Validation Set Research Articles

A quality assessment tool for focused abdominal sonography for trauma examinations using artificial intelligence.

Current tools to review focused abdominal sonography for trauma (FAST) images for quality have poorly defined grading criteria or are developed to grade the skills of the sonographer and not the examination. The purpose of this study is to establish a grading system with substantial agreement among coders, thereby enabling the development of an automated assessment tool for FAST examinations using artificial intelligence (AI). Five coders labeled a set of FAST clips. Each coder was responsible for a different subset of clips (10% of the clips were labeled in triplicate to evaluate intercoder reliability). The clips were labeled with a quality score from 1 (lowest quality) to 5 (highest quality). Clips of 3 or greater were considered passing. An AI training model was developed to score the quality of the FAST examination. The clips were split into a training set, a validation set, and a test set. The predicted scores were rounded to the nearest quality level to distinguish passing from failing clips. A total of 1,514 qualified clips (1,399 passing and 115 failing clips) were evaluated in the final data set. This final data set had a 94% agreement between pairs of coders on the pass/fail prediction, and the set had a Krippendorff α of 66%. The decision threshold can be tuned to achieve the desired tradeoff between precision and sensitivity. Without using the AI model, a reviewer would, on average, examine roughly 25 clips for every 1 failing clip identified. In contrast, using our model with a decision threshold of 0.015, a reviewer would examine roughly five clips for every one failing clip - a fivefold reduction in clips reviewed while still correctly identifying 85% of passing clips. Integration of AI holds significant promise in improving the accurate evaluation of FAST images while simultaneously alleviating the workload burden on expert physicians. Diagnostic Test/Criteria; Level II.

Comparing MLR and ANN models for school building electrical energy prediction in Osijek-Baranja County in Croatia

This paper presents a study conducted in Osijek-Baranja County, Croatia, to predict electrical energy consumption in school buildings. Data from the Energy Management Information System (EMIS) database for primary and secondary schools were analyzed using Multiple Linear Regression (MLR) and Artificial Neural Networks (ANN). The ANN model achieved a high R² of 0957 in the training set, with lower Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) than the MLR model, which had an R² of 0950. On the validation set, the ANN model maintained an R² of 0954 and showed slightly better performance with a lower Coefficient of Variation of RMSE (CVRMSE) of 19,79 %, compared to the MLR model’s CVRMSE of 20,50 %. These results indicate that the ANN model generally provides more accurate and reliable predictions for energy consumption in school buildings. However, both models provided a robust positive correlation between the predicted and actual values.

Effect of aberrant fructose metabolism following SARS-CoV-2 infection on colorectal cancer patients' poor prognosis.

Most COVID-19 patients have a positive prognosis, but patients with additional underlying diseases are more likely to have severe illness and increased fatality rates. Numerous studies indicate that cancer patients are more prone to contract SARS-CoV-2 and develop severe COVID-19 or even dying. In the recent transcriptome investigations, it is demonstrated that the fructose metabolism is altered in patients with SARS-CoV-2 infection. However, cancer cells can use fructose as an extra source of energy for growth and metastasis. Furthermore, enhanced living conditions have resulted in a notable rise in fructose consumption in individuals' daily dietary habits. We therefore hypothesize that the poor prognosis of cancer patients caused by SARS-CoV-2 may therefore be mediated through fructose metabolism. Using CRC cases from four distinct cohorts, we built and validated a predictive model based on SARS-CoV-2 producing fructose metabolic anomalies by coupling Cox univariate regression and lasso regression feature selection algorithms to identify hallmark genes in colorectal cancer. We also developed a composite prognostic nomogram to improve clinical practice by integrating the characteristics of aberrant fructose metabolism produced by this novel coronavirus with age and tumor stage. To obtain the genes with the greatest potential prognostic values, LASSO regression analysis was performed, In the TCGA training cohort, patients were randomly separated into training and validation sets in the ratio of 4: 1, and the best risk score value for each sample was acquired by lasso regression analysis for further analysis, and the fifteen genes CLEC4A, FDFT1, CTNNB1, GPI, PMM2, PTPRD, IL7, ALDH3B1, AASS, AOC3, SEPINE1, PFKFB1, FTCD, TIMP1 and GATM were finally selected. In order to validate the model's accuracy, ROC curve analysis was performed on an external dataset, and the results indicated that the model had a high predictive power for the prognosis prediction of patients. Our study provides a theoretical foundation for the future targeted regulation of fructose metabolism in colorectal cancer patients, while simultaneously optimizing dietary guidance and therapeutic care for colorectal cancer patients in the context of the COVID-19 pandemic.

Development and validation of a risk prediction model for older adults with social isolation in China

BackgroundOlder adults are vulnerable to social isolation due to declining physical and cognitive function, decreased interpersonal interactions, and reduced outdoor activities after retirement. This study aimed to develop and validate a predictive model to assess the risk of social isolation among older adults in China.MethodsUsing data from the 2011 China Health and Retirement Longitudinal Study (CHARLS). The study cohort was randomly divided into training and validation groups in a 70:30 ratio. We used least absolute shrinkage and selection operator (LASSO) regression analysis with tenfold cross-validation to identify optimal predictive factors and examined the correlates of social isolation using logistic regression. A nomogram was constructed for the predictive model, and its accuracy was assessed using calibration curves. The predictive performance of the model was assessed using area under the receiver operating characteristic (ROC) curve and decision curve analysis (DCA).ResultsFrom the 2011 CHARLS database, 4,747 older adults were included in the final analysis, of whom 1,654 (34.8%) experienced social isolation. Multifactorial logistic regression identified educational level, marital status, gender, physical activity, physical self -maintenance ability, and number of children as predictive factors for social isolation. The predictive model achieved an AUC of 0.739 (95%CI = 0.722–0.756) in the training set and 0.708 (95%CI = 0.681–0.735) in the validation set. The Hosmer-Lemeshow test yielded P values of 0.111 and 0.324, respectively (both P > 0.05), indicating significant agreement between the nomogram and observed outcomes. The nomogram showed excellent predictive ability according to ROC and DCA.ConclusionsThe predictive model developed to assess the risk of social isolation in the Chinese older adults shows promising utility for early screening and intervention by clinical healthcare professionals.

Risk factor analysis and establishment of a predictive model for complications of elderly advanced gastric cancer with Clavien-Dindo classification ≥ II grade

BackgroundThe occurrence of complications following radical gastrectomy for gastric cancer significantly impacts patients’ quality of life. Elderly patients are susceptible to postoperative complications. This study seeks to investigate the risk factors associated with Clavien-Dindo ≥IIgrade complications following radical gastrectomy for advanced gastric cancer in elderly patients, develop a nomogram risk prediction model, and validate its accuracy.MethodsRetrospective collection of clinical and pathological data was conducted on 442 elderly patients with advanced gastric cancer who underwent radical gastrectomy at Shaanxi Provincial People’s Hospital from January 2015 to December 2020. They were randomly divided into a training set (n = 310) and a validation set (n = 132) in a 7:3 ratio. The severity of postoperative complications was graded using the Clavien-Dindo classification system, resulting in two complication groups: Clavien-Dindo <II group (n = 229) and Clavien-Dindo ≥II group (n = 81). Univariate and multivariate logistic regression analyses were performed to identify independent risk factors affecting the occurrence of Clavien-Dindo ≥II grade complications, and a predictive model was established based on the results. The model was then validated using the validation set.ResultsAmong the 442 patients included in the study, 121 cases (27.38%) experienced postoperative complications, with 111 cases (25.11%) classified as Clavien-Dindo ≥II grade complications. Multivariable logistic analysis revealed that the Prognostic Nutritional Index (PNI), surgical duration, age, and history of Diabetes mellitus were independent risk factors for the occurrence of Clavien-Dindo ≥IIgrade complications in elderly patients with advanced gastric cancer after surgery (P < 0.05). The nomogram model constructed based on these factors demonstrated good discriminative ability, as indicated by the area under the Receiver Operating Characteristic (ROC) curve. Calibration plots showed that the predicted probability of gastric cancer lymph node metastasis using the nomogram model was well aligned with actual outcomes. Decision curve analysis indicated the clinical utility of the nomogram model across a wide range of thresholds, demonstrating its practicality and potential for clinical benefit.ConclusionThis study identified Prognostic Nutritional Index (PNI), surgical duration, age, and history of Diabetes mellitus as risk factors for the occurrence of Clavien-Dindo ≥ II grade complications in elderly patients with advanced gastric cancer after surgery. Based on these four risk factors, a nomogram risk prediction model was constructed. This model can be used to personalize the prediction of the risk of Clavien-Dindo ≥ II grade complications occurring after surgery in elderly patients with advanced gastric cancer.

A novel online calculator based on inflammation-related endotypes and clinical features to predict postoperative pulmonary infection in patients with cervical spinal cord injury

BackgroundPostoperative pulmonary infection (POI) of patients with cervical spinal cord injury (CSCI) is highly heterogeneous, while the potential endotypes and related risk factors remain unclear. MethodsA retrospective collection of 290 CSCI patients was conducted from January 2010 to July 2024 using 1:1 propensity score matching to compare POI (n = 145) and non-POI (n = 145) groups. We generated laboratory examination data from admission patients and identified endotypes using unsupervised consensus clustering and machine learning. CSCI patients were randomly assigned to the training set (n = 203) and internal validation set (n = 87). A separate cohort comprising 245 CSCI patients were used for external validation. Independent predictors for POI were identified using univariate and multivariate logistic regression. A nomogram and an online calculator were developed and validated, both internally and externally. ResultsTwo inflammation-related endotypes were identified: high inflammation endotype (endotype C1) and low inflammation endotype (endotype C2). Eight predictors for POI were identified (including age, operation duration, number of surgical segments, time between injury and surgery, preoperative steroid pulse, American Spinal Injury Association (ASIA) grade, smoking history, and inflammation-related endotype). A nomogram integrating the risk factors showed excellent discrimination in the training set (AUC, 0.976; 95% CI 0.956–0.996), internal validation set (AUC, 0.993; 95% CI 0.981–1.000), and external validation set (AUC, 0.799; 95%CI 0.744–0.854). Calibration curves demonstrated excellent fit, and decision curves highlighted its favorable clinical value. An online calculator (https://tjspine.shinyapps.io/dynnomapp/) was constructed to improve the convenience and efficiency of our prediction model. ConclusionsWe identified inflammation-related endotype and constructed a web-based calculator for predicting POI in patients with CSCI, exhibiting excellent clinical utility.

Application of Machine Learning Techniques for Countering Side-Channel Attacks in Cryptographic Systems

The use of machine learning algorithms in order to, not only, detect the adversarial intend behind side-channel attacks on cryptographic systems, but also to resist Differential Power Analysis (DPA) attacks. In particular, with the help of the DPA Challenge Dataset containing power traces of AES encryption operations, we propose a detailed step-by-step approach that includes data acquisition, preprocessing, feature extraction, and model assessment. The pre-processing includes noise reduction, normalization and segmental processing of the collected data for which basic statistical and frequency domain analysis can be used for extraction of relevant features. Support Vector Machines (SVMs) are then trained and tested in order to classify and in turn predict attack scenarios as per the subsequently derived features. As the outcome of the result pages show, the SVM model successfully classifies attack and non-attack traces at a rate of 88% on the validation set, which underlines the usage of machine learning to boost cryptographic security. Investigation of the feature relevance demonstrates that frequency-domain features, namely FFT coefficients are most impactful. The findings of this research prove that machine learning can be useful in preventing side-channel attacks apart from providing valuable information on enhancing the understanding of different defenses in cryptographic systems as well as future development of this domain.

Explainable machine learning model for predicting paratracheal lymph node metastasis in cN0 papillary thyroid cancer

Prophylactic dissection of paratracheal lymph nodes in clinically lymph node-negative (cN0) papillary thyroid carcinoma (PTC) remains controversial. This study aims to integrate preoperative and intraoperative variables to compare traditional nomograms and machine learning (ML) models, developing and validating an interpretable predictive model for paratracheal lymph node metastasis (PLNM) in cN0 PTC patients. We retrospectively selected 3213 PTC patients treated at the First Affiliated Hospital of Chongqing Medical University from 2016 to 2020. They were randomly divided into the training and test datasets with a 7:3 ratio. The 533 PTC patients treated at the Guangyuan Central Hospital from 2019 to 2022 were used as an external test sets. We developed and validated nine ML models using 10-fold cross-validation and grid search for hyperparameter tuning. The predictive performance was evaluated using ROC curves, decision curve analysis (DCA), calibration curves, and precision-recall curves. The best model was compared to a traditional logistic regression-based nomogram. The XGBoost model achieved AUC values of 0.935, 0.857, and 0.775 in the training, validation, and test sets, respectively, significantly outperforming the traditional nomogram model with AUCs of 0.85, 0.844, and 0.769, respectively. SHapley Additive exPlanations (SHAP)-based visualization identified the top 10 predictive features of the XGBoost model, and a web-based calculator was created based on these features. ML is a reliable tool for predicting PLNM in cN0 PTC patients. The SHAP method provides valuable insights into the XGBoost model, and the resultant web-based calculator is a clinically useful tool to assist in the surgical planning for paratracheal lymph node dissection.

Predictive nomogram of the clinical outcomes of colorectal cancer based on methylated SEPT9 and intratumoral IL-10+ Tregs infiltration

BackgroundGene methylation and the immune-related tumor microenvironment (TME) are highly correlated in tumor progression and therapeutic efficacy. Although both of them can be used to predict the clinical outcomes of colorectal cancer (CRC) patients, their predictive value is still unsatisfactory. Whether a combination risk model comprising these two prediction parameters performs better predictive effectiveness than independent factor is still unclear. Methylated Septin9 (mSEPT9) is an early diagnosis biomarker of CRC, in this study, we aimed to investigate mSEPT9-related biomarkers of immunosuppressive TME and identify the value of the combination risk model in predicting the clinical outcomes of CRC.MethodsImmunofluorescence staining was performed to clarify the correlation between intratumoral IL-10+ Treg infiltration and mSEPT9 in peripheral blood. Survival time, response to 5-fluorouracil (5-FU)-based chemotherapy and PD-1 blockade, and the probability of recurrence or metastasis were analyzed in study (197 CRC samples) and validation (195 CRC samples) sets to evaluate the efficacy of combination risk model. Potential mechanisms were explored by mRNA sequencing.ResultsHypermethylated SEPT9 in the peripheral blood of patients with CRC (stage I-III, and stage IV with resectable M1) before radical resection was positively correlated with high intratumoral IL-10+ Treg infiltration. The high-risk model revealed poor overall survival and progression-free survival, inferior therapeutic response to 5-FU-based chemotherapy and PD-1 blockade, and high probability of recurrence or metastasis. The underlying mechanisms may be associated with the increase in mSEPT9 and mediation of IL-10 via methionine metabolic reprogramming-induced infiltration of IL-10+ Tregs in the TME, which promotes tumor progression and resistance to 5-FU-based chemotherapy and PD-1 blockade.ConclusionsThe combination risk model of peripheral mSETP9 and intratumoral IL-10+ Treg infiltration in CRC can effectively predict prognosis, responsiveness to 5-FU-based chemotherapy and PD-1 blockade, and the probability of recurrence or metastasis. Therefore, this model can be used for precision treatment of CRC.

Development and external validation of clinical predictive model for stress urinary incontinence in Chinese women : a multicenter retrospective study

BackgroundStress urinary incontinence (SUI), the prevalent form of urinary incontinence, significantly impairs women’s quality of life. This study aims to create a visual nomogram to estimate the risk of SUI within one year postpartum for early intervention in high-risk Chinese women.MethodsWe recruited 1,531 postpartum women who gave birth at two hospitals in Kunshan City from 2021 to 2022. Delivery details were meticulously extracted from the hospitals’ medical records system, while one-year postpartum follow-ups were conducted via phone surveys specifically designed to ascertain SUI status. Utilizing data from one hospital as the training set, logistic regression analysis was performed to pinpoint significant factors and subsequently construct the nomogram. To ensure robustness, an independent dataset sourced from the second hospital served as the external validation cohort. The model’s performance was rigorously evaluated using calibration plots, ROC curves, AUC values, and DCA curves.ResultsThe study population was 1,125 women. The SUI incidence within one year postpartum was 26% (293/1125). According to the regression analysis, height, pre-pregnancy BMI, method of induction, mode of delivery, perineal condition, neonatal weight, SUI during pregnancy, and SUI during the first pregnancy were incorporated into the nomogram. The AUC of the nomogram was 0.829 (95% CI 0.790–0.867), and the external validation set was 0.746 (95% CI 0.689–0.804). Subgroup analysis based on parity showed good discrimination. The calibration curve indicated concordance. The DCA curve showed a significant net benefit.ConclusionDrawing from real-world data, we have successfully developed an SUI predictive model tailored for postpartum Chinese women. Upon successful external validation, this model holds immense potential as an effective screening tool for SUI, enabling timely interventions and ultimately may improve women’s quality of life.

Noninvasive multi-cancer detection using blood-based cell-free microRNAs.

Patients diagnosed with early-stage cancers have a substantially higher chance of survival than those with late-stage diseases. However, the option for early cancer screening is limited, with most cancer types lacking an effective screening tool. Here we report a miRNA-based blood test for multi-cancer early detection based on examination of serum microRNA microarray data from cancer patients and controls. First, a large multi-cancer training set that included 1,408 patients across 7 cancer types and 1,408 age- and gender-matched non-cancer controls was used to develop a 4-microRNA diagnostic model using 10-fold cross-validation. In three independent validation sets comprising a total of 4,875 cancer patients across 13 cancer types and 3,722 non-cancer participants, the 4-microRNA model achieved greater than 90% sensitivity for 9 cancer types (lung, biliary tract, bladder, colorectal, esophageal, gastric, glioma, pancreatic, and prostate cancers) and 75-84% sensitivity for 3 cancer types (sarcoma, liver, and ovarian cancer), while maintaining greater than 99% specificity. The sensitivity remained to be > 99% for patients with stage 1 lung cancer. Our study provided novel evidence to support the development of an inexpensive and accurate miRNA-based blood test for multi-cancer early detection.

Development of an Algorithm for Semantic Segmentation of Earth Remote Sensing Data to Determine Phytoplankton Populations

Introduction. Computer vision is widely used for semantic segmentation of Earth remote sensing (ERS) data. The method allows monitoring ecosystems, including aquatic ones. Algorithms that maintain the quality of semantic segmentation of ERS images are in demand, specifically, to identify areas with phytoplankton, where water blooms— the cause of suffocation — are possible. The objective of the study is to create an algorithm that processes satellite data as input information for the formation and checking of mathematical models of hydrodynamics, which are used to monitor the state of water bodies. Various algorithms for semantic segmentation are described in the literature. New research focuses on enhancing the reliability of recognition — often using neural networks. This approach is modified in the presented work. To develop the direction, a new set of information from open sources and synthetic data are proposed. They are aimed at improving the generalization ability of the model. For the first time, the contour area of the phytoplankton population is compared to the database — and thus the boundary conditions are formed for the implementation of mathematical models and the construction of boundary-adaptive grids.Materials and Methods. The set of remote sensing images was supplemented with the author's augmentation algorithm in Python. Computer vision segmented areas of phytoplankton populations in the images. The U-Net convolutional neural network (CNN) was trained on the basis of NVIDIA Tesla T4 computing accelerators.Results. To automate the detection of phytoplankton distribution areas, a computer vision algorithm based on the U-Net CNN was developed. The model was evaluated by the calculated values of the main quality metrics related to segmentation tasks. The following metric values were obtained: Precision = 0.89, Recall = 0.88, F1 = 0.87, Dice = 0.87, and IoU = 0.79. Graphical visualization of the results of CNN learning on the training and validation sets showed good quality of model learning. This is evidenced by small changes in the loss function at the end of training. The segmentation performed by the model turned out to be close to manual marking, which indicated the high quality of the proposed solution. The area of the segmented region of the phytoplankton population was calculated by the area of one pixel. The result obtained for the original image was 51202.5 (based on information about the number of pixels related to the bloom of blue-green algae). The corresponding result of the modeling was 51312.Discussion and Conclusion. The study expands theoretical and practical knowledge on the use of convolutional neural networks for semantic segmentation of space imagery data. Given the results of the work, it is possible to assess the potential for automating the process of semantic segmentation of remote sensing data to determine the boundaries of phytoplankton populations using artificial intelligence. The use of the proposed computer vision model to obtain contours of water bloom due to phytoplankton will provide for the creation of databases — the basis for environmental monitoring of water resources and predictive modeling of hydrobiological processes.

Water Resources’ AI–ML Data Uncertainty Risk and Mitigation Using Data Assimilation

Artificial intelligence (AI), including machine learning (ML) and deep learning (DL), learns by training and is restricted by the amount and quality of training data. Training involves a tradeoff between prediction bias and variance controlled by model complexity. Increased model complexity decreases prediction bias, increases variance, and increases overfitting possibilities. Overfitting is a significantly smaller training prediction error relative to the trained model prediction error for an independent validation set. Uncertain data generate risks for AI–ML because they increase overfitting and limit generalization ability. Specious confidence in predictions from overfit models with limited generalization ability, leading to misguided water resource management, is the uncertainty-related negative consequence. Improved data is the way to improve AI–ML models. With uncertain water resource data sets, like stream discharge, there is no quick way to generate improved data. Data assimilation (DA) provides mitigation for uncertainty risks, describes data- and model-related uncertainty, and propagates uncertainty to results using observation error models. A DA-derived mitigation example is provided using a common-sense baseline, derived from an observation error model, for the confirmation of generalization ability and a threshold identifying overfitting. AI–ML models can also be incorporated into DA to provide additional observations for assimilation or as a forward model for prediction and inverse-style calibration or training. The mitigation of uncertain data risks using DA involves a modified bias–variance tradeoff that focuses on increasing solution variability at the expense of increased model bias. Increased variability portrays data and model uncertainty. Uncertainty propagation produces an ensemble of models and a range of predictions.

Enhancing skin aging parameter assessment in clinical trials: AI-Driven analysis of ultrasound images

BackgroundHigh-frequency ultrasound (HFUS) imaging offers valuable insights into skin aging parameters in clinical trials. However, manual measurements are time-consuming and operator-dependent, necessitating the exploration of automated approaches. ObjectiveThis study aimed to develop, train, and validate an artificial intelligence model for automated HFUS image parameter acquisition in dermatological assessments. A secondary goal was its application in evaluating data from an anti-aging skin therapy clinical trial. MethodsSupervised machine learning algorithms were used to predict target variables from ultrasound images. The best models were selected through independent validation. A training dataset of 144 ultrasound skin images and a validation set of 40 images were employed. A clinical trial involved 34 female subjects with aging skin signs, randomized into two groups: (1) cosmetic formulation plus therapeutic unfocused ultrasound (TUS) and (2) placebo formulation plus TUS. HFUS imaging was conducted before and after a 4-week and 8-week study period. ResultsOur algorithms successfully predicted echogenicity and thickness variables, showing competitive performance. In the clinical trial, the treatment group exhibited a significant improvement in skin echogenicity, while no significant changes were observed in the placebo group. ConclusionThis study demonstrates the feasibility of automating the prediction of bioimage variables for anti-aging strategy development. Mathematical algorithms reduced HFUS image evaluations to minutes, expediting clinical trial progress.

Development and validation of a nomogram for predicting COPD: A nationwide population-based study in South Korea.

Chronic obstructive pulmonary disease (COPD) remains a significant global health burden exacerbated by tobacco smoking, occupational exposure, and air pollution. COPD is one of the top 3 causes of death worldwide. In South Korea, the COPD burden is expected to increase due to ongoing exposure to risk factors and the aging population. COPD is extensively underdiagnosed or underestimated, owing to a lack of public awareness. This study aimed to develop and validate a nomogram for COPD by using national data to promote early diagnosis and intervention. This study drew on a dataset from the 7th Korea National Health and Nutrition Examination Survey from 2016 to 2018, including 10,819 subjects aged 40 years or older with spirometry results. Influence of demographic, socioeconomic, and health-related factors on the incidence. Multivariable logistic regression was used to identify the significant predictors of the nomogram. The nomogram was validated using receiver operating characteristic curves, calibration plots, and concordance index (C-index). Internal validation was performed by bootstrapping. In the final analysis, 1059 (14.0%) participants had COPD. Key risk factors associated with increased COPD risk included being male, aged 70 and older, lower educational level, living in a rural area, current smoking status, underweight, and history of tuberculosis and asthma. The area under the curve (AUC) of the model was 0.822 (95% CI: 0.810-0.832), indicating that the nomogram has a high ability to identify COPD. The nomogram demonstrated solid predictive performance, as confirmed by calibration plots with a C-index (of 0.822) for the validation set with 1000 bootstrap samples. In conclusion, we developed a tool for the early detection of COPD with good properties in primary care settings, without spirometry. Appropriate and early diagnosis of COPD can have a crucial impact on public health.

Machine learning-based prediction of gastroparesis risk following complete mesocolic excision

BackgroundGastroparesis is a major complication following complete mesocolic excision (CME) and significantly impacts patient outcomes. This study aimed to create a machine learning model to pinpoint key risk factors before, during, and after surgery, effectively predicting the risk of gastroparesis after CME.MethodsThe study involved 1146 patients with colon cancer, out of which 95 developed gastroparesis. Data were collected on 34 variables, including demographics, chronic conditions, pre-surgery test results, types of surgery, and intraoperative details. Four machine learning techniques were employed: extreme gradient boosting (XGBoost), random forest (RF), support vector machine (SVM), and k-nearest neighbor (KNN). The evaluation involved k-fold cross-validation, receiver operating characteristic (ROC) analysis, calibration curves, decision curve analysis (DCA), and external validation.ResultsXGBoost excelled in its performance for predictive models. ROC analysis showed high accuracy for XGBoost, with area under the curve (AUC) scores of 0.976 for the training set and 0.906 for the validation set. K-fold cross-validation confirmed the model's stability, and calibration curves indicated high predictive accuracy. Additionally, DCA highlighted XGBoost’s superior patient benefits for intervention treatments. An AUC of 0.77 in external validation demonstrated XGBoost's strong generalization ability.ConclusionThe XGBoost-fueled predictive model for post-surgery colon cancer patients proved highly effective. It underlined gastroparesis as a significant post-operative issue, associated with advanced age, prolonged surgeries, extensive intraoperative blood loss, surgical techniques, low serum protein levels, anemia, diabetes, and hypothyroidism.

A nomogram for enhanced risk stratification for predicting cervical lymph node metastasis in papillary thyroid carcinoma patients

BackgroundCervical lymph node metastasis (CLNM) significantly impacts the prognosis of papillary thyroid carcinoma (PTC) patients. Accurate CLNM prediction is crucial for surgical planning and patient outcomes. This study aimed to develop and validate a nomogram-based risk stratification system to predict CLNM in PTC patients.MethodsThis retrospective study included 1069 patients from Zhongshan Hospital and 253 from the Qingpu Branch of Zhongshan Hospital. Preoperative ultrasound (US) data and various nodule characteristics were documented. Patients underwent lobectomy with central lymph node dissection and lateral dissection if suspicious. Multivariate logistic regression, least absolute shrinkage and selection operator (LASSO) regression, and the random forest algorithm were used to identify CLNM risk factors. A nomogram was constructed and validated internally and externally. Model performance was assessed via receiver operating characteristic (ROC) curves, calibration plots, DeLong’s test, decision curve analysis (DCA), and the clinical impact curve (CIC).ResultsSix independent CLNM risk factors were identified: age, sex, tumor size, calcification, internal vascularity, and US-reported CLNM status. The model's area under the curve (AUC) was 0.77 for both the training and the external validation sets. Calibration plots and Hosmer‒Lemeshow (HL) tests showed good calibration. The optimal cutoff value was 0.57, with a sensitivity of 58.02% and a specificity of 83.43%. Risk stratification on the basis of the nomogram categorized patients into low-, intermediate-, and high-risk groups, effectively differentiating the likelihood of CLNM, and an online calculator was created for clinical use.ConclusionThe nomogram accurately predicts CLNM risk in PTC patients, aiding personalized surgical decisions and improving patient management.

Multi-sequence MRI-based radiomics model to preoperatively predict the WHO/ISUP grade of clear Cell Renal Cell Carcinoma: a two-center study

ObjectivesTo develop radiomics models based on multi-sequence MRI from two centers for the preoperative prediction of the WHO/ISUP grade of Clear Cell Renal Cell Carcinoma (ccRCC).MethodsThis retrospective study included 334 ccRCC patients from two centers. Significant clinical factors were identified through univariate and multivariate analyses. MRI sequences included Dynamic contrast-enhanced MRI, axial fat-suppressed T2-weighted imaging, diffusion-weighted imaging, and in-phase/out-of-phase images. Feature selection methods and logistic regression (LR) were used to construct clinical and radiomics models, and a combined model was developed using the Rad-score and significant clinical factors. Additionally, seven classifiers were used to construct the combined model and different folds LR was used to construct the combined model to evaluate its performance. Models were evaluated using receiver operating characteristic (ROC) curves, area under the curve (AUC), and decision curve analysis (DCA). The Delong test compared ROC performance, with p < 0.050 considered significant.ResultsMultivariate analysis identified intra-tumoral vessels as an independent predictor of high-grade ccRCC. In the external validation set, the radiomics model (AUC = 0.834) outperformed the clinical model (AUC = 0.762), with the combined model achieving the highest AUC (0.855) and significantly outperforming the clinical model (p = 0.003). DCA showed that the combined model had a higher net benefit within the 0.04–0.54 risk threshold range than clinical model. Additionally, the combined model constructed using logistic regression has a higher priority compared to other classifiers. Additionally, 10-fold cross-validation with LR for the combined model showed consistent AUC values (0.849–0.856) across different folds.ConclusionThe radiomics models based on multi-sequence MRI might be a noninvasive and effective tool, demonstrating good efficacy in preoperatively predicting the WHO/ISUP grade of ccRCC.

Ultrasound-based Radiomics Analysis for Assessing Risk Factors Associated With Early Recurrence Following Surgical Resection of Hepatocellular Carcinoma

ObjectiveThe aim of this study was to explore the value of ultrasound-based radiomics analysis for early recurrence (ER) after surgical resection of hepatocellular carcinoma (HCC). MethodsThis retrospective study included 127 patients who underwent primary surgical resection for HCC between October 2019 and November 2021. The patients were subsequently divided into training and validation sets (7:3 ratio). All patients received preoperative routine ultrasound and contrast-enhanced ultrasound (CEUS), with postoperative pathological confirmation of HCC. Radiomics features were extracted from maximum section of two-dimensional ultrasound image. The least absolute shrinkage and selection operation (LASSO) logistic regression algorithm with 10-fold cross-validation was employed to establish ultrasonic radiomics features. Logistic regression modeling was employed to build models based on clinical and ultrasonic features (model 1, clinical-ultrasonic model), radiomics signature (model 2, ultrasonic radiomics model), and the combination (model 3, clinical-ultrasonic-radiomics model). Then a nomogram model was established to predict the risk of ER, and the application value of nomogram through internal verification was evaluated. ResultsModel 3 showed optimal diagnostic performance in both training set (AUC=0.907) and validation set (AUC=0.925), followed by the model 1 in training set (AUC=0.846) and validation set (AUC=0.855), both above two models performed better than model 2 in training set (AUC=0.751) and validation set (AUC=0.702) (p<0.05). In training set and validation set of model 3, the sensitivity were 83.3%, 77.8%, the specificity ware 95.8%, 100.0% and the C-index were 0.791, 0.778. ConclusionThe preoperative clinical-ultrasonic-radiomics model is anticipated to be a reliable tool for predicting the ER of surgical resection of HCC.

Sports-Related Genomic Predictors Are Associated with Athlete Status in Chinese Sprint/Power Athletes.

Objectives: The aim of this study was to assess the relationship between variant loci significantly associated with sports-related traits in the GWAS Catalog database and sprint/power athlete status, as well as to explore the polygenic profile of elite athletes. Methods: Next-generation sequencing and microarray technology were used to genotype samples from 211 elite athletes who had achieved success in national or international competitions in power-based sports and from 522 non-athletes, who were healthy university students with no history of professional sports training. Variant loci collected from databases were extracted after imputation. Subsequently, 80% of the samples were randomly selected as the training set, and the remaining 20% as the validation set. Results: Association analysis of variant loci was conducted in the training set, and individual Total Genotype Score (TGS) were calculated using genotype dosage and lnOR, followed by the establishment of a logistic model, with predictive performance evaluated in the validation set. Association analysis was performed on 2075 variant loci, and after removing linked loci (r2 > 0.2), 118 Tag SNPs (p ≤ 0.05) were identified. A logistic model built using 30 Tag SNPs (p ≤ 0.01) showed better performance in the validation set (AUC = 0.707). Conclusions: Our study identified 30 new genetic molecular markers and demonstrated that elite sprint/power athlete status is polygenic.