External Validation Dataset Research Articles

Abstract 4138334: Identification and Immunological Characterization of Arrhythmia-Related Molecular Clusters in Ischemic Cardiomyopathy

Ischemic cardiomyopathy, a severe cardiac condition resulting from prolonged myocardial ischemia, is characterized by ventricular dilation, dysfunction, and an increased risk of life-threatening arrhythmias. Arrhythmia, a common complication in ischemic cardiomyopathy, is associated with poor clinical outcomes. Recent studies have indicated that the molecular heterogeneity of ischemic cardiomyopathy may contribute to the development of arrhythmias. Therefore, our study aimed to explore arrhythmia-related molecular clusters in ischemic cardiomyopathy and establish a prediction model for risk stratification. Based on gene expression profiles of ischemic cardiomyopathy patients, we analyzed the dysregulation of arrhythmia-associated genes and the associated immune characteristics (Figure 1A-C). Using unsupervised clustering analysis, we identified distinct arrhythmia-related molecular clusters in ischemic cardiomyopathy patients (Figure 1D-F). Cluster-specific differentially expressed genes were further identified using weighted gene co-expression network analysis (WGCNA). Functional enrichment analysis revealed significant differences in biological processes and pathways between the clusters (Figure 1G-H). Subsequently, we developed a prediction model using a random forest algorithm, which showed superior performance in distinguishing patients with different arrhythmia risks (Figure 1I-J). This forest model demonstrated excellent performance in one external validation datasets with AUC values of 0.891. In conclusion, our study provides novel insights into the molecular basis of arrhythmias in ischemic cardiomyopathy and establishes a prediction model with potential clinical utility for risk stratification and targeted therapies. Further investigation into the specific mechanisms underlying the identified molecular clusters may reveal new therapeutic targets for the management of arrhythmias in ischemic cardiomyopathy patients.

Explainable machine learning for early prediction of sepsis in traumatic brain injury: A discovery and validation study

Background People with traumatic brain injury (TBI) are at high risk for infection and sepsis. The aim of the study was to develop and validate an explainable machine learning(ML) model based on clinical features for early prediction of the risk of sepsis in TBI patients. Methods We enrolled all patients with TBI in the Medical Information Mart for Intensive Care IV database from 2008 to 2019. All patients were randomly divided into a training set (70%) and a test set (30%). The univariate and multivariate regression analyses were used for feature selection. Six ML methods were applied to develop the model. The predictive performance of different models were determined based on the area under the curve (AUC) and calibration curves in the test cohort. In addition, we selected the eICU Collaborative Research Database version 1.2 as the external validation dataset. Finally, we used the Shapley additive interpretation to account for the effects of features attributed to the model. Results Of the 1555 patients enrolled in the final cohort, 834 (53.6%) patients developed sepsis after TBI. Six variables were associated with concomitant sepsis and were used to develop ML models. Of the 6 models constructed, the Extreme Gradient Boosting (XGB) model achieved the best performance with an AUC of 0.807 and an accuracy of 74.5% in the internal validation cohort, and an AUC of 0.762 for the external validation. Feature importance analysis revealed that use mechanical ventilation, SAPSII score, use intravenous pressors, blood transfusion on admission, history of diabetes, and presence of post-stroke sequelae were the top six most influential features of the XGB model. Conclusion As shown in the study, the ML model could be used to predict the occurrence of sepsis in patients with TBI in the intensive care unit.

Early Modeling of the Upcoming Landsat Next Constellation for Soybean Yield Prediction Under Varying Levels of Water Availability

The upcoming Landsat Next will provide more frequent land surface observations at higher spatial and spectral resolutions that will greatly benefit the agricultural sector. Early modeling of the upcoming Landsat Next products for soybean yield prediction is essential for long-term satellite monitoring strategies. In this context, this article evaluates the contribution of Landsat Next’s improved spectral resolution for soybean yield prediction under varying levels of water availability. Ground-based hyperspectral data collected over five cropping seasons at the Brazilian Agricultural Research Corporation were resampled to Landsat Next spectral resolution. The spectral dataset (n = 384) was divided into calibration and external validation datasets and investigated using three strategies for soybean yield prediction: (1) using the reflectance from each spectral band; (2) using existing and new vegetation indices developed based on three general equations: Normalized Difference Vegetation Index (NDVI-like), Band Ratio Vegetation Index (RVI-like), and Band Difference Vegetation Index (DVI-like), replacing the traditional spectral bands by all possible combinations between two bands for index calculation; and (3) using a partial least squares regression (PLSR) model composed of all Landsat Next spectral bands, in comparison to PLSR models using Landsat OLI and Sentienel-2 MSI bands. The results show the distribution of the new spectral bands over the most prominent changes in leaf reflectance due to water deficit, particularly in the visible and shortwave infrared spectrum. (1) Band 18 (centered at 1610 nm) had the highest correlation with yield (R2 = 0.34). (2) A new vegetation index, called Normalized Difference Shortwave Vegetation Index (NDSWVI), is proposed and calculated from bands 19 and 20 (centered at 2028 and 2108 nm). NDSWVI showed the best performance (R2 = 0.37) compared to traditional existing and new vegetation indices. (3) The PLSR model gave the best results (R2 = 0.65), outperforming the Landsat OLI and Sentinel-2 MSI sensors. The improved spectral resolution of Landsat Next is expected to contribute to improved crop monitoring, especially for soybean crops in Brazil, increasing the sustainability of the production systems and strengthening food security in Brazil and globally.

Compare three deep learning-based artificial intelligence models for classification of calcified lumbar disc herniation: a multicenter diagnostic study

ObjectiveTo develop and validate an artificial intelligence diagnostic model for identifying calcified lumbar disc herniation based on lateral lumbar magnetic resonance imaging(MRI).MethodsDuring the period from January 2019 to March 2024, patients meeting the inclusion criteria were collected. All patients had undergone both lumbar spine MRI and computed tomography(CT) examinations, with regions of interest (ROI) clearly marked on the lumbar sagittal MRI images. The participants were then divided into separate sets for training, testing, and external validation. Ultimately, we developed a deep learning model using the ResNet-34 algorithm model and evaluated its diagnostic efficacy.ResultsA total of 1,224 eligible patients were included in this study, consisting of 610 males and 614 females, with an average age of 53.34 ± 10.61 years. Notably, the test datasets displayed an impressive classification accuracy rate of 91.67%, whereas the external validation datasets achieved a classification accuracy rate of 88.76%. Among the test datasets, the ResNet34 model outperformed other models, yielding the highest area under the curve (AUC) of 0.96 (95% CI: 0.93, 0.99). Additionally, the ResNet34 model also exhibited superior performance in the external validation datasets, exhibiting an AUC of 0.88 (95% CI: 0.80, 0.93).ConclusionIn this study, we established a deep learning model with excellent performance in identifying calcified intervertebral discs, thereby offering a valuable and efficient diagnostic tool for clinical surgeons.

Artificial intelligence tools trained on human-labeled data reflect human biases: a case study in a large clinical consecutive knee osteoarthritis cohort

Humans have been shown to have biases when reading medical images, raising questions about whether humans are uniform in their disease gradings. Artificial intelligence (AI) tools trained on human-labeled data may have inherent human non-uniformity. In this study, we used a radiographic knee osteoarthritis external validation dataset of 50 patients and a six-year retrospective consecutive clinical cohort of 8,273 patients. An FDA-approved and CE-marked AI tool was tested for potential non-uniformity in Kellgren-Lawrence grades between the right and left sides of the images. We flipped the images horizontally so that a left knee looked like a right knee and vice versa. According to human review, the AI tool showed non-uniformity with 20–22% disagreements on the external validation dataset and 13.6% on the cohort. However, we found no evidence of a significant difference in the accuracy compared to senior radiologists on the external validation dataset, or age bias or sex bias on the cohort. AI non-uniformity can boost the evaluated performance against humans, but image areas with inferior performance should be investigated.

Using CT images to assist the segmentation of MR images via generalization: Segmentation of the renal parenchyma of renal carcinoma patients.

Developing deep learning models for segmenting medical images in multiple modalities with less data and annotation is an attractive and challenging task, which was previously discussed as being accomplished by complex external frameworks for bridging the gap between different modalities. Exploring the generalization ability of networks in medical images in different modalities could provide more simple and accessible methods, yet comprehensive testing could still be needed. To explore the feasibility and robustness of using computed tomography (CT) images to assist the segmentation of magnetic resonance (MR) images via the generalization, in the segmentation of renal parenchyma of renal cell carcinoma (RCC) patients. Nephrographic CT images and fat-suppressed T2-weighted (fs-T2W) images were retrospectively collected. The pure CT dataset included 116 CT images. Additionally, 240 MR images were randomly divided into subsets A and B. From subset A, three training datasets were constructed, each containing 40, 80, and 120 images, respectively. Similarly, three datasets were constructed from subset B. Subsequently, datasets with mixed modality were created by combining these pure MR datasets with the 116 CT images. The 3D-UNET models for segmenting the renal parenchyma in two steps were trained using these 13 datasets: segmenting kidneys and then the renal parenchyma. These models were evaluated in internal MR (n=120), CT (n=65) validation datasets, and an external validation dataset of CT (n=79), using the mean of the dice similarity coefficient (DSC). To demonstrate the robustness of generalization ability over different proportions of modalities, we compared the models trained with mixed modality in three different proportions and pure MR, using repeated measures analysis of variance (RM-ANOVA). We developed a renal parenchyma volume quantification tool by the trained models. The mean differences and Pearson correlation coefficients between the model segmentation volume and the ground truth segmentation volume were calculated for its evaluation. The mean DSCs of models trained with 116 data in CT in the validation of MR were 0.826, 0.842, and 0.953, respectively, for the predictions of kidney segmentation model on whole image, renal parenchymal segmentation model on kidneys with RCC and without RCC. For all models trained with mixed modality, the means of DSC were above 0.9, in all validations of CT and MR. According to the results of the comparison between models trained with mixed modality and pure MR, the means of DSC of the former were significantly greater or equal to the latter, at all three different proportions of modalities. The differences of volumes were all significantly lower than one-third of the volumetric quantification error of a previous method, and the Pearson correlation coefficients of volumes were all above 0.96 on kidneys with and without RCC of three validations. CT images could be used to assist the segmentation of MR images via the generalization, with or without the supervision of MR data. This ability showed acceptable robustness. A tool for accurately measuring renal parenchymal volume on CT and MR images was established.

Artificial intelligence-enabled electrocardiogram for mortality and cardiovascular risk estimation: a model development and validation study

Artificial intelligence (AI)-enabled electrocardiography (ECG) can be used to predict risk of future disease and mortality but has not yet been adopted into clinical practice. Existing model predictions do not have actionability at an individual patient level, explainability, or biological plausibi. We sought to address these limitations of previous AI-ECG approaches by developing the AI-ECG risk estimator (AIRE) platform. The AIRE platform was developed in a secondary care dataset (Beth Israel Deaconess Medical Center [BIDMC]) of 1 163 401 ECGs from 189 539 patients with deep learning and a discrete-time survival model to create a patient-specific survival curve with a single ECG. Therefore, AIRE predicts not only risk of mortality, but also time-to-mortality. AIRE was validated in five diverse, transnational cohorts from the USA, Brazil, and the UK (UK Biobank [UKB]), including volunteers, primary care patients, and secondary care patients. AIRE accurately predicts risk of all-cause mortality (BIDMC C-index 0·775, 95% CI 0·773-0·776; C-indices on external validation datasets 0·638-0·773), future ventricular arrhythmia (BIDMC C-index 0·760, 95% CI 0·756-0·763; UKB C-index 0·719, 95% CI 0·635-0·803), future atherosclerotic cardiovascular disease (0·696, 0·694-0·698; 0·643, 0·624-0·662), and future heart failure (0·787, 0·785-0·789; 0·768, 0·733-0·802). Through phenome-wide and genome-wide association studies, we identified candidate biological pathways for the prediction of increased risk, including changes in cardiac structure and function, and genes associated with cardiac structure, biological ageing, and metabolic syndrome. AIRE is an actionable, explainable, and biologically plausible AI-ECG risk estimation platform that has the potential for use worldwide across a wide range of clinical contexts for short-term and long-term risk estimation. British Heart Foundation, National Institute for Health and Care Research, and Medical Research Council.

Development of a risk score for predicting one-year mortality in patients with atrial fibrillation using XGBoost-assisted feature selection.

There are no tools specifically designed to assess mortality risk in patients with atrial fibrillation (AF). This study aimed to utilize machine learning methods to identify pertinent variables and develop an easily applicable prognostic score to predict 1-year mortality in AF patients. This study, based on the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database, focused on patients aged 18 years and older with AF. A critical care database from China was the external validation set. The importance of variables from XGBoost guided the development of a logistic model, forming the basis for an AF scoring model. Records of of 26 365 AF patients were obtained from the MIMIC-IV database. The external validation dataset included 231 AF patients. The CRAMB score (Charlson comorbidity index, readmission, age, metastatic solid tumor, and maximum blood urea nitrogen concentration) outperformed the CCI and CHA2DS2-VASc scores, demonstrating superior predictive value for 1-year mortality. In the test set, the area under the receiver operating characteristic (AUC) for the CRAMB score was 0.765 (95% confidence interval [CI], 0.753-0.776), while in the external validation set, it was 0.582 (95% CI, 0.502-0.657). The simplicity of the CRAMB score makes it user-friendly, allowing for coverage of a broader and more heterogeneous AF population.

Superior performance of a center-point AI model over VFLDNet in automated cobb angle estimation for scoliosis assessment.

Aims to establish the superiority of our proposed model over the state-of-the-art vertebra-focused landmark detection network (VFLDNet) in automating Cobb angle estimation from spinal radiographs. Utilizing a private dataset for external validation, we compared the performance of our center-point detection-based vertebra localization and tilt estimation network (VLTENet) with the key-point detection-based VFLDNet. Both models' Cobb angle predictions were rigorously evaluated against manual consensus score using metrics such as mean absolute error (MAE), correlation coefficient, intraclass correlation coefficient (ICC), Fleiss' kappa, Bland-Altman analysis, and classification metrics [sensitivity (SN), specificity, accuracy] focusing on major curve estimation and scoliosis severity classification. A retrospective analysis of 118 cases with 342 Cobb angle measurements revealed that our model achieved a MAE of 2.15° for total Cobb angles and 1.89° for the major curve, significantly outperforming VFLDNet's MAE of 2.80°and 2.57°, respectively. Both models demonstrated robust correlation and ICC, but our model excelled in classification consistency, particularly in predicting major curve magnitude (ours: kappa = 0.83; VFLDNet: kappa = 0.67). In subgroup analyses by scoliosis severity, our model consistently surpassed VFLDNet, displaying superior mean (SD) differences, narrower limits of agreement, and higher SN, specificity, and accuracy, most notably in moderate (ours: SN = 86.84%; VFLDNet: SN = 83.16%) to severe (ours: SN = 92.86%; VFLDNet: SN = 85.71%) scoliosis. Our model emerges as the superior choice for automated Cobb angle estimation, particularly in assessing major curve and moderate to severe scoliosis, underscoring its potential to revolutionize clinical workflows and enhance patient care.

Identification of a disulfidptosis-related genes signature for diagnostic and immune infiltration characteristics in endometriosis

Endometriosis (EMs) is the prevalent gynecological disease with the typical features of intricate pathogenesis and immune-related factors. Currently, there is no effective therapeutic intervention for EMs. Disulfidptosis, the cell death pattern discovered recently, may show close relationships to immunity and EMs. In this study, bioinformatics analysis was used to investigate the role of disulfide breakdown related genes (DRGs) in EMs. The EMs gene expression matrix was subjected to differential analysis for identifying overlap between differentially expressed genes (DEGs) in EMs and genes associated with disulfide poisoning. Immunoinfiltration analysis was performed. In addition, the association of hub genes with immune cells was examined. Multiple machine learning methods were employed to identify hub genes, construction of predictive models, and validation using external datasets and clinical samples. Totally 15 overlapping genes were identified. Immune-correlation analysis showed that NK cells played a vital role, and these 15 genes were closely related to NK cells. PDLIM1 was further determined as the hub gene through machine learning techniques. Clinical samples and external datasets were adopted for validating the performance in diagnosis. According to the above findings, we built the predictive model, and calculated the AUCs obtained from three external validation datasets to demonstrate the model accuracy. RT-qPCR and IHC analyses were applied to confirm the results. Colony formation was used to verify the effect of PDLIM1 on the proliferation of primary EMs cells. A strong correlation between disulfidptosis and EMs was identified in this study, highlighting its close correlation with the immune microenvironment. Moreover, our results shed new lights on exploring biomarkers and potential therapeutic targets for EMs.

Left atrial enlargement: a crucial indicator for identifying atrial fibrillation in patients with hypertension

Abstract Background Left atrial enlargement resulting from hypertension is closely linked to the development and persistence of atrial fibrillation (AF). The newly proposed staging recognizes AF as disease continuum, which makes us aware that AF prevention should focus on the Pre-AF stage, and atrial enlargement is one of the important manifestations in this stage. Previous scoring systems, such as CHA2DS2-VASc and C2HEST, along with the recently highlighted left atrial diameter (LAD), have been significant tools for predicting AF occurrence. However, a comprehensive assessment of their utility is currently lacking. Purpose This study aims to explore the role of left atrial size in identifying atrial fibrillation (AF) among hospitalized hypertensives, and to compare its recognition effectiveness with previous scoring systems. Methods We conducted a cross-sectional analysis within hospitalized hypertensives. The discovery, internal and external validation datasets were established. The XGBoost was employed to identify key variables related to AF occurrence, which were ranked based on their importance scores. To gauge the predictive prowess of LAD regarding AF occurrence, we plotted the receiver operating characteristic curve (ROC) and calculated the area under the curve (AUC). This enabled us to pinpoint the LAD cutoff value corresponding to the maximum Youden index, indicative of susceptibility to AF. Subsequently, Youden index determined the optimal cutoff value from the ROC curve. Delong’s test compared the identification abilities of different tools within the same dataset. Logistic regression analysis assessed the correlation between clinical variables and left atrial size. Results The XGBoost revealed 17 variables affecting AF occurrence in hypertensives, with LAD, age, and HF ranking among the top 3 (Figure 2a). LAD had a more pronounced impact on AF occurrence than other variables. LAD-based identification of AF in hypertensives had AUC of 0.827 (95%CI: 0.816-0.837), with an optimal cutoff of 38mm. Subgrouping the study population by LAD≥38mm, ROC curves showed AUCs of 0.757 (95%CI: 0.745-0.768) for the discovery set, 0.750 (95%CI: 0.731-0.768) for internal validation, and 0.759 (95%CI: 0.743-0.776) for external validation. LAD≥38mm had a more higher and stable AUC value for judging the AF occurrence than CHA2DS2-VASc and C2HEST scores (Figure 2b). Hypertensives with LAD≥38mm had a 9.72-fold higher AF risk than those with LAD<38mm. Conclusion LAD is a pivot determinant for AF occurrence. LAD≥38mm is a valuable identifier for high-risk hypertensives prone to AF.Figure 1.Center illustrationFigure 2.XGboost and Delong’s test of ROC

Automated quantification of cerebral microbleeds in susceptibility-weighted MRI: association with vascular risk factors, white matter hyperintensity burden, and cognitive function.

To train and validate a deep learning (DL)-based segmentation model for cerebral microbleeds (CMB) on susceptibility-weighted MRI; and to find associations between CMB, cognitive impairment, and vascular risk factors. Participants in this single-institution retrospective study underwent brain MRI to evaluate cognitive impairment between January-September 2023. For training the DL model, the nnU-Net framework was used without modifications. The DL model's performance was evaluated on independent internal and external validation datasets. Linear regression analysis was used to find associations between log-transformed CMB numbers, cognitive function (mini-mental status examination [MMSE]), white matter hyperintensity (WMH) burden, and clinical vascular risk factors (age, sex, hypertension, diabetes, lipid profiles, and body mass index). Training of the DL model (n = 287) resulted in a robust segmentation performance with an average dice score of 0.73 (95% CI, 0.67-0.79) in an internal validation set, (n = 67) and modest performance in an external validation set (dice score = 0.46, 95% CI, 0.33-0.59, n = 68). In a temporally independent clinical dataset (n = 448), older age, hypertension, and WMH burden were significantly associated with CMB numbers in all distributions (total, lobar, deep, and cerebellar; all P <.01). MMSE was significantly associated with hyperlipidemia (β = 1.88, 95% CI, 0.96-2.81, P <.001), WMH burden (β = -0.17 per 1% WMH burden, 95% CI, -0.27-0.08, P <.001), and total CMB number (β = -0.01 per 1 CMB, 95% CI, -0.02-0.001, P = .04) after adjusting for age and sex. The DL model showed a robust segmentation performance for CMB. In all distributions, CMB had significant positive associations with WMH burden. Increased WMH burden and CMB numbers were associated with decreased cognitive function. CMB = cerebral microbleed; DL = deep learning, DSC = dice similarity coefficient; MMSE = mini-mental status examination; SVD = small vessel disease; SWI = susceptibility-weighted image; WMH = white matter hyperintensity.

Machine Learning-Based In Silico Prediction of the Inhibitory Activity of Chemical Substances Against Rat and Human Cytochrome P450s.

The prediction of cytochrome P450 inhibition by a computational (quantitative) structure-activity relationship approach using chemical structure information and machine learning would be useful for toxicity research as a simple and rapid in silico tool. However, there are few in silico models focusing on the species differences between rat and human in the P450s inhibition. This study aimed to establish in silico models to classify chemical substances as inhibitors or non-inhibitors of various rat and human P450s, using only molecular descriptors. Using the in-house test results from our in vitro experiments, we used 326 substances for model construction and internal validation data. Apart from the 326 substances, 60 substances were used as external validation data set. We focused on seven rat P450s (CYP1A1, CYP1A2, CYP2B1, CYP2C6, CYP2D1, CYP2E1, and CYP3A2) and 11 human P450s (CYP1A1, CYP1A2, CYP1B1, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4). Most of the models established using XGBoost showed an area under the receiver operating characteristic curve (ROC-AUC) of 0.8 or more in the internal validation. When we set an applicability domain for the models and confirmed their generalization performance through external validation, most of the models showed an ROC-AUC of 0.7 or more. Interestingly, for CYP1A1 and CYP1A2, we discovered that a human P450 inhibitory activity model can predict rat P450 inhibitory activity and vice versa. These models are the first attempts to predict inhibitory activity against a wide variety of P450s in both rats and humans using chemical structure information. Our experimental results and in silico models would be helpful to support information for species similarities and differences in chemical-induced toxicity.

Machine learning and explainable artificial intelligence to predict pathologic stage in men with localized prostate cancer.

Though several nomograms exist, machine learning (ML) approaches might improve prediction of pathologic stage in patients with prostate cancer. To develop ML models to predict pathologic stage that outperform existing nomograms that use readily available clinicopathologic variables. Patients with prostate adenocarcinoma who underwent surgery were identified in the National Cancer Database. Seven ML models were trained to predict organ-confined (OC) disease, extracapsular extension, seminal vesicle invasion (SVI), and lymph node involvement (LNI). Model performance was measured using area under the curve (AUC) on a holdout testing data set. Clinical utility was evaluated using decision curve analysis (DCA). Performance metrics were confirmed on an external validation data set. The ML-based extreme gradient boosted trees model achieved the best performance with an AUC of 0.744, 0.749, 0.816, 0.811 for the OC, ECE, SVI, and LNI models, respectively. The MSK nomograms achieved an AUC of 0.708, 0.742, 0.806, 0.802 for the OC, ECE, SVI, and LNI models, respectively. These models also performed the best on DCA. Findings were consistent on both a holdout internal validation data set as well as an external validation data set. Our ML models better predicted pathologic stage relative to existing nomograms at predicting pathologic stage. Accurate prediction of pathologic stage can help oncologists and patients determine optimal definitive treatment options for patients with prostate cancer.

Triple and quadruple optimization for feature selection in cancer biomarker discovery

The proliferation of omics data has advanced cancer biomarker discovery but often falls short in external validation, mainly due to a narrow focus on prediction accuracy that neglects clinical utility and validation feasibility. We introduce three- and four-objective optimization strategies based on genetic algorithms to identify clinically actionable biomarkers in omics studies, addressing classification tasks aimed at distinguishing hard-to-differentiate cancer subtypes beyond histological analysis alone. Our hypothesis is that by optimizing more than one characteristic of cancer biomarkers, we may identify biomarkers that will enhance their success in external validation. Our objectives are to: (i) assess the biomarker panel’s accuracy using a machine learning (ML) framework; (ii) ensure the biomarkers exhibit significant fold-changes across subtypes, thereby boosting the success rate of PCR or immunohistochemistry validations; (iii) select a concise set of biomarkers to simplify the validation process and reduce clinical costs; and (iv) identify biomarkers crucial for predicting overall survival, which plays a significant role in determining the prognostic value of cancer subtypes. We implemented and applied triple and quadruple optimization algorithms to renal carcinoma gene expression data from TCGA. The study targets kidney cancer subtypes that are difficult to distinguish through histopathology methods. Selected RNA-seq biomarkers were assessed against the gold standard method, which relies solely on clinical information, and in external microarray-based validation datasets. Notably, these biomarkers achieved over 0.8 of accuracy in external validations and added significant value to survival predictions, outperforming the use of clinical data alone with a superior c-index. The provided tool also helps explore the trade-off between objectives, offering multiple solutions for clinical evaluation before proceeding to costly validation or clinical trials.

Development, validation, and visualization of a web-based nomogram for predicting chronic kidney disease incidence at health examination centers

Purpose To develop and validate a web-based nomogram for predicting new incident chronic kidney disease (CKD) within 4 years in a cohort undergoing routine physical examination from two health examination centers. Methods One center was utilized for training and internal validation of a nomogram model involving 6515 patients, while a separate center was employed for external validation with 3152 patients. Sixteen candidate predictors, including patient demographics, medical histories, physical examination, and laboratory test data, were included in this study to ascertain factors linked to new incident CKD. A nomogram was created to predict CKD risks using a logistic model. The nomogram’s performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration plot, and decision curve analysis. Results Out of the 9667 healthy individuals included in the study with mean age of 46 years, sex ratio (male/female) of 1.69 (6075/3592), 118 (2.59%), 51 (2.61%), and 60 (1.90%) individuals developed CKD in the training (n = 4563), internal validation (n = 1952), and external validation (n = 3152) datasets, respectively. Age, history of diabetes mellitus, systolic blood pressure, serum creatinine, albumin, and triglyceride levels were used to build the nomogram, which yielded excellent discrimination ability (training cohort, AUC = 0.8806, 95% confidence interval [CI] 0.8472–0.9141; internal validation cohort, AUC = 0.8506, 95% CI 0.7856–0.9156; external validation cohort, AUC = 0.9183, 95% CI 0.8698–0.9669). We further developed a web-based calculator for convenient application (https://luochuxuan.shinyapps.io/dynnomapp/). Conclusion Our web-based nomogram accurately predicted CKD risks in Chinese health individuals and can be easily used in clinical settings.

Enhancing the convenience of frailty index assessment for elderly Chinese people with machine learning methods.

Frailty is a state that is closely associated with adverse health outcomes in the aging process. The frailty index (FI), which measures frailty in terms of cumulative deficits, has been widely used for frailty assessment in elderly people, and its advantage of self-reported information collection makes it applicable to a broader group of elderly people. Our study aims to simplify the Frailty Index Assessment Scale, while maintaining its reliability and accuracy, to easily and quickly assess frailty in elderly people. In this study, participants (age ≥ 65 years) from the Chinese Longitudinal Healthy Longevity Survey (CLHLS), which had 13,339, 372 and 1214 participants in 2008, 2011, and 2014, respectively, were used. The 2008 dataset was split into 80% for training and 20% for internal validation, and the data from 2011 to 2014 as external validation. In order to obtain effective predictors, we used Lasso regression, Boruta algorithm and random forest classifier score for feature selection. We used six models for predictive model construction and evaluated the models in the validation dataset. Model performance was measured by area under the curve (AUC), accuracy and F1 score. Logistic regression was found to be the best performing and most interpretable algorithm with AUC, accuracy and F1 of 0.974, 0.932 and 0.880 for the validation dataset, respectively. The AUCs for the external independent validation dataset were 0.963 and 0.977, respectively. Subgroup analysis showed that the model had good predictive power in both males and females. The predictive power was stronger among the elderly people over 80 years old, with AUC, accuracy and F1 of 0.973,0.914, and 0.893, respectively. The model also obtained good predictive power in the case of FI measured by different indicators. The model showed good robustness in the follow-up assessment of frailty status in elderly people, with the AUC remaining above 0.95 and accuracy above 0.9 over the long-term follow-up. Using machine learning techniques, we have successfully developed a simple frailty assessment prediction model based on 10 key features to shorten the frailty assessment scale with near full-scale accuracy. A user-friendly website was created to facilitate the application of this prediction model ( https://healthy-aging.shinyapps.io/Frailty_Assessment/ ).

External validation of an artificial intelligence multi-label deep learning model capable of ankle fracture classification

BackgroundAdvances in medical imaging have made it possible to classify ankle fractures using Artificial Intelligence (AI). Recent studies have demonstrated good internal validity for machine learning algorithms using the AO/OTA 2018 classification. This study aimed to externally validate one such model for ankle fracture classification and ways to improve external validity.MethodsIn this retrospective observation study, we trained a deep-learning neural network (7,500 ankle studies) to classify traumatic malleolar fractures according to the AO/OTA classification. Our internal validation dataset (IVD) contained 409 studies collected from Danderyd Hospital in Stockholm, Sweden, between 2002 and 2016. The external validation dataset (EVD) contained 399 studies collected from Flinders Medical Centre, Adelaide, Australia, between 2016 and 2020. Our primary outcome measures were the area under the receiver operating characteristic (AUC) and the area under the precision-recall curve (AUPR) for fracture classification of AO/OTA malleolar (44) fractures. Secondary outcomes were performance on other fractures visible on ankle radiographs and inter-observer reliability of reviewers.ResultsCompared to the weighted mean AUC (wAUC) 0.86 (95%CI 0.82–0.89) for fracture detection in the EVD, the network attained wAUC 0.95 (95%CI 0.94–0.97) for the IVD. The area under the precision-recall curve (AUPR) was 0.93 vs. 0.96. The wAUC for individual outcomes (type 44A-C, group 44A1-C3, and subgroup 44A1.1-C3.3) was 0.82 for the EVD and 0.93 for the IVD. The weighted mean AUPR (wAUPR) was 0.59 vs 0.63. Throughout, the performance was superior to that of a random classifier for the EVD.ConclusionAlthough the two datasets had considerable differences, the model transferred well to the EVD and the alternative clinical scenario it represents. The direct clinical implications of this study are that algorithms developed elsewhere need local validation and that discrepancies can be rectified using targeted training. In a wider sense, we believe this opens up possibilities for building advanced treatment recommendations based on exact fracture types that are more objective than current clinical decisions, often influenced by who is present during rounds.

Multicenter evaluation of CT deep radiomics model in predicting Leibovich score risk groups for non-metastatic clear cell renal cell carcinoma

BackgroundNon-metastatic clear cell renal cell carcinoma (nccRCC) poses a significant risk of postoperative recurrence and metastasis, underscoring the importance of accurate preoperative risk assessment. While the Leibovich score is effective, it relies on postoperative histopathological data. This study aims to evaluate the efficacy of CT radiomics and deep learning models in predicting Leibovich score risk groups in nccRCC, and to explore the interrelationship between CT and pathological features. Patients and MethodsThis research analyzed 600 nccRCC patients from four datasets, dividing them into low (Leibovich scores of 0–2) and intermediate to high risk (Leibovichscores exceeding 3) groups. Radiological models were developed from CT subjective features, and radiomics and deep learning models were constructed from CT images. Additionally, a deep radiomics model and a fusion model integrating radiological, radiomics, and deep learning features were introduced. Model performance was assessed by AUC values, while survival differences across predicted groups were analyzed using survival curves and the DeLong test. Moreover, the research investigated the connection between CT and pathological features derived from whole-slide pathological images in the second validation dataset. ResultsWithin the training dataset, four radiological, three radiomics, and thirteen deep learning features were selected to develop models predicting nccRCC Leibovich score risk groups. The deep radiomics model demonstrated superior predictive accuracy, evidenced by AUC values of 0.881, 0.829, and 0.819 in external validation datasets. Notably, significant differences in overall survival were observed among patients classified by this model (DeLong test p < 0.05 across all datasets). Furthermore, a correlation and complementarity were observed between deep radiomics features and pathological deep learning features. ConclusionsThe CT deep radiomics model precisely predicts nccRCC Leibovich score risk groups preoperatively and highlights the synergistic effect between CT and pathological data.

Machine Learning Algorithm to Predict Atrial Fibrillation Using Serial 12-Lead ECGs Based on Left Atrial Remodeling.

We hypothesized that analysis of serial ECGs could predict new-onset atrial fibrillation (AF) more accurately than analysis of a single ECG by detecting the subtle cardiac remodeling that occurs immediately before AF occurrence. Our aim in this study was to compare the performance of 2 types of machine learning (ML) algorithms. Standard 12-lead ECGs of patients selected by cardiologists between January 2010 and May 2021 were used for ML model development. Two ML models (single ECG and serial ECG) were developed using a light gradient boosting machine-learning algorithm. Model performance was evaluated based on the area under the receiver operating characteristic curve, sensitivity, specificity, accuracy, and F1 score. We trained the ML models on 415 964 ECGs from 176 090 patients. When testing the 2 ML models using external validation data sets, the performance of the serial-ML model was significantly better than that of the single-ML model for predicting new-onset AF (single- versus serial-ML model: sensitivity 0.744 versus 0.810; specificity 0.742 versus 0.822; accuracy 0.743 versus 0.816; F1 score 0.743 versus 0.815; area under the receiver operating characteristic curve 0.812 versus 0.880; P<0.001). The Shapley Additive Explanations analysis ranked P-wave duration and amplitude among the top 10 ECG parameters. An ML model based on serial ECGs from an individual had greater ability to predict new-onset AF than the ML model based on a single ECG. P-wave morphologies were associated with future AF prediction.