Moreover, patients with previous HZ infection exhibited a higher incidence of AEFIs after the second dose (68% vs. 38%, p = 0.001). In conclusion, sex, age, and clinical history significantly influenced AEFI incidence and manifestations. Therefore, it is important to personalize vaccination strategies in frail populations, by tailored administration and monitoring plans, especially in females and individuals with prior HZ infection, to improve vaccine safety and patient outcomes.

PMID:40005570 | DOI:10.3390/pathogens14020195

J Chem Inf Model. 2025 Mar 10;65(5):2256-2267. doi: 10.1021/acs.jcim.4c01320. Epub 2025 Feb 25.

ABSTRACT

Drug safety is affected by multiple molecular properties and safety assessment is critical for clinical application. Evaluating a drug candidate's therapeutic potential is facilitated by machine learning models trained on extensive compound bioactivity data sets, presenting a promising approach to drug safety assessment. Here, we introduce SynthMol, a deep learning framework that integrates pre-trained 3D structural features, graph attention networks, and molecular fingerprints to achieve high accuracy in molecular property prediction. Evaluation of SynthMol on 22 data sets, including MoleculeNet, MolData and published drug safety data, showed that it could provide higher prediction accuracy than state-of-the-art model in most tasks. SynthMol achieved an ROC-AUC value of 0.944 in the BBBP data set, 2.61% higher than the next best model, and an ROC-AUC of 0.906 on the hERG data set, a 2.38% improvement. Validation of SynthMol in real-world applications with experimentally determined hERG toxicity and CYP inhibition data supported its capacity to distinguish functional changes for drug development. The implementation code and data are available at https://github.com/ThomasSu1/SynthMol.

PMID:40000610 | DOI:10.1021/acs.jcim.4c01320

Lung Cancer. 2025 Mar;201:108449. doi: 10.1016/j.lungcan.2025.108449. Epub 2025 Feb 19.

ABSTRACT

BACKGROUND AND AIMS: NUT carcinoma (NC) is a sporadic, highly aggressive tumor that primarily affects children, adolescents, and young adults and is characterized by the presence of somatic NUTM1 rearrangements. This analysis by the European Cooperative Study Group for Pediatric Rare Tumors (EXPeRT) aims to fill the knowledge gap regarding the clinical characteristics of children with NC.

METHODS: A retrospective case series of NC-patients aged 0-18 years treated between 2011 and 2023 was conducted using the EXPeRT database. Relevant clinical characteristics, including treatment and outcome were recorded.

RESULTS: Twenty-seven patients with a median age of 13 years (range 7-18) were analyzed. Thirteen patients were initially misdiagnosed. Sixteen patients had thoracic and 11 extra-thoracic tumors, including three in the nasal/sinus region and two in the submandibular glands. Despite intense multimodal treatment, median event-free and overall survivals were 1.5 and 6.5 months, respectively.

CONCLUSIONS: Early diagnosis of NC by examination of the NUTM1 rearrangement in undifferentiated or poorly differentiated carcinomas is crucial in order to initiate specific and intensive therapy as quickly as possible. Similar to adult patients, only a minority of pediatric patients achieved prolonged survival. Therefore, the development of novel therapeutic strategies in future joint clinical trials is essential.

PMID:39999637 | DOI:10.1016/j.lungcan.2025.108449

BMC Cancer. 2025 Feb 25;25(1):352. doi: 10.1186/s12885-025-13782-0.

ABSTRACT

BACKGROUND: Patients with a rare cancer in rural, regional, and remote Australia experience heightened challenges in their illness journey, including significant psychosocial impacts. Although peer support has shown benefits for common cancer patients living in urban areas, these programs often do not reach underserved groups for instance those with a rare cancer, or those living in rural, regional or remote areas. This study aimed to explore the characteristics of peer support programs for patients with a rare cancer living in rural, regional or remote areas.

METHODS: Focus groups and interviews were conducted with 39 people with a rare cancer and 10 healthcare providers to explore key points for inclusion in a peer support service for people diagnosed with a rare cancer living in rural, regional or remote areas. Data were transcribed verbatim and analysed thematically, using Nvivo.

RESULTS: Participants described their peer support needs using the key terms who, what, how, where, and when. Participants advocated for a flexible, multicomponent intervention that could meet the varied and fluctuating needs of this group. Participants also noted challenges with the practical delivery of such a service, specifically, the risk of receiving misinformation, adverse emotional reactions, interpersonal challenges and implementation issues.

CONCLUSIONS: This study highlights the role of peer support in addressing unmet needs of patients with a rare cancer, particularly in rural areas, emphasising the importance of tailored, flexible, and multimodal interventions for the delivery of peer support that addresses diverse needs.

PMID:40001049 | DOI:10.1186/s12885-025-13782-0

Sci Rep. 2025 Feb 25;15(1):6758. doi: 10.1038/s41598-025-90780-z.

ABSTRACT

Diabetic foot ulcers (DFUs) are a common and serious complication of diabetes, presenting as open sores or wounds on the sole. They result from impaired blood circulation and neuropathy associated with diabetes, increasing the risk of severe infections and even amputations if untreated. Early detection, effective wound care, and diabetes management are crucial to prevent and treat DFUs. Artificial intelligence (AI), particularly through deep learning, has revolutionized DFU diagnosis and treatment. This work introduces the DFU_XAI framework to enhance the interpretability of deep learning models for DFU labeling and localization, ensuring clinical relevance. The framework evaluates six advanced models-Xception, DenseNet121, ResNet50, InceptionV3, MobileNetV2, and Siamese Neural Network (SNN)-using interpretability techniques like SHAP, LIME, and Grad-CAM. Among these, the SNN model excelled with 98.76% accuracy, 99.3% precision, 97.7% recall, 98.5% F1-score, and 98.6% AUC. Grad-CAM heat maps effectively identified ulcer locations, aiding clinicians with precise and visually interpretable insights. The DFU_XAI framework integrates explainability into AI-driven healthcare, enhancing trust and usability in clinical settings. This approach addresses challenges of transparency in AI for DFU management, offering reliable and efficient solutions to this critical healthcare issue. Traditional DFU methods are labor-intensive and costly, highlighting the transformative potential of AI-driven systems.

PMID:40000748 | DOI:10.1038/s41598-025-90780-z

J Imaging Inform Med. 2025 Feb 25. doi: 10.1007/s10278-025-01436-3. Online ahead of print.

ABSTRACT

Ultrasound is widely used to monitor and diagnose thyroid nodules, but accurately segmenting these nodules in ultrasound images remains a challenge due to the presence of noise and artifacts, which often blur nodule boundaries. While several deep learning algorithms have been developed for this task, their performance is frequently suboptimal. In this study, we introduce the use of EfficientNet-B7 as the backbone for the DeepLabV3+ architecture in thyroid nodule segmentation, marking its first application in this area. We evaluated the proposed method using a dataset from the First Affiliated Hospital of Zhengzhou University, along with two public datasets. The results demonstrate high performance, with a pixel accuracy (PA) of 97.67%, a Dice similarity coefficient of 0.8839, and an Intersection over Union (IoU) of 79.69%. These outcomes outperform most traditional segmentation networks.

PMID:40000546 | DOI:10.1007/s10278-025-01436-3

Insights Imaging. 2025 Feb 25;16(1):48. doi: 10.1186/s13244-025-01923-9.

ABSTRACT

OBJECTIVES: To develop and validate a deep learning model based on nnU-Net combined with radiomics to achieve autosegmentation of gastric cancer (GC) and preoperative prediction via the Lauren classification.

METHODS: Patients with a pathological diagnosis of GC were retrospectively enrolled in three medical centers. The nnU-Net autosegmentation model was developed using manually segmented datasets and evaluated by the Dice similarity coefficient (DSC). The CT images were processed by the nnU-Net model to obtain autosegmentation results and extract radiomic features. The least absolute shrinkage and selection operator (LASSO) method selects optimal features for calculating the Radscore and constructing a radiomic model. Clinical characteristics and the Radscore were integrated to construct a combined model. Model performance was evaluated via the receiver operating characteristic (ROC) curve.

RESULTS: A total of 433 GC patients were divided into the training set, internal validation set, external test set-1, and external test set-2. The nnU-Net model achieved a DSC of 0.79 in the test set. The areas under the curve (AUCs) of the internal validation set, external test set-1, and external test set-2 were 0.84, 0.83, and 0.81, respectively, for the radiomic model; and 0.81, 0.81, and 0.82, respectively, for the combined model. The AUCs of the radiomic and combined models showed no statistically significant difference (p > 0.05). The radiomic model was selected as the optimal model.

CONCLUSIONS: The nnU-Net model can efficiently and accurately achieve automatic segmentation of GCs. The radiomic model can preoperatively predict the Lauren classification of GC with high accuracy.

CRITICAL RELEVANCE STATEMENT: This study highlights the potential of nnU-Net combined with radiomics to noninvasively predict the Lauren classification in gastric cancer patients, enhancing personalized treatment strategies and improving patient management.

KEY POINTS: The Lauren classification influences gastric cancer treatment and prognosis. The nnU-Net model reduces doctors' manual segmentation errors and workload. Radiomics models aid in preoperative Lauren classification prediction for patients with gastric cancer.

PMID:40000513 | DOI:10.1186/s13244-025-01923-9

Acad Radiol. 2025 Feb 24:S1076-6332(25)00104-7. doi: 10.1016/j.acra.2025.02.002. Online ahead of print.

ABSTRACT

RATIONALE AND OBJECTIVES: This study assesses the image quality of temporal bone ultra-high-resolution (UHR) Computed tomography (CT) scans in adults and children using hybrid iterative reconstruction (HIR) and a novel, vendor-specific deep learning-based reconstruction (DLR) algorithm called AiCE Inner Ear.

MATERIAL AND METHODS: In a retrospective, single-center study (February 1-July 30, 2023), UHR-CT scans of 57 temporal bones of 35 patients (5 children, 23 male) with at least one anatomical unremarkable temporal bone were included. There is an adult computed tomography dose index volume (CTDIvol 25.6 mGy) and a pediatric protocol (15.3 mGy). Images were reconstructed using HIR at normal resolution (0.5-mm slice thickness, 512² matrix) and UHR (0.25-mm, 1024² and 2048² matrix) as well as with a vendor-specific DLR advanced intelligent clear-IQ engine inner ear (AiCE Inner Ear) at UHR (0.25-mm, 1024² matrix). Three radiologists evaluated 18 anatomic structures using a 5-point Likert scale. Signal-to-noise (SNR) and contrast-to-noise ratio (CNR) were measured automatically.

RESULTS: In the adult protocol subgroup (n=30; median age: 51 [11-89]; 19 men) and the pediatric protocol subgroup (n=5; median age: 2 [1-3]; 4 men), UHR-CT with DLR significantly improved subjective image quality (p<0.024), reduced noise (p<0.001), and increased CNR and SNR (p<0.001). DLR also enhanced visualization of key structures, including the tendon of the stapedius muscle (p<0.001), tympanic membrane (p<0.009), and basal aspect of the osseous spiral lamina (p<0.018).

CONCLUSION: Vendor-specific DLR-enhanced UHR-CT significantly improves temporal bone image quality and diagnostic performance.

PMID:40000329 | DOI:10.1016/j.acra.2025.02.002

Acad Radiol. 2025 Feb 24:S1076-6332(25)00108-4. doi: 10.1016/j.acra.2025.02.007. Online ahead of print.

ABSTRACT

PURPOSE: This meta-analysis aims to evaluate the diagnostic performance of CT-based artificial intelligence (AI) in diagnosing cervical lymph node metastasis (LNM) of papillary thyroid cancer (PTC).

METHODS: A systematic search was conducted in PubMed, Embase, and Web of Science databases through December 2024, following PRISMA-DTA guidelines. Studies evaluating CT-based AI models for diagnosing cervical LNM in patients with pathologically confirmed PTC were included. The methodological quality was assessed using a modified QUADAS-2 tool. A bivariate random-effects model was used to calculate pooled sensitivity, specificity, and area under the curve (AUC). Heterogeneity was evaluated using I2 statistics, and meta-regression analyses were performed to explore potential sources of heterogeneity.

RESULTS: 17 studies comprising 1778 patients in internal validation sets and 4072 patients in external validation sets were included. In internal validation sets, AI demonstrated a sensitivity of 0.80 (95% CI: 0.71-0.86), specificity of 0.79 (95% CI: 0.73-0.84), and AUC of 0.86 (95% CI: 0.83-0.89). Radiologists suggested comparable performance with sensitivity of 0.77 (95% CI: 0.64-0.87), specificity of 0.79 (95% CI: 0.72-0.85), and AUC of 0.85 (95% CI: 0.81-0.88). Subgroup analyses revealed that deep learning methods outperformed machine learning in sensitivity (0.86 vs 0.72, P<0.05). No significant publication bias was found in internal validation sets for AI diagnosis (P=0.78).

CONCLUSION: CT-based AI showed comparable diagnostic performance to radiologists for detecting cervical LNM in PTC patients, with deep learning models showing superior sensitivity. AI could potentially serve as a valuable diagnostic support tool, though further prospective validation is warranted. Limitations include high heterogeneity among studies and insufficient external validation in diverse populations.

PMID:40000328 | DOI:10.1016/j.acra.2025.02.007