Training Set Research Articles

Prediction and stratification for the surgical adverse events after minimally invasive esophagectomy: A two-center retrospective study

BACKGROUND Minimally invasive esophagectomy (MIE) is a widely accepted treatment for esophageal cancer, yet it is associated with a significant risk of surgical adverse events (SAEs), which can compromise patient recovery and long-term survival. Accurate preoperative identification of high-risk patients is critical for improving outcomes. AIM To establish and validate a risk prediction and stratification model for the risk of SAEs in patients with MIE. METHODS This retrospective study included 747 patients who underwent MIE at two centers from January 2019 to February 2024. Patients were separated into a train set (n = 549) and a validation set (n = 198). After screening by least absolute shrinkage and selection operator regression, multivariate logistic regression analyzed clinical and intraoperative variables to identify independent risk factors for SAEs. A risk stratification model was constructed and validated to predict the probability of SAEs. RESULTS SAEs occurred in 10.2% of patients in train set and 13.6% in the validation set. Patients with SAE had significantly higher complication rate and a longer hospital stay after surgery. The key independent risk factors identified included chronic obstructive pulmonary disease, a history of alcohol consumption, low forced expiratory volume in the first second, and low albumin levels. The stratification model has excellent prediction accuracy, with an area under the curve of 0.889 for the training set and an area under the curve of 0.793 for the validation set. CONCLUSION The developed risk stratification model effectively predicts the risk of SAEs in patients undergoing MIE, facilitating targeted preoperative interventions and improving perioperative management.

Computationally efficient machine-learned model for GST phase change materials via direct and indirect learning.

Phase change materials such as Ge2Sb2Te5 (GST) are ideal candidates for next-generation, non-volatile, solid-state memory due to the ability to retain binary data in the amorphous and crystal phases and rapidly transition between these phases to write/erase information. Thus, there is wide interest in using molecular modeling to study GST. Recently, a Gaussian Approximation Potential (GAP) was trained for GST to reproduce Density Functional Theory (DFT) energies and forces at a fraction of the computational cost [Zhou et al., Nat. Electron. 6, 746 (2023)]; however, simulations of large length and time scales are still challenging using this GAP model. Here, we present a machine-learned (ML) potential for GST implemented using the Atomic Cluster Expansion (ACE) framework. This ACE potential shows comparable accuracy to the GAP potential but performs orders of magnitude faster. We train the ACE potentials both directly from DFT and also using a recently introduced indirect learning approach where the potential is trained instead from an intermediate ML potential, in this case, GAP. Indirect learning allows us to consider a significantly larger training set than could be generated using DFT alone. We compare the directly and indirectly learned potentials and find that both reproduce the structure and thermodynamics predicted by the GAP and also match experimental measures of GST structure. The speed of the ACE model, particularly when using graphics processing unit acceleration, allows us to examine repeated transitions between crystal and amorphous phases in device-scale systems with only modest computational resources.

PE-CycleGAN network based CBCT-sCT generation for nasopharyngeal carsinoma adaptive radiotherapy.

To explore the synthesis of high-quality CT (sCT) from cone-beam CT (CBCT) using PE-CycleGAN for adaptive radiotherapy (ART) for nasopharyngeal carcinoma. A perception-enhanced CycleGAN model "PE-CycleGAN" was proposed, introducing dual-contrast discriminator loss, multi-perceptual generator loss, and improved U-Net structure. CBCT and CT data from 80 nasopharyngeal carcinoma patients were used as the training set, with 7 cases as the test set. By quantifying the mean absolute error (MAE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), as well as the dose gamma pass rate and the relative dose deviations of the target area and organs at risk (OAR) between sCT and reference CT, the image quality and dose calculation accuracy of sCT were evaluated. The MAE of sCT generated by PE-CycleGAN compared to the reference CT was (56.89±13.84) HU, approximately 30% lower than CBCT's (81.06±15.86) HU (P<0.001). PE-CycleGAN's PSNR and SSIM were 26.69±2.41dB and 0.92±0.02 respectively, significantly higher than CBCT's 21.54±2.37dB and 0.86±0.05 (P<0.001), indicating substantial improvements in image quality and structural similarity. In gamma analysis, under the 2 mm/2% criterion, PE-CycleGAN's sCT achieved a pass rate of (90.13±3.75)%, significantly higher than CBCT's (81.65±3.92)% (P<0.001) and CycleGAN's (87.69±3.50)% (P<0.05). Under the 3 mm/3% criterion, PE-CycleGAN's sCT pass rate of (90.13±3.75)% was also significantly superior to CBCT's (86.92±3.51)% (P<0.001) and CycleGAN's (94.58±2.23)% (P<0.01). The mean relative dose deviation of the target area and OAR between sCT and planned CT was within ±3% for all regions, except for the Lens Dmax (Gy), which had a deviation of 3.38% (P=0.09). The mean relative dose deviations for PTVnx HI, PTVnd HI, PTVnd CI, PTV1 HI, PRV_SC, PRV_BS, Parotid, Larynx, Oral, Mandible, and PRV_ON were all less than ±1% (P>0.05). PE-CycleGAN demonstrates the ability to rapidly synthesize high-quality sCT from CBCT, offering a promising approach for CBCT-guided adaptive radiotherapy in nasopharyngeal carcinoma.

Research on the Prediction Model of Coal Spontaneous Combustion Hazard Level Based on IGWO-GRNN

ABSTRACT To accurately predict the risk level of spontaneous coal combustion, we introduce an improved grey wolf optimization strategy (IGWO) to optimize the generalized regression neural network (GRNN), thereby establishing the IGWO-GRNN coal spontaneous combustion risk level prediction model. The superiority of the IGWO algorithm over the grey wolf optimization algorithm (GWO), whale optimization algorithm (WOA) and particle swarm optimization algorithm (PSO) was verified. Fifty sets of sample data of spontaneous coal combustion in the mining area were selected as the research object, and the data were randomly divided into the training set and the test set according to the ratio of 8:2. The prediction results of the IGWO-GRNN prediction model were obtained and compared with those of the GWO-GRNN, GRNN, BP (Backpropagation), and GRU (Gated Recurrent Unit) models, and the absolute value of the relative error was introduced as an evaluation index of the model prediction accuracy. The results demonstrate that IGWO outperforms GWO, WOA, and PSO in terms of optimization efficiency, with the IGWO-GRNN prediction model yielding the most accurate results. Specifically, the maximum absolute value of the relative error does not exceed 4.27%, and the average relative error is only 3.29%, indicating a substantial improvement over the other models. Therefore, this study achieves high-precision prediction under small-sample data, which provides an important reference value for the small-sample prediction problem in the mining field.

A data-driven latent variable approach to validating the research domain criteria framework

Despite the widespread use of the Research Domain Criteria (RDoC) framework in psychiatry and neuroscience, recent studies suggest that the RDoC is insufficiently specific or excessively broad relative to the underlying brain circuitry it seeks to elucidate. To address these concerns, we employ a latent variable approach using bifactor analysis. We examine 84 whole-brain task-based fMRI (tfMRI) activation maps from 19 studies with 6192 participants. A curated subset of 37 maps with a balanced representation of RDoC domains constitute the training set, and the remaining held-out maps form the internal validation set. External validation is conducted using 36 peak coordinate activation maps from Neurosynth, using terms of RDoC constructs as seeds for topic meta-analysis. Here, we show that a bifactor model incorporating a task-general domain and splitting the cognitive systems domain better fits the examined corpus of tfMRI data than the current RDoC framework. We also identify the domain of arousal and regulatory systems as underrepresented. Our data-driven validation supports revising the RDoC framework to reflect underlying brain circuitry more accurately.

Whole lung radiomic features are associated with overall survival in patients with locally advanced non-small cell lung cancer treated with definitive radiotherapy

BackgroundSeveral studies have suggested that lung tissue heterogeneity is associated with overall survival (OS) in lung cancer. However, the quantitative relationship between the two remains unknown. The purpose of this study is to investigate the prognostic value of whole lung-based and tumor-based radiomics for OS in LA-NSCLC treated with definitive radiotherapy.MethodsA total of 661 patients with LA-NSCLC treated with definitive radiotherapy in combination with chemotherapy were enrolled in this study, with 292 patients in the training set, 57 patients from the same hospital from January to December 2017 as an independent test set (test-set-1), 83 patients from a multi-institutional prospective clinical trial data set (RTOG0617) as test-set-2, and 229 patients from a Dutch radiotherapy center as test-set-3. Tumor-based radiomic features and whole lung-based radiomic features were extracted from primary tumor and whole lungs (excluding the primary tumor) delineations in planning CT images. Feature selection of radiomic features was done by the least absolute shrinkage (LASSO) method embedded with a Cox proportional hazards (CPH) model with 5-fold cross-internal validation, with 1000 bootstrap samples. Radiomics prognostic scores (RS) were calculated by CPH regression based on selected features. Three models based on a tumor RS, and a lung RS separately and their combinations were constructed. The Harrell concordance index (C-index) and calibration curves were used to evaluate the discrimination and calibration performance. Patients were stratified into high and low risk groups based on median RS, and a log-rank test was performed.ResultsThe discrimination ability of lung- and tumor-based radiomics model was similar in terms of C-index, 0.69 vs. 0.68 in training set, 0.68 vs. 0.66 in test-set-1, 0.61 vs. 0.62 in test-set-2, 0.65 vs. 0.64 in test-set-3. The combination of tumor- and lung-based radiomics model performed best, with C-index of 0.71 in training set, 0.70 in test-set-1, 0.69 in test-set-2, and 0.68 in test-set-3. The calibration curve showed good agreement between predicted values and actual values. Patients were well stratified in training set, test-set-1 and test-set-3. In test-set-2, it was only whole lung-based RS that could stratify patients well and tumor-based RS performed bad.ConclusionLung- and tumor-based radiomic features have the power to predict OS in LA-NSCLC. The combination of tumor- and lung-based radiomic features can achieve optimal performance.

Discovery of Methylated DNA Biomarkers for Potential Non-Endoscopic Detection of Barrett's Esophagus and Esophageal Adenocarcinoma.

We sought to develop a minimally-invasive, robust, accessible nonendoscopic strategy to diagnose Barrett's esophagus (BE), esophageal adenocarcinoma (EAC), and its immediate precursor lesion, high-grade dysplasia (HGD) based on methylated DNA biomarkers applied to a retrievable sponge-capsule device in a cohort representative of the BE population (i.e., mostly short-segment, non-dysplastic BE, NDBE). We identified 12 candidate methylation markers to distinguish normal vs. abnormal esophagus. These 12 markers were first assayed in 21-paired matched NDBE-normal esophageal tissues, then assessed in a case-control study of 234 esophageal samples collected using a sponge-capsule device. A classification algorithm was developed using Least Absolute Shrinkage and Selection Operator (LASSO) in a 199-patient training set and tested in an independent 35-patient test set. Twelve markers (A1BG, C9orf50, cg00720137, FLI1, GRAMD1B, HOXB13, IRF4, KCNQ3, NTNG1, SPX, TBC1D30, and USP44) were significantly hypermethylated (i.e., all P<0.05) in BE vs. matched normal esophageal biopsies. A discriminatory three-gene LASSO panel (USP44, TBCD1D30, NELL1), adjusted for age and sex, accurately distinguished HGD or EAC from normal control patients in both training (AUC=0.911, 95%CI=0.863-0.959) and test (AUC=0.969, 95%CI=0.911-1.00) sets. In normal vs. NDBE/LGD/HGD/EAC patients, this algorithm exhibited AUCs of 0.862 (95%CI=0.812-0.912) and 0.864 (95%CI=0.745-0.982) in training and test sets, respectively. In normal vs. NDBE patients, the algorithm yielded AUCs of 0.819 (95%CI=0.748-0.889) and 0.776 (95%CI=0.583-0.968) in training and test sets, respectively. This discriminatory biomarker panel algorithm exemplifies a practical nonendoscopic strategy to diagnose BE, HGD, and EAC using a minimally-invasive sponge-capsule device coupled with DNA methylation markers.

Nomogram to predict cumulative live birth rate following in vitro fertilization/intracytoplasmic sperm injection cycles in patients with endometriosis

BackgroundThe success of in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) in endometriosis patients varies widely, and predicting the likelihood of achieving a live birth remains a clinical challenge. This study aims to develop a predictive nomogram for assessing the cumulative live birth (CLB) rate following IVF/ICSI cycles among patients with endometriosis.MethodA retrospective cohort study was conducted to analyze the clinical data of 1457 patients with endometriosis after IVF/ICSI treatment from January 2017 to August 2022. The patients were divided into a training set (70%) and a validation set (30%) using a random number table. Univariate analysis and multifactorial logistic regression analysis were employed to identify relevant predictive factors affecting CLB rate. A predictive model was then established based on the identified factors.ResultsUnivariate analysis and multifactorial logistic regression analysis revealed that patients with concurrent adenomyosis had a decreased CLB rate (OR = 0.51, 95% CI: 0.31—0.82). As the duration of infertility increased, the CLB rate decreased (OR = 0.93, 95% CI: 0.88—0.99). Higher numbers of fertilized oocytes and high-quality embryos were associated with an increased likelihood of CLB. A nomogram predictive model for CLB rate, based on age, concurrent adenomyosis, duration of infertility, number of fertilization, and number of high-quality embryos, was developed. The area under the curve (AUC) for the training set and validation set was 0.823 (95% CI: 0.798–0.849) and 0.773 (95% CI: 0.729–0.818), respectively. The stratified analysis demonstrated the applicability of the model in the validation cohort.ConclusionThis predictive nomogram for CLB rate in patients with endometriosis provides valuable and precise information for personalized decision-making, which could be a visual and easily applied tool for IVF/ICSI counselling.

Development of a clinical prediction model for benign and malignant pulmonary nodules with a CTR ≥ 50% utilizing artificial intelligence-driven radiomics analysis

ObjectiveIn clinical practice, diagnosing the benignity and malignancy of solid-component-predominant pulmonary nodules is challenging, especially when 3D consolidation-to-tumor ratio (CTR) ≥ 50%, as malignant ones are more invasive. This study aims to develop and validate an AI-driven radiomics prediction model for such nodules to enhance diagnostic accuracy.MethodsData of 2,591 pulmonary nodules from five medical centers (Zhengzhou People’s Hospital, etc.) were collected. Applying exclusion criteria, 370 nodules (78 benign, 292 malignant) with 3D CTR ≥ 50% were selected and randomly split 7:3 into training and validation cohorts. Using R programming, Lasso regression with 10-fold cross-validation filtered features, followed by univariate and multivariate logistic regression to construct the model. Its efficacy was evaluated by ROC, DCA curves and calibration plots.ResultsLasso regression picked 18 non-zero coefficients from 108 features. Three significant factors—patient age, solid component volume and mean CT value—were identified. The logistic regression equation was formulated. In the training set, the ROC AUC was 0.721 (95%CI: 0.642–0.801); in the validation set, AUC was 0.757 (95%CI: 0.632–0.881), showing the model’s stability and predictive ability.ConclusionThe model has moderate accuracy in differentiating benign from malignant 3D CTR ≥ 50% nodules, holding clinical potential. Future efforts could explore more to improve its precision and value.Clinical trial numberNot applicable.

Deep learning-aided diagnosis of acute abdominal aortic dissection by ultrasound images.

Acute abdominal aortic dissection (AD) is a serious disease. Early detection based on ultrasound (US) can improve the prognosis of AD, especially in emergency settings. We explored the ability of deep learning (DL) to diagnose abdominal AD in US images, which may help the diagnosis of AD by novice radiologists or non-professionals. There were 374 US images from patients treated before June 30, 2022. The images were classified as AD-positive and AD-negative images. Among them, 90% of images were used as the training set, and 10% of images were used as the test set. A Densenet-169 model and a VGG-16 model were used in this study and compared with two human readers. DL models demonstrated high sensitivity and AUC for diagnosing abdominal AD in US images, and DL models showed generally better performance than human readers. Our findings demonstrated the efficacy of DL-aided diagnosis of abdominal AD in US images, which can be helpful in emergency settings.

PLASMA: Partial LeAst Squares for Multiomics Analysis

Background/Objectives: Recent growth in the number and applications of high-throughput “omics” technologies has created a need for better methods to integrate multiomics data. Much progress has been made in developing unsupervised methods, but supervised methods have lagged behind. Methods: Here we present the first algorithm, PLASMA, that can learn to predict time-to-event outcomes from multiomics data sets, even when some samples have only been assayed on a subset of the omics data sets. PLASMA uses two layers of existing partial least squares algorithms to first select components that covary with the outcome and then construct a joint Cox proportional hazards model. Results: We apply PLASMA to the stomach adenocarcinoma (STAD) data from The Cancer Genome Atlas. We validate the model both by splitting the STAD data into training and test sets and by applying them to the subset of esophageal cancer (ESCA) containing adenocarcinomas. We use the other half of the ESCA data, which contains squamous cell carcinomas dissimilar to STAD, as a negative comparison. Our model successfully separates both the STAD test set (p = 2.73 × 10−8) and the independent ESCA adenocarcinoma data (p = 0.025) into high-risk and low-risk patients. It does not separate the negative comparison data set (ESCA squamous cell carcinomas, p = 0.57). The performance of the unified multiomics model is superior to that of individually trained models and is also superior to an unsupervised method (Multi-Omics Factor Analysis; MOFA), which finds latent factors to be used as putative predictors in a post hoc survival analysis. Conclusions: Many of the factors that contribute strongly to the PLASMA model can be justified from the biological literature.

Clinical impact of an explainable machine learning with amino acid PET imaging: application to the diagnosis of aggressive glioma.

Radiomics-based machine learning (ML) models of amino acid positron emission tomography (PET) images have shown efficiency in glioma prediction tasks. However, their clinical impact on physician interpretation remains limited. This study investigated whether an explainable radiomics model modifies nuclear physicians' assessment of glioma aggressiveness at diagnosis. Patients underwent dynamic 6-[18F]fluoro-L-DOPA PET acquisition. With a 75%/25% split for training (n = 63) and test sets (n = 22), an ensemble ML model was trained using radiomics features extracted from static/dynamic parametric PET images to classify lesion aggressiveness. Three explainable ML methods-Local Interpretable Model-agnostic Explanations (LIME), Anchor, and SHapley Additive exPlanations (SHAP)-generated patient-specific explanations. Eighteen physicians from eight institutions evaluated the test samples. During the first phase, physicians analyzed the 22 cases exclusively through magnetic resonance and static/dynamic PET images, acquired within a maximum interval of 30 days. In the second phase, the same physicians reevaluated the same cases (n = 22), using all available data, including the radiomics model predictions and explanations. Eighty-five patients (54[39-62] years old, 41 women) were selected. In the second phase, physicians demonstrated a significant improvement in diagnostic accuracy compared to the first phase (0.775 [0.750-0.802] vs. 0.717 [0.694-0.737], p = 0.007). The explainable radiomics model augmented physician agreement, with a 22.72% increase in Fleiss's kappa, and significantly enhanced physician confidence (p < 0.001). Among all physicians, Anchor and SHAP showed efficacy in 75% and 72% of cases, respectively, outperforming LIME (p ≤ 0.001). Our results highlight the potential of an explainable radiomics model using amino acid PET scans as a diagnostic support to assist physicians in identifying glioma aggressiveness.

Deep Learning-Derived Quantitative Scores for Chronic Rhinosinusitis Assessment: Correlation With Quality of Life Outcomes.

Computed tomography (CT) plays a crucial role in assessing chronic rhinosinusitis, but lacks objective quantifiable indicators. This study aimed to use deep learning for automated sinus segmentation to generate distinct quantitative scores and explore their correlations with disease-specific quality of life. From July 2021 to August 2022, 445 CT data were collected from 2 medical centers. A deep learning model based on nnU-Net was trained for automatic sinus segmentation and internally validated using 300 cases. The remaining 145 cases were split into an external testing set (74 cases) and an independent testing set (71 cases). Two quantitative scores, the quantitative Lund-MacKay score and the quantitative opacification score (QOS), were derived from the segmentation results. The quantitative scores' efficacy was assessed by comparing them with the Lund-MacKay score (LMS), the 22-item Sinonasal Outcome Test score (SNOT-22), and other clinical variables through correlation analyses. Furthermore, the relationship between quantitative scores and postoperative quality of life improvement was explored using single-factor logistic regression. The segmentation model achieved average Dice similarity coefficients of 0.993, 0.978, 0.958, and 0.871 for the training, validation, external testing, and independent testing sets, respectively. Both quantitative scores significantly correlated with the LMS (rho = 0.87 and rho = 0.70, P < .001). Neither score correlated with the total SNOT-22 score, although the modified QOS showed significant correlations with the nasal and sleep subdomains (rho = 0.26 and rho = 0.27, P < .05). No significant association was found between quantitative score and postoperative improvement in quality of life. Deep learning enables the automated segmentation of sinuses on CT scans, producing quantitative scores of sinus opacification. These automatic quantitative scores may serve as tools for chronic rhinosinusitis assessment.

Dispersionless Nonhybrid Density Functional.

A dispersion-corrected density functional theory (DFT+D) method has been developed. It includes a nonhybrid dispersionless generalized gradient approximation (GGA) functional paired with a literature-parametrized dispersion function. The functional's 9 adjustable parameters were optimized using a training set of 589 benchmark interaction energies. The resulting method performs better than other GGA-based DFT+D methods, giving a mean unsigned error of 0.33 kcal/mol. It even performs better than some more expensive meta-GGA or hybrid dispersion-corrected functionals. An important advantage of using the new functional is that its dispersion energy given by the D component is very close to the true dispersion energy at all intermolecular separations, whereas in other similarly accurate DFT+D approaches, such a dispersion contribution in the van der Waals minimum region is only a small fraction of the true value.

A nomogram for predicting central venous catheter-related infections in maintenance hemodialysis: A 4-year single-center study.

To develop and validate a nomogram model for predicting central venous catheter-related infections (CRI) in patients with maintenance hemodialysis (MHD). MHD patients with central venous catheters (CVCs) visiting the outpatient hemodialysis (HD) center of Xuzhou Medical University Affiliated Hospital from January 2020 to December 2023 were retrospectively selected through a HD monitoring system. Patient data were collected, and the patients were divided into training and validation sets in a 7:3 ratio. The training set was used to establish the model, which was verified using the validation set. Multiple logistic regression analysis was performed to identify risk factors for central venous CRI and develop a nomogram prediction model. A total of 300 MHD patients were enrolled. Multivariate analysis showed that catheter duration, catheter site, catheter reinsertion, history of catheter infection, diabetes, and albumin <35 g/L were risk factors for central venous CRI. The area under the receiver operating characteristic (ROC) curve (AUC) for the training set was 0.902 (95% confidence interval (CI) = 0.862-0.941), with a sensitivity of 85.7%, specificity of 80%, and a Youden index of 65.7%, and that for the validation set was 0.826 (95% CI = 0.726-0.905), with a sensitivity of 80.5%, specificity of 77.9%, and a Youden index of 58.4%. The model demonstrated good discrimination and calibration (Hosmer-Lemeshow goodness-of-fit test statistics: training set: χ2 = 4.709, p = 0.788; validation set: χ2 = 7.171, p = 0.518). This study identified six risk factors associated with central venous CRI in MHD patients. This predictive model demonstrates good prognostic performance and can be used by clinicians to screen for high-risk patients with central venous CRI, thereby enabling the early implementation of risk management strategies.

End-to-End CT Radiomics-Based Pipeline for Predicting Renal Interstitial Fibrosis Grade in CKD Patients.

Non-invasive assessment of renal fibrosis in patients with chronic kidney disease (CKD) remains a clinical challenge. This study aims to integrate radiomics and clinical factors to develop an end-to-end pipeline for predicting interstitial fibrosis (IF) in CKD patients. This retrospective study included 80 patients with CKD, with 53 patients in training set and 27 patients in test set. All patients underwent renal computed tomography (CT) scans and biopsy. Patients were classified into two groups based on their renal IF grade: mild-moderate and severe. Radiomics features were extracted from the automatically segmented right renal region on CT images, and univariate analysis along with multiple Least Absolute Shrinkage and Selection Operator (LASSO) was employed to construct the radiomics signature. Subsequently, logistic regression models were developed to create the radiomics model and the combined model. The predictive performance of both models was evaluated through discrimination, calibration, and decision curve analysis (DCA), and a nomogram was constructed for the model demonstrating superior performance. The combined model significantly outperformed the radiomics model, achieving a cross-validated AUC of 0.935±0.041 in the training set, compared to 0.804±0.024 for the radiomics model. In the test set, the combined model outperformed the radiomics model, with an AUC of 0.918 [95% CI 0.799-1] vs. 0.764 [95% CI 0.549-0.979], p=0.031 (DeLong test, Statistic: -2.152). Calibration curves and DCA indicated that the combined model demonstrated good calibration and better clinical net benefit. This end-to-end workflow could serve as a potential non-invasive tool to predict renal IF grade (mild-moderate vs. severe) in CKD patients.

High-throughput methylation sequencing reveals novel biomarkers for the early detection of renal cell carcinoma

PurposeRenal cell carcinoma (RCC) is a common malignancy, with patients frequently diagnosed at an advanced stage due to the absence of sufficiently sensitive detection technologies, significantly compromising patient survival and quality of life. Advances in cell-free DNA (cfDNA) methylation profiling using liquid biopsies offer a promising non-invasive diagnostic option, but robust biomarkers for early detection are current not available. This study aimed to identify methylation biomarkers for RCC and establish a DNA methylation signature-based prognostic model for this disease.MethodsHigh-throughput methylation sequencing was performed on peripheral blood samples obtained from 49 primarily Stage I RCC patients and 44 healthy controls. Comparative analysis and Least Absolute Shrinkage and Selection Operator (LASSO) regression methods were employed to identify RCC methylation signatures.Subsequently, methylation markers-based diagnostic and prognostic models for RCC were independently trained and validated using random forest and Cox regression methodologies, respectively.ResultsComparative analysis revealed 864 differentially methylated CpG islands (DMCGIs), 96.3% of which were hypermethylated. Using a training set from The Cancer Genome Atlas (TCGA) dataset of 443 early-stage RCC tumors and matched normal tissues, we applied LASSO regression and identified 23 methylation signatures. We then constructed a random forest-based diagnostic model for early-stage RCC and validated the model using two independent datasets: a TCGA set of 460 RCC tumors and controls, and a blood sample set from our study of 15 RCC cases and 29 healthy controls. For Stage I RCC tissue, the model showed excellent discrimination (AUC-ROC: 0.999, sensitivity: 98.5%, specificity: 100%). Blood sample validation also yielded commendable results (AUC-ROC: 0.852, sensitivity: 73.9%, specificity: 89.7%). Further analysis using Cox regression identified 7 of the 23 DMCGIs as prognostic markers for RCC, allowing the development of a prognostic model with strong predictive power for 1-, 3-, and 5-year survival (AUC-ROC > 0.7).ConclusionsOur findings highlight the critical role of hypermethylation in RCC etiology and progression, and present these identified biomarkers as promising candidates for diagnostic and prognostic applications.

Diagnosing Epilepsy with Normal Interictal EEG Using Dynamic Network Models.

Whereas a scalp electroencephalogram (EEG) is important for diagnosing epilepsy, a single routine EEG is limited in its diagnostic value. Only a small percentage of routine EEGs show interictal epileptiform discharges (IEDs) and overall misdiagnosis rates of epilepsy are 20% to 30%. We aim to demonstrate how network properties in EEG recordings can be used to improve the speed and accuracy differentiating epilepsy from mimics, such as functional seizures - even in the absence of IEDs. In this multicenter study, we analyzed routine scalp EEGs from 218 patients with suspected epilepsy and normal initial EEGs. The patients' diagnoses were later confirmed based on an epilepsy monitoring unit (EMU) admission. About 46% ultimately being diagnosed with epilepsy and 54% with non-epileptic conditions. A logistic regression model was trained using spectral and network-derived EEG features to differentiate between epilepsy and non-epilepsy. Of the 218 patients, 90% were used for training and 10% were held out for testing. Within the training set, 10-fold cross validation was performed. The resulting tool was named "EpiScalp." EpiScalp achieved an area under the curve (AUC) of 0.940, an accuracy of 0.904, a sensitivity of 0.835, and a specificity of 0.963 in classifying patients as having epilepsy or not. EpiScalp provides an accurate diagnostic aid from a single initial EEG recording, even in more challenging epilepsy cases with normal initial EEGs. This may represent a paradigm shift in epilepsy diagnosis by deriving an objective measure of epilepsy likelihood from previously uninformative EEGs. ANN NEUROL 2025.

Development of a disease diagnostic model to predict the occurrence of central precocious puberty of female.

To develop a clinical model for predicting the occurrence of Central Precocious Puberty based on the breast development outcomes in chinese girls. This is a retrospective study, which included a total of 1,001 girls aged 6-9 years old who visited the outpatient clinic of Beijing Children's Hospital from January 2017 to October 2022 for "breast development". Participants were categorized into pubertal development (PD) cohort and simple premature breast development (PT) according to the criteria, and information was collected and tested for relevant indicators. After dealing with missing data, logistic regression, LASSO regression and random forest were used to screen the variables, and support vector machine models were built with SMOTE oversampling and ten-fold cross-validation to assess the effectiveness of the models in the training and validation sets. 1,001 girls were included in the analysis, of whom 369 (36.9 %) were diagnosed with PD and 632 (63.1 %) with PT. Body mass index (BMI), bone age (BA), luteinizing hormone (LH), follicle stimulating hormone (FSH), estradiol (E2), uterine diameter, and ovary volume were identified as the final predictor variables by three variable screening methods. The AUC of the constructed disease diagnostic model was 0.9457 in the developmental cohort and 0.8357 inthe external validation group, and sensitivity analyses revealed that the performance of the constructed models with different variable selection strategies was similar. A disease diagnostic model was developed that may help predict a girl's risk of diagnosing central precocious puberty.

Predictive risk model of mild cognitive impairment in patients with malignant haematological diseases after haematopoietic stem cell transplantation

ObjectiveThis study is to develop and validate a robust risk prediction model for mild cognitive impairment (MCI) in patients with malignant haematological diseases after haematopoietic stem cell transplantation (HSCT).MethodsIn this study, we analysed the clinical data of the included patients. Logistic regression analysis was used to identify independent risk factors for cognitive impairment after HSCT in patients with malignant haematological diseases, and a risk prediction model was constructed. Multiple cohorts of patients with haematological malignancies after HSCT (282 cases) from the Affiliated Hospital of Xuzhou Medical University and the First People’s Hospital of Yancheng City between April 2019 and February 2022, and patients from the Affiliated Huaian No.1 People’s Hospital of Nanjing Medical University between March 2022 and July 2023 were used for external validation. Logistic regression analysis was performed to develop the predictive model. The predictive value and consistency of the model were evaluated using the area under the curve (AUC) and calibration method, respectively. Decision curve analysis (DCA) was performed to access the utility of the model.ResultsApproximately half (52.26%) of the patients in the study developed mild cognitive impairment (MCI). Older age, allogeneic HSCT, anxiety, graft-versus-host disease, and longer hospital stay were associated with a higher risk of developing MCI. ROC curve analysis confirmed the sound performance of the predictive model and external validation, with AUC of 0.897 and 0.789 respectively. The direction of the calibration curves of the training and validation sets is closer to the diagonal (ideal curve), indicating good model consistency; the DCA curves also show that the model has good predictive ability and stability.ConclusionsWe conclude that it is possible to predict mild cognitive impairment with readily available, mostly pretransplant predictors. The accuracy of the risk prediction models can be improved for use in clinical practice, possibly by adding pretransplant patient-reported functioning and comorbidities.