Explore the vast range of titles available.

Radiomics aims to develop novel biomarkers and provide relevant deeper subvisual information about pathology, immunophenotype, and tumor microenvironment. It uses automated or semiautomated quantitative analysis of high‐dimensional images to improve characterization, diagnosis, and prognosis. Recent years have seen a rapid increase in radiomics applications in nuclear medicine, leading to some promising research results in digestive system oncology, which have been driven by big data analysis and the development of artificial intelligence. Although radiomics advances one step further toward the non‐invasive precision medical analysis, it is still a step away from clinical application and faces many challenges. This review article summarizes the available literature on digestive system tumors regarding radiomics in nuclear medicine. First, we describe the workflow and steps involved in radiomics analysis. Subsequently, we discuss the progress in clinical application regarding the utilization of radiomics for distinguishing between various diseases and evaluating their prognosis, and demonstrate how radiomics advances this field. Finally, we offer our viewpoint on how the field can progress by addressing the challenges facing clinical implementation. Recent years have seen a rapid increase in radiomics applications in nuclear medicine, leading to some promising research results in digestive system oncology, which have been driven by big data analysis and the development of artificial intelligence. This review article summarizes the available literature on digestive system tumors regarding radiomics in nuclear medicine. The authors discuss that the progress in the clinical application regarding the utilization of radiomics for distinguishing between various diseases and evaluating their prognosis, and demonstrate how radiomics advances this field.

This study evaluates the independent correlation between inner cell mass (ICM) and trophectoderm (TE) grading with preimplantation genetic testing for aneuploidy (PGT-A) outcomes in embryos biopsied on Day 5 or Day 6 to determine whether morphology-based selection can serve as an alternative to PGT-A. A retrospective analysis of 1,292 embryos from patients undergoing IVF at a single academic center was conducted between October 2019 and March 2023. Blastocysts were graded using Society for Assisted Reproductive Technology (SART) morphology criteria, and biopsy results were categorized as euploid, aneuploid, mosaic, inconclusive, or low DNA. Maternal age (24–45 years) was analyzed both by SART age groups (< 35, 35–37, 38–40, 41–42, > 42) and in a multivariable logistic regression (euploid vs aneuploid) adjusting for age, biopsy day (Day 5 vs Day 6), inner cell mass (ICM), and trophectoderm (TE). Embryos with mosaic, inconclusive, or low-DNA results were excluded from inferential analyses. Statistical analysis was performed using the chi-square test (p < 0.05). Of the 1,292 embryos, 632 (48.9%) were euploid, 508 (39.3%) aneuploid, and 152 (11.8%) mosaic or inconclusive. Good ICM embryos had significantly higher euploidy rates than aneuploidy on both biopsy days (p < 0.001), while poor ICM embryos had higher aneuploidy rates (p < 0.001). Similarly, good TE embryos were more likely to be euploid than poor TE embryos (p < 0.001). Day 5 embryos had significantly higher euploidy rates than Day 6 embryos across all grading categories (p < 0.05). However, no statistical difference was found between Day 6 good ICM embryos and Day 5 fair ICM embryos (p = 0.1335), while Day 6 good TE embryos had higher euploidy than Day 5 fair TE embryos (p < 0.01). Euploidy declined stepwise across SART age groups. In the adjusted model, each additional year of maternal age after 35 lowered the odds of euploidy by ~ 6%. Both ICM and TE grades predict embryo euploidy, supporting the use of morphological assessment to optimize embryo selection and potentially reduce reliance on PGT-A.

This study developed a non‐invasive model using PIVKA‐II and MRI features to predict microvascular invasion in hepatocellular carcinoma, providing a reliable tool for early risk assessment and personalized treatment planning. The study included 98 patients with pathologically confirmed HCC (Child‐Pugh A, BCLC stage A), comprising 43 MVI‐positive and 55 MVI‐negative cases. Baseline clinical characteristics and MRI features were collected. Univariate analysis identified candidate variables associated with MVI, which were subsequently included in multivariate logistic regression analysis to identify independent influencing factors. A nomogram prediction model was developed based on the independent factors. The model's diagnostic performance, calibration, and clinical applicability were evaluated. The diagnostic value of PIVKA‐II alone for MVI was analyzed. The MVI‐positive group showed significantly higher alpha‐fetoprotein ≥ 400 ng/mL, PIVKA‐II levels, and aspartate aminotransferase levels. ROC curve analysis showed that PIVKA‐II alone had an AUC of 0.682 for diagnosing MVI, with a maximum Youden's index of 33.57, a cut‐off value of 741.5 mAU/mL, 37.21% sensitivity, 96.36% specificity. Tumor diameter, peritumoral abnormal enhancement, intratumoral arteries, and mean tumor ADC value showed significant differences between the two groups. Elevated PIVKA‐II, intratumoral arteries, and mean tumor ADC value were independent influencing factors for MVI in HCC. A nomogram incorporating these factors achieved an AUC of 0.81, outperforming PIVKA‐II alone. The model demonstrated good calibration and clinical utility. The non‐invasive predictive nomogram constructed by combining PIVKA‐II with MRI features demonstrates good predictive value and clinical applicability for assessing MVI in HCC patients. [Correction added on 25 March 2026, after first online publication: In the sentence “A nomogram incorporating these factors achieved...” the value “0.85” is changed to “0.81” in this version.]

KCI Citation Count: 0

The relationship between dysbiosis of the gastrointestinal microbiota and gastric cancer (GC) has been extensively studied. However, microbiota alterations in GC patients vary widely across studies, and reproducible diagnostic biomarkers for early GC are still lacking in multiple populations. Thus, this study aimed to characterize the gastrointestinal microbial communities involved in gastric carcinogenesis through a meta-analysis of multiple published and open datasets. We analyzed 16S rRNA sequencing data from 1,642 gastric biopsy samples and 394 stool samples across 11 independent studies. VSEARCH, QIIME and R packages such as vegan, phyloseq, cooccur, and random forest were used for data processing and analysis. PICRUSt software was employed to predict functions. The α-diversity results indicated significant differences in the intratumoral microbiota of cancer patients compared to non-cancer patients, while no significant differences were observed in the fecal microbiota. Network analysis showed that the positive correlation with GC-enriched bacteria increased, and the positive correlation with GC-depleted bacteria decreased compared to healthy individuals. Functional analyses indicated that pathways related to carbohydrate metabolism were significantly enriched in GC, while biosynthesis of unsaturated fatty acids was diminished. Additionally, we investigated non- commensals, which are crucial in both -negative and -positive GC. Random forest models, constructed using specific taxa associated with GC identified from the LEfSe analysis, revealed that the combination of Lactobacillus and Streptococcus included alone could effectively discriminate between GC patients and healthy individuals in fecal samples (area under the curve (AUC) = 0.7949). This finding was also validated in an independent cohort (AUC = 0.7712). This study examined the intratumoral and fecal microbiota of GC patients from a dual microecological perspective and identified and as intratumoral and intestinal-specific co-differential bacteria. Furthermore, it confirmed the validity of the combination of and as GC-specific microbial markers across multiple populations, which may aid in the early non-invasive diagnosis of GC.

Background Accurate measurement of hemoglobin concentration is essential for various medical scenarios, including preoperative evaluations and determining blood loss. Traditional invasive methods are inconvenient and not suitable for rapid, point-of-care testing. Moreover, current models, due to their complex parameters, are not well-suited for mobile medical settings, which limits the ability to conduct frequent and rapid testing. This study aims to introduce a novel, compact, and efficient system that leverages deep learning and smartphone technology to accurately estimate hemoglobin levels, thereby facilitating rapid and accessible medical assessments. Methods The study employed a smartphone application to capture images of the eye, which were subsequently analyzed by a deep neural network trained on data from invasive blood test data. Specifically, the EGE-Unet model was utilized for eyelid segmentation, while the DHA(C3AE) model was employed for hemoglobin level prediction. The performance of the EGE-Unet was evaluated using statistical metrics including mean intersection over union (MIOU), F1 Score, accuracy, specificity, and sensitivity. The DHA(C3AE) model’s performance was assessed using mean absolute error (MAE), mean-square error (MSE), root mean square error (RMSE), and R^2. Results The EGE-Unet model demonstrated robust performance in eyelid segmentation, achieving an MIOU of 0.78, an F1 Score of 0.87, an accuracy of 0.97, a specificity of 0.98, and a sensitivity of 0.86. The DHA(C3AE) model for hemoglobin level prediction yielded promising outcomes with an MAE of 1.34, an MSE of 2.85, an RMSE of 1.69, and an R^2 of 0.34. The overall size of the model is modest at 1.08 M, with a computational complexity of 0.12 FLOPs (G). Conclusions This system presents a groundbreaking approach that eliminates the need for supplementary devices, providing a cost-effective, swift, and accurate method for healthcare professionals to enhance treatment planning and improve patient care in perioperative environments. The proposed system has the potential to enable frequent and rapid testing of hemoglobin levels, which can be particularly beneficial in mobile medical settings. Trial Registration The clinical trial was registered on the Chinese Clinical Trial Registry (No. ChiCTR2100044138) on 20/02/2021.

The integration of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) with radiomics has emerged as a transformative approach for non-invasive prediction of breast cancer molecular subtypes. This review systematically evaluates methodological innovations, clinical validation milestones, and translational applications: (1) Methodological Advancements: Standardized DCE-MRI protocols combined with multidimensional radiomic features (morphological, textural, and wavelet-transformed parameters) significantly improved discriminative performance for ER, HER2, and triple-negative subtypes. (2) Deep Learning Integration: Multitask predictive models achieved early treatment response assessment and recurrence risk stratification through spatiotemporal heterogeneity analysis. (3) Clinical Validation: Prospective multicenter trials demonstrated that radiomic models showed strong concordance with 21-gene assays and could potentially replace 38% of repeat biopsies. Despite these advancements, challenges persist in data heterogeneity and mechanistic interpretation of radiomic biomarkers. Emerging strategies integrating radiogenomic analyses and organoid validation platforms are establishing new paradigms for precision imaging-guided therapy.

Embryo implantation failure is considered a leading cause of infertility and a significant bottleneck for fertilization (IVF) treatment. Confirmed factors that lead to implantation failure involve unhealthy embryos, unreceptive endometrium, and asynchronous development and communication between the two. The quality of embryos is further dependent on sperm parameters, oocyte quality, and early embryo development after fertilization. The extensive involvement of such different factors contributes to the variability of implantation potential across different menstrual cycles. An ideal approach to predict the implantation outcome should not compromise embryo implantation. The use of clinical material, including follicular fluid, cumulus cells, sperm, seminal exosomes, spent blastocyst culture medium, blood, and uterine fluid, that can be collected relatively non-invasively without compromising embryo implantation in a transfer cycle opens new perspectives for the diagnosis of embryo implantation potential. Compositional comparison of these samples between fertile women and women or couples with implantation failure has identified both quantitative and qualitative differences in the expression of microRNAs (miRs) that hold diagnostic potential for implantation failure. Here, we review current findings of secreted miRs that have been identified to potentially be useful in predicting implantation outcome using material that can be collected relatively non-invasively. Developing non-invasive biomarkers of implantation potential would have a major impact on implantation failure and infertility.

Background: The transcription factor SOX9 plays a critical role in various diseases, including hepatocellular carcinoma (HCC), and has been implicated in resistance to sorafenib treatment. Accurate assessment of SOX9 expression is important for guiding personalized therapy in HCC patients; however, a reliable non-invasive method for evaluating SOX9 status remains lacking. This study aims to develop a deep learning (DL) model capable of preoperatively and non-invasively predicting SOX9 expression from CT images in HCC patients. Methods: We retrospectively analyzed a dataset comprising 4011 CT images from 101 HCC patients who underwent surgical resection followed by sorafenib therapy at West China Hospital, Sichuan University. A deep reinforcement learning (DRL) approach was proposed to enhance prediction accuracy by identifying and focusing on image regions highly correlated with SOX9 expression, thereby reducing the impact of background noise. Results: Our DRL-based model achieved an area under the curve (AUC) of 91.00% (95% confidence interval: 88.64–93.15%), outperforming conventional DL methods by over 10%. Furthermore, survival analysis revealed that patients with SOX9-positive tumors had significantly shorter recurrence-free survival (RFS) and overall survival (OS) compared to SOX9-negative patients, highlighting the prognostic value of SOX9 status. Conclusions: This study demonstrates that a DRL-enhanced DL model can accurately and non-invasively predict SOX9 expression in HCC patients using preoperative CT images. These findings support the clinical utility of imaging-based SOX9 assessment in informing treatment strategies and prognostic evaluation for patients with advanced HCC.