Area Under The Receiver Operating Characteristic Research Articles

Frailty in intracranial meningioma resection: the risk analysis index demonstrates strong discrimination for predicting non-home discharge and in-hospital mortality.

Frailty is an independent risk factor for adverse postoperative outcomes following intracranial meningioma resection (IMR). The role of the Risk Analysis Index (RAI) in predicting postoperative outcomes following IMR is nascent but may inform preoperative patient selection and surgical planning. IMR patients from the Nationwide Inpatient Sample were identified using diagnostic and procedural codes (2019-2020). The relationship between preoperative RAI-measured frailty and primary outcomes (non-home discharge (NHD), in-hospital mortality) and secondary outcomes (extended length of stay (eLOS), complication rates) was assessed via multivariate analyses. The discriminatory accuracy of the RAI for primary outcomes was measured in area under the receiver operating characteristic (AUROC) curve analysis. A total of 23,230 IMR patients (mean age = 59) were identified, with frailty statuses stratified by RAI score: 0-20 "robust" (R)(N = 10,665, 45.9%), 21-30 "normal" (N)(N = 8,895, 38.3%), 31-40 "frail" (F)(N = 2,605, 11.2%), and 41+ "very frail" (VF)(N = 1,065, 4.6%). Rates of NHD (R 11.5%, N 29.7%, F 60.8%, VF 61.5%), in-hospital mortality (R 0.5%, N 1.8%, F 3.8%, VF 7.0%), eLOS (R 13.2%, N 21.5%, F 40.9%, VF 46.0%), and complications (R 7.5%, N 11.6%, F 15.7%, VF 16.0%) significantly increased with increasing frailty thresholds (p < 0.001). The RAI demonstrated strong discrimination for NHD (C-statistic: 0.755) and in-hospital mortality (C-statistic: 0.754) in AUROC curve analysis. Increasing RAI-measured frailty is significantly associated with increased complication rates, eLOS, NHD, and in-hospital mortality following IMR. The RAI demonstrates strong discrimination for predicting NHD and in-hospital mortality following IMR, and may aid in preoperative risk stratification.

Differential diagnosis of thyroid nodules using heterogeneity quantification software on ultrasound images: correlation with the Bethesda system and surgical pathology

Ultrasonography (US)-guided fine-needle aspiration cytology (FNAC) is the primary modality for evaluating thyroid nodules. However, in cases of atypia of undetermined significance (AUS) or follicular lesion of undetermined significance (FLUS), supplemental tests are necessary for a definitive diagnosis. Accordingly, we aimed to develop a non-invasive quantification software using the heterogeneity scores of thyroid nodules. This cross-sectional study retrospectively enrolled 188 patients who were categorized into four groups according to their diagnostic classification in the Bethesda system and surgical pathology [II-benign (B) (n = 24); III-B (n = 52); III-malignant (M) (n = 54); V/VI-M (n = 58)]. Heterogeneity scores were derived using an image pixel-based heterogeneity index, utilized as a coefficient of variation (CV) value, and analyzed across all US images. Differences in heterogeneity scores were compared using one-way analysis of variance with Tukey’s test. Diagnostic accuracy was determined by calculating the area under the receiver operating characteristic (AUROC) curve. The results of this study indicated significant differences in mean heterogeneity scores between benign and malignant thyroid nodules, except in the comparison between III-M and V/VI-M nodules. Among malignant nodules, the Bethesda classification was not observed to be associated with mean heterogeneity scores. Moreover, there was a positive correlation between heterogeneity scores and the combined diagnostic category, which was based on the Bethesda system and surgical cytology grades (R = 0.639, p < 0.001). AUROC for heterogeneity scores showed the highest diagnostic performance (0.818; cut-off: 30.22% CV value) for differentiating the benign group (normal/II-B/III-B) from the malignant group (III-M/V&VI-M), with a diagnostic accuracy of 72.5% (161/122). Quantitative heterogeneity measurement of US images is a valuable non-invasive diagnostic tool for predicting the likelihood of malignancy in thyroid nodules, including AUS or FLUS.

Deep learning-based model for difficult transfemoral access prediction compared with human assessment in stroke thrombectomy

BackgroundIn mechanical thrombectomy (MT), extracranial vascular tortuosity is among the main determinants of procedure duration and success. Currently, no rapid and reliable method exists to identify the anatomical features precluding...

Machine learning prediction model for postoperative ileus following colorectal surgery.

Postoperative ileus (POI) continues to be a major cause of morbidity following colorectal surgery. Despite best efforts, the incidence of POI in colorectal surgery remains high (~30%). This study aimed to investigate machine learning techniques to identify risk factors for POI in colorectal surgery patients, to help guide further preventative strategies. A TRIPOD-guideline-compliant retrospective study was conducted for major colorectal surgery patients at a single tertial care centre (2018-2022). The primary outcome was the occurrence of POI, defined as not achieving GI-2 (outcome measure of time to first stool and tolerance of oral diet) by day four. Multivariate logistic regression, decision trees, radial basis function and multilayer perceptron (MLP) models were trained using a random allocation of patients to training/testing data sets (80/20). The area under the receiver operating characteristic (AUROC) curves were used to evaluate model performance. Of 504 colorectal surgery patients, 183 (36%) experienced POI. Multivariate logistic regression, decision trees, radial basis function and MLP models returned an AUROC of 0.722, 0.706, 0.712 and 0.800, respectively. The MLP model had the highest sensitivity and specificity values. In addition to well-known risk factors for POI, such as postoperative hypokalaemia, surgical approach, and opioid use, the MLP model identified sarcopenia (ranked 4/30) as a potentially modifiable risk factor for POI. MLP outperformed other models in predicting POI. Machine learning can provide valuable insights into the importance and ranking of specific predictive variables for POI. Further research into the predictive value of preoperative sarcopenia for POI is required.

The influence of gender-specific factors influencing severe anxiety in psychotic major depression: role of thyroid hormones and depression severity.

Psychotic major depression (PMD) is characterized by major depressive disorder (MDD) accompanied by delusions or hallucinations. While the prevalence of PMD and its association with anxiety have been studied, gender-specific differences and the role of thyroid hormones in PMD-related anxiety remain less explored. A total of 1718 first-episode and drug-naïve MDD patients was assessed for the presence of PMD and severe anxiety. Clinical assessments, including Hamilton Depression Rating Scale (HAMD), Hamilton Anxiety Rating Scale (HAMA), Positive and Negative Syndrome Scale (PANSS), and Clinical Global Impressions-Severity (CGI-S) scale, were conducted to assess depression, anxiety, psychotic symptoms, and clinical severity, respectively. Blood samples were collected to measure thyroid function parameters. The prevalence of severe anxiety was higher in PMD patients compared to non-psychotic MDD patients (71.3% vs. 5.3%). No significant gender differences were observed in the prevalence of severe anxiety among PMD patients. However, elevated thyroid-stimulating hormone (TSH) levels and increased depression severity (HAMD scores) were identified as independent risk factors for severe anxiety in female PMD patients. In contrast, no significant risk factors were found in male PMD patients. The area under the receiver operating characteristic (AUCROC) analysis revealed that the HAMD score and TSH level showed acceptable discriminatory capacity for distinguishing between female PMD patients with and without severe anxiety. This study highlights the heightened prevalence of severe anxiety in PMD patients, with TSH levels and depression severity emerging as gender-specific risk factors for anxiety in females. These findings suggest the importance of thyroid hormone assessment and tailored interventions for managing anxiety in female PMD patients.

Validating the Baveno Elastography Criteria of Advanced Liver Fibrosis in Two-Dimensional Shear Wave Elastography: A Prospective Pathology-Based Study

Introduction: The Baveno criteria for assessing advanced liver fibrosis were mainly determined by transient elastography (TE), and its pathology-based validation studies in two-dimensional shear wave elastography (2D-SWE) remain limited. We aimed to validate the Baveno criteria through use of 2D-SWE. Method: Consecutive patients who underwent liver biopsies for various benign liver diseases were prospectively recruited. Liver stiffness measurement (LSM) was simultaneously evaluated by TE and 2D-SWE. The optimal cutoff value to predict advanced liver fibrosis was determined by the Youden Index, and the diagnostic performance was estimated using area under the receiver operating characteristic (AUROC) analysis. Results: A total of 101 patients were enrolled having a median age of 55.0 (IQR: 46.0–63.5) years, with 53 (52.48%) of them being male. Using <9 and >14 kPa as the optimal dual cutoffs, the AUROC values in TE and 2D-SWE were 0.92 (95% CI: 0.83–0.97) and 0.93 (95% CI: 0.84–0.98), respectively (p = 0.61). The sensitivity and specificity of LSM by TE/2D-SWE achieved rates of 94.44%/94.44% and 86.00%/88.00%, respectively. However, using the Baveno criteria, the AUROC values in TE and 2D-SWE could remain achieving 0.91 (95% CI: 0.82–0.97) and 0.93 (95% CI: 0.84–0.98), respectively (p = 0.36). The sensitivity and specificity in TE/2D-SWE were 88.24%/88.24% and 86.79%/90.57%, respectively. Conclusion: This study establishes the compatibility of the Baveno dual cutoff criteria with 2D-SWE, positioning it as an easily used criteria in clinical practice and research.

Abstract PO3-07-04: Prediction of PAM50 molecular subtypes from H&amp;E-stained breast cancer specimens using tumor microenvironment features and additive multiple instance learning models

Abstract Background: PAM50, a 50-gene signature, classifies breast cancers into one of five subtypes (basal, luminal A, luminal B, HER2-enriched, and normal-like), revealing information about underlying tumor biology, and has emerged as a key prognostic indicator influencing treatment decisions. There is growing interest in bridging the gap between expression-based metrics and histopathology, where immunohistochemistry (IHC) and sequencing-based approaches have been proposed for this purpose. However, hematoxylin and eosin (H&amp;E)-stained slides are ubiquitously utilized by pathologists for cancer diagnosis, while IHC and sequencing-based approaches require additional tissue and specialized processing and/or analysis. Here, we describe a computer vision-based approach to predict PAM50 classification using H&amp;E-stained whole slide images (WSIs). Methods: We obtained expression-based PAM50 subtype labels and corresponding H&amp;E-stained WSIs for 961 breast carcinomas from the TCGA BRCA cohort. We used two separate machine learning (ML) approaches to predict PAM50 subtypes from WSIs. In the first approach, we deployed previously trained PathExplore models to extract quantitative human-interpretable features (HIFs) that summarize the TME. We subsequently trained random forest classification models on these HIFs to predict PAM50 subtypes. For the second approach, we developed additive multiple instance learning (aMIL) models. Additionally, we explored the effects of PAM50 subtype labeling and aggregation strategies beyond the 5-class approach. Our 3-class approach combines Luminal A and B, as seen in IHC efforts to increase agreement with PAM50 assays, while excluding Normal, a category containing few and heterogeneous samples. We also performed binary classification for each subtype in the 3-class model (e.g. luminal vs. other). Slides were split into training (60%), validation (20%), and test (20%) sets, stratified by PAM50 labels, and model performance was assessed using the area under the receiver operator curve (AUROC) metric on the held-out test set, using a one vs. rest approach for multi-class models. To establish a baseline for PAM50 prediction, we developed random forest classification models using only clinical covariates (tumor stage, histologic grade, histological subtype, and BRCA1/2 status). Results: We compared the performance of our two ML models (HIF and aMIL) to that of the baseline model, and we report the AUROC values in Table 1. These models both performed well in predicting Basal, Luminal A, Luminal B, and Luminal (A+B), while the model performance was less strong for predictions of the HER2 and Normal classifications. The three-class model showed improved performance of predicting Luminal classifications relative to the five-class model that separates Luminal A and B. Although simplifying classification problems to a binary use case typically provides improved performance, this phenomenon was not observed for any of the PAM50 subtypes. Conclusions: These results demonstrate that AI-powered digital pathology can accurately and reproducibly perform molecular-based classification tasks, such as predicting PAM50 classifications, using WSIs, suggesting a more efficient path toward clinically relevant breast cancer characterization. Table 1. Performance of all models in predicting PAM50 molecular subtypes. AUROC values are shown. Shaded cells represent the best test-set performance for each class (row). Citation Format: Maria Guramare, Syed Ashar Javed, Christian Kirkup, Dinkar Juyal, Jacqueline Brosnan-Cashman, Victoria Mountain, Ryan Leung, Bahar Rahsepar, John Abel, Amaro Taylor-Weiner, Jake Conway. Prediction of PAM50 molecular subtypes from H&amp;E-stained breast cancer specimens using tumor microenvironment features and additive multiple instance learning models [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-07-04.

Automated annotation of disease subtypes

BackgroundDistinguishing diseases into distinct subtypes is crucial for study and effective treatment strategies. The Open Targets Platform (OT) integrates biomedical, genetic, and biochemical datasets to empower disease ontologies, classifications, and potential gene targets. Nevertheless, many disease annotations are incomplete, requiring laborious expert medical input. This challenge is especially pronounced for rare and orphan diseases, where resources are scarce. MethodsWe present a machine learning approach to identifying diseases with potential subtypes, using the approximately 23,000 diseases documented in OT. We derive novel features for predicting diseases with subtypes using direct evidence. Machine learning models were applied to analyze feature importance and evaluate predictive performance for discovering both known and novel disease subtypes. ResultsOur model achieves a high (89.4%) ROC AUC (Area Under the Receiver Operating Characteristic Curve) in identifying known disease subtypes. We integrated pre-trained deep-learning language models and showed their benefits. Moreover, we identify 515 disease candidates predicted to possess previously unannotated subtypes. ConclusionsOur models can partition diseases into distinct subtypes. This methodology enables a robust, scalable approach for improving knowledge-based annotations and a comprehensive assessment of disease ontology tiers. Our candidates are attractive targets for further study and personalized medicine, potentially aiding in the unveiling of new therapeutic indications for sought-after targets.

Validating the 2023 FIGO staging system: A nomogram for endometrioid endometrial cancer and adenocarcinoma.

To find the factors impacting overall survival (OS) prognosis in patients with endometrioid endometrial carcinoma (EEC) and adenocarcinoma and to establish a nomogram model to validate the 2023 International Federation of Obstetrics and Gynecology (FIGO) staging system for endometrial cancer. Data were obtained from the Surveillance, Epidemiology, and End Results (SEER) training cohort. An independent validation cohort was obtained from the First Affiliated Hospital of Anhui Medical University between 2008 and 2023. Cox regression analysis identified independent prognostic factors for OS in EEC and adenocarcinoma patients. A nomogram predicting OS was developed and validated utilizing the C-index, calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA). The relationship between the tumor grade and prognosis of EEC and adenocarcinoma was quantified using net reclassification improvement (NRI), propensity score matching (PSM), and Kaplan-Meier curves. Cox regression analysis identified age, race, marital status, tumor grade, tumor stage, tumor size, and chemotherapy as independent prognostic factors for OS. A nomogram for predicting OS was developed based on these factors. The C-indexes for the OS nomogram was 0.743 and 0.720 for the SEER training set and external validation set, respectively. The area under the ROC (AUC) for the OS nomogram was 0.755, 0.757, and 0.741 for the SEER data subsets and 0.844, 0.719, and 0.743 for the external validation subsets. Calibration plots showed high concordance between the nomogram-predicted and observed OS. DCA also demonstrated the clinical utility of the OS nomogram. NRI, PSM, and survival analyses revealed that tumor grade was the most important histopathological factor for EEC and adenocarcinoma prognosis. Seven independent prognostic variables for the OS of patients with EEC and adenocarcinoma were identified. The established OS nomogram has good predictive ability and clinical utility and validates the 2023 endometrial cancer FIGO staging system.

Development of a Novel Statistical Model for Predicting Clinical Outcomes in Stroke Patients With Tandem Occlusions After Endovascular Therapy.

Tandem occlusions are intracranial large vessel occlusions (LVOs) with a concomitant ipsilateral extracranial internal carotid artery occlusionand can cause more severe stroke symptoms. To develop a simple, rigorously cross-validated novel tool to predict clinical outcomes following tandem occlusion in patients with acute LVO stroke, based on data that are easily available to clinicians. To have used machine learning approaches to utilize the available information from clinical and angiographic data to make predictive models able to distinguish between mortality versus survival and good (modified Rankin Scale (mRS) ≤ 2) versus unfavorable neurological outcomes (mRs ≥ 3) Materials and methods: Retrospective data from 87 consecutive patients with anterior circulation stroke and tandem occlusions who underwent mechanical thrombectomy and stenting between December 2009 and January 2020 were analyzed. Patients were stratified into three groups based on the location of their LVO, and these groups were compared using statistical tests. Predictive models were built and cross-validated 1000 times to estimate their predictive power, measured by accuracy and area under the receiver operating curve (AUROC). For distinguishing good outcome (mRS ≤ 2) versus poor outcome (mRS ≥ 3), the model comprised age, initial National Institute of Health Stroke Scale (NIHSS) score, Alberta Stroke Program Early CT Score (ASPECTS), NIHSS at 24 hours, NIHSS at discharge and intracranial haemorrhage and yielded an accuracy of 83% and the AUROC of 0.91. For mortality prediction, the model comprised age, initial NIHSS, intravenous thrombolysis, NIHSS at 24 hours and NIHSS at discharge and yielded an accuracy of 91% and an AUROC of 0.94. Models developed exhibit strong predictive performance and can distinguish between both the instances of survival versus mortality and good versus poor outcome with an aim to support clinicians in deciding on optimal management for these complex patients.The developed model will help identify those at risk of poorer outcomes and the prospective better selection of patients with acute ischaemic large vessel stroke secondary to tandem occlusions.

Preclinical identification of acute coronary syndrome without high sensitivity troponin assays using machine learning algorithms

Preclinical management of patients with acute chest pain and their identification as candidates for urgent coronary revascularization without the use of high sensitivity troponin essays remains a critical challenge in emergency medicine. We enrolled 2760 patients (average age 70 years, 58.6% male) with chest pain and suspected ACS, who were admitted to the Emergency Department of the University Hospital Tübingen, Germany, between August 2016 and October 2020. Using 26 features, eight Machine learning models (non-deep learning models) were trained with data from the preclinical rescue protocol and compared to the “TropOut” score (a modified version of the “preHEART” score which consists of history, ECG, age and cardiac risk but without troponin analysis) to predict major adverse cardiac event (MACE) and acute coronary artery occlusion (ACAO). In our study population MACE occurred in 823 (29.8%) patients and ACAO occurred in 480 patients (17.4%). Interestingly, we found that all machine learning models outperformed the “TropOut” score. The VC and the LR models showed the highest area under the receiver operating characteristic (AUROC) for predicting MACE (AUROC = 0.78) and the VC showed the highest AUROC for predicting ACAO (AUROC = 0.81). A SHapley Additive exPlanations (SHAP) analyses based on the XGB model showed that presence of ST-elevations in the electrocardiogram (ECG) were the most important features to predict both endpoints.

Impact of calcification on Murray law-based quantitative flow ratio for physiological assessment of intermediate coronary stenoses.

To investigate the influence of coronary calcification on the diagnostic performance of Murray law-based quantitative flow ratio (μQFR) in identifying hemodynamically significant coronary lesions referenced to fractional flow reserve (FFR). A total of 571 intermediate lesions from 534 consecutive patients (66.1 ± 10.0 years, 67.2% males) who underwent coronary angiography and simultaneous FFR measurement were included. Calcific deposits were graded by angiography as none or mild (spots), moderate (involving ≤ 50% of the reference vessel diameter), and severe (> 50%). Performance of μQFR to detect functional ischemia (FFR ≤ 0.80) was evaluated, including diagnostic parameters and areas under the receiver-operating curves (AUCs). The discrimination of ischemia by μQFR was comparable between none/mild and moderate/severe calcification (AUC: 0.91 [95% confidence interval: 0.88-0.93] vs. 0.87 [95% confidence interval: 0.78-0.94]; p = 0.442). No statistically significant difference was observed for μQFR between the two categories in sensitivity (0.70 vs. 0.69, p = 0.861) and specificity (0.94 vs. 0.90, p = 0.192). Moreover, μQFR showed significantly higher AUCs than quantitative coronary angiographic diameter stenosis in both vessels with none/mild (0.91 vs. 0.78, p < 0.001) and moderate/severe calcification (0.87 vs. 0.69, p < 0.001). By multivariable analysis, there was no association between calcification and μQFR-FFR discordance (adjusted odds ratio: 1.529, 95% confidence interval: 0.788-2.968, p = 0.210) after adjustment for other confounding factors. μQFR demonstrated robust and superior diagnostic performance for lesion-specific ischemia compared with angiography alone regardless of coronary calcification.

Predicting osteoporosis from kidney-ureter-bladder radiographs utilizing deep convolutional neural networks

Osteoporosis is a common condition that can lead to fractures, mobility issues, and death. Although dual-energy X-ray absorptiometry (DXA) is the gold standard for osteoporosis, it is expensive and not widely available. In contrast, kidney-ureter-bladder (KUB) radiographs are inexpensive and frequently ordered in clinical practice. Thus, it is a potential screening tool for osteoporosis. In this study, we explored the possibility of predicting the bone mineral density (BMD) and classifying high-risk patient groups using KUB radiographs. We proposed DeepDXA-KUB, a deep learning model that predicts the BMD values of the left hip and lumbar vertebrae from an input KUB image. The datasets were obtained from Taiwanese medical centers between 2006 and 2019, using 8913 pairs of KUB radiographs and DXA examinations performed within 6 months. The images were randomly divided into training and validation sets in a 4:1 ratio. To evaluate the model's performance, we computed a confusion matrix and evaluated the sensitivity, specificity, accuracy, precision, positive predictive value, negative predictive value, F1 score, and area under the receiver operating curve (AUROC). Moderate correlations were observed between the predicted and DXA-measured BMD values, with a correlation coefficient of 0.858 for the lumbar vertebrae and 0.87 for the left hip. The model demonstrated an osteoporosis detection accuracy, sensitivity, and specificity of 84.7 %, 81.6 %, and 86.6 % for the lumbar vertebrae and 84.2 %, 91.2 %, and 81 % for the left hip, respectively. The AUROC was 0.939 for the lumbar vertebrae and 0.947 for the left hip, indicating a satisfactory performance in osteoporosis screening. The present study is the first to develop a deep learning model based on KUB radiographs to predict lumbar spine and femoral BMD. Our model demonstrated a promising correlation between the predicted and DXA-measured BMD in both the lumbar vertebrae and hip, showing great potential for the opportunistic screening of osteoporosis.

The use of artificial intelligence in reconstructive surgery for head and neck cancer: a systematic review.

The popularity of artificial intelligence (AI) in head and neck cancer (HNC) management is increasing, but postoperative complications remain prevalent and are the main factor that impact prognosis after surgery. Hence, recent studies aim to assess new AI models to evaluate their ability to predict free flap complications more effectively than traditional algorithms. This systematic review aims to summarize current evidence on the utilization of AI models to predict complications following reconstructive surgery for HNC. A combination of MeSH terms and keywords was used to cover the following three subjects: "HNC," "artificial intelligence," and "free flap or reconstructive surgery." The electronic literature search was performed in three relevant databases: Medline (Ovid), Embase (Ovid), and Cochrane. Quality appraisal of the included study was conducted using the TRIPOD Statement. The review included a total of 5 manuscripts (n = 5) for a total of 7524 patients. Across studies, the highest area under the receiver operating characteristic (AUROC) value achieved was 0.824 by the Auto-WEKA model. However, only 20% of reported AUROCs exceeded 0.70. One study concluded that most AI models were comparable or inferior in performance to conventional logistic regression. The highest predictors of complications were flap type, smoking status, tumour location, and age. Some models showed promising results. Predictors identified across studies were different than those found in existing literature, showing the added value of AI models. However, the algorithms showed inconsistent results, underlying the need for better-powered studies with larger databases before clinical implementation.

Risk score prediction for bacteriologically confirmed tuberculosis among adults with HIV on antiretroviral therapy in northwest Ethiopia: prognostic model development.

This study was aimed at developing a risk score prediction model for bacteriologically confirmed tuberculosis (TB) among adults with HIV receiving antiretroviral therapy in Ethiopia. An institutional-based retrospective follow-up study was conducted among 569 adults with HIV on ART. We used demographic and clinical prognostic factors to develop a risk prediction model. Model performance was evaluated by discrimination and calibration using the area under the receiver operating characteristic (AUROC) curve and calibration plot. Bootstrapping was used for internal validation. A decision curve analysis was used to evaluate the clinical utility. Opportunistic infection, functional status, anemia, isoniazid preventive therapy, and WHO clinical stages were used to develop risk prediction. The AUROC curve of the original model was 87.53% [95% confidence interval (CI): 83.88-91.25] and the calibration plot ( P -value = 0.51). After internal validation, the AUROC curve of 86.61% (95% CI: 82.92-90.29%) was comparable with the original model, with an optimism coefficient of 0.0096 and good calibration ( P -value = 0.10). Our model revealed excellent sensitivity (92.65%) and negative predictive value (NPV) (98.60%) with very good specificity (70.06%) and accuracy (72.76%). After validation, accuracy (74.85%) and specificity (76.27%) were improved, but sensitivity (86.76%) and NPV (97.66%) were relatively reduced. The risk prediction model had a net benefit up to 7.5 threshold probabilities. This prognostic model had very good performance. Moreover, it had very good sensitivity and excellent NPV. The model could help clinicians use risk estimation and stratification for early diagnosis and treatment to improve patient outcomes and quality of life.

Endoscopic Ultrasound-based Shear Wave Elastography for Detection of Advanced Liver Disease.

Endoscopic ultrasound shear wave elastography (EUS-SWE) is a novel modality for liver stiffness measurement. The aims of this study are to evaluate the performance and reliability of EUS-SWE for detecting advanced liver disease in a prospective cohort. EUS-SWE measurements were prospectively obtained from patients undergoing EUS between August 2020 and March 2023. Liver stiffness measurements were compared between patients with and without advanced liver disease (ALD), defined as stage ≥3, to determine diagnostic accuracy for advanced fibrosis and portal hypertension. Logistic regression was performed to identify variables that impact the reliability of EUS-SWE readings. Select patients underwent paired magnetic resonance elastography (MRE) for liver fibrosis correlation. Patients with ALD demonstrated higher liver stiffness compared to healthy controls (left lobe: 17.6 vs. 12.7 kPa, P<0.001; median right lobe: 24.8 vs. 11.0 kPa, P<0.001). The area under the receiver operator characteristic (AUROC) for the detection of ALD was 0.73 and 0.80 for left and right lobe measurements, respectively. General anesthesia was associated with reliable EUS-SWE liver readings (odds ratio: 2.73, 95% CI: 1.07-7.39, P=0.040). Left lobe measurements correlated significantly with MRE with an increase of 0.11 kPa (95% CI: 0.05-0.17 kPA) for every 1 kPa increase on EUS-SWE. D. SWE is a promising technology that can readily be incorporated into standard EUS examinations for the assessment of ALD.

Development and validation of a nomogram model for predicting the risk of MAFLD in the young population

This study aimed to develop and validate a nomogram model that includes clinical and laboratory indicators to predict the risk of metabolic-associated fatty liver disease (MAFLD) in young Chinese individuals. This study retrospectively analyzed a cohort of young population who underwent health examination from November 2018 to December 2021 at The Affiliated Hospital of Southwest Medical University in Luzhou City, Sichuan Province, China. We extracted the clinical and laboratory data of 43,040 subjects and randomized participants into the training and validation groups (7:3). Univariate logistic regression analysis, the least absolute shrinkage and selection operator regression, and multivariate logistic regression models identified significant variables independently associated with MAFLD. The predictive accuracy of the model was analyzed in the training and validation sets using area under the receiver operating characteristic (AUROC), calibration curves, and decision curve analysis. In this study, we identified nine predictors from 31 variables, including age, gender, body mass index, waist-to-hip ratio, alanine aminotransferase, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, uric acid, and smoking. The AUROC for the subjects in the training and validation groups was 0.874 and 0.875, respectively. The calibration curves show excellent accuracy of the nomogram. This nomogram which was based on demographic characteristics, lifestyle habits, anthropometrics, and laboratory data can visually and individually predict the risk of developing MAFLD. This nomogram is a quick and effective screening tool for assessing the risk of MAFLD in younger populations and identifying individuals at high risk of MAFLD, thereby contributing to the improvement of MAFLD management.

Clinical evaluation of pulsatile tinnitus: history and physical examination techniques to predict vascular etiology

BackgroundPulsatile tinnitus (PT) may be due to a spectrum of cerebrovascular etiologies, ranging from benign venous turbulence to life threatening dural arteriovenous fistulas. A focused clinical history and physical examination...

Machine learning for persistent free radicals in biochar: dual prediction of contents and types using regression and classification models

Persistent free radicals (PFRs) are emerging substances with diverse impacts in biochar applications, necessitating accurate prediction of their content and types prior to their optimal use and minimal adverse effects. This prediction task is challenging due to the nonlinearity and intricate variable relationships of biochar. Herein, we employed data-driven techniques to compile a dataset from peer-reviewed publications, aiming to systematically predict the PFRs by developing supervised machine learning models. Notably, extreme gradient boosting (XGBoost) model exhibited the best predictive performance for both regression and classification tasks in predicting the PFRs, achieving a test R2 value of 0.95 for PFR content prediction, along with an Area Under the Receiver Operating Curve (AUROC) of 0.92 for PFR type prediction, respectively. Based on XGBoost model, a graphical user interface (GUI) was developed to access PFRs predictions. Analysis of feature importance revealed that the biochar properties, such as metal/non-metal doping, pyrolysis temperature, carbon content, and specific surface area were identified as the four most significant factors influencing PFRs contents. Regarding the types of PFRs in biochar, specific surface area, pyrolysis temperature, carbon content, and feedstock were top-ranked influencing factors. These findings provide valuable guidance for accurately predicting both the contents and types of PFRs in biochar, and also hold significant potential for highly efficient utilization of biochar across various applications.Graphical

Machine learning for predicting colon cancer recurrence

Colorectal cancer (CRC) is a global public health concern, ranking among the most commonly diagnosed malignancies worldwide. Despite advancements in treatment modalities, the specter of CRC recurrence remains a significant challenge, demanding innovative solutions for early detection and intervention. The integration of machine learning into oncology offers a promising avenue to address this issue, providing data-driven insights and personalized care. This retrospective study analyzed data from 396 patients who underwent surgical procedures for colon cancer (CC) between 2010 and 2021. Machine learning algorithms were employed to predict CC recurrence, with a focus on demographic, clinicopathological, and laboratory characteristics. A range of evaluation metrics, including AUC (Area Under the Receiver Operating Characteristic), accuracy, recall, precision, and F1 scores, assessed the performance of machine learning algorithms. Significant risk factors for CC recurrence were identified, including sex, carcinoembryonic antigen (CEA) levels, tumor location, depth, lymphatic and venous invasion, and lymph node involvement. The CatBoost Classifier demonstrated exceptional performance, achieving an AUC of 0.92 and an accuracy of 88% on the test dataset. Feature importance analysis highlighted the significance of CEA levels, albumin levels, N stage, weight, platelet count, height, neutrophil count, lymphocyte count, and gender in determining recurrence risk. The integration of machine learning into healthcare, exemplified by this study's findings, offers a pathway to personalized patient risk stratification and enhanced clinical decision-making. Early identification of individuals at risk of CC recurrence holds the potential for more effective therapeutic interventions and improved patient outcomes. Machine learning has the potential to revolutionize our approach to CC recurrence prediction, emphasizing the synergy between medical expertise and cutting-edge technology in the fight against cancer. This study represents a vital step toward precision medicine in CC management, showcasing the transformative power of data-driven insights in oncology.