Search

Purpose To assess the performance of different machine learning (ML) approaches in identifying risk factors for diabetic ketoacidosis (DKA) and predicting DKA. Methods This study applied flexible ML (XGBoost, distributed random forest [DRF] and feedforward network) and conventional ML approaches (logistic regression and least absolute shrinkage and selection operator [LASSO]) to 3,400 DKA cases and 11,780 controls nested in adults with type 1 diabetes identified from Optum® de-identified Electronic Health Record dataset (2007-2018). Area under the curve (AUC), accuracy, sensitivity and specificity were computed using 5-fold cross validation, and their 95% confidence intervals (CI) were established using 1,000 bootstrap samples. The importance of predictors was compared across these models. Results In the training set, XGBoost and feedforward network yielded higher AUC values (0.89 and 0.86, respectively) than logistic regression (0.83), LASSO (0.83) and DRF (0.81). However, the AUC values were similar (0.82) among these approaches in the test set (95% CI range, 0.80-0.84). While the accuracy values >0.8 and the specificity values >0.9 for all models, the sensitivity values were only 0.4. The differences in these metrics across these models were minimal in the test set. All approaches selected some known risk factors for DKA as the top ten features. XGBoost and DRF included more laboratory measurements or vital signs compared with conventional ML approaches, while feedforward network included more social demographics. Conclusions In our empirical study, all ML approaches demonstrated similar performance, and identified overlapping, but different, top ten predictors. The difference in selected top predictors needs further research. This article is protected by copyright. All rights reserved.

This paper aims to discover a suitable combination of contemporary feature selection techniques and robust prediction classifiers. As such, to examine the impact of the feature selection method on classifier performance, we use two Chinese and three other real-world credit scoring datasets. The utilized feature selection methods are the least absolute shrinkage and selection operator (LASSO), multivariate adaptive regression splines (MARS). In contrast, the examined classifiers are the classification and regression trees (CART), logistic regression (LR), artificial neural network (ANN), and support vector machines (SVM). Empirical findings confirm that LASSO’s feature selection method, followed by robust classifier SVM, demonstrates remarkable improvement and outperforms other competitive classifiers. Moreover, ANN also offers improved accuracy with feature selection methods; LR only can improve classification efficiency through performing feature selection via LASSO. Nonetheless, CART does not provide any indication of improvement in any combination. The proposed credit scoring modeling strategy may use to develop policy, progressive ideas, operational guidelines for effective credit risk management of lending, and other financial institutions. The finding of this study has practical value, as to date, there is no consensus about the combination of feature selection method and prediction classifiers.

Objective: This paper aims at proposing a new machine-learning based model to improve the calculation of mealtime insulin boluses (MIB) in type 1 diabetes (T1D) therapy using continuous glucose monitoring (CGM) data. Indeed, MIB is still often computed through the standard formula (SF), which does not account for glucose rate-of-change ( $\Delta$ G), causing critical hypo/hyperglycemic episodes. Methods: Four candidate models for MIB calculation, based on multiple linear regression (MLR) and least absolute shrinkage and selection operator (LASSO) are developed. The proposed models are assessed in silico , using the UVa/Padova T1D simulator, in different mealtime scenarios and compared to the SF and three $\Delta$ G-accounting variants proposed in the literature. An assessment on real data, by retrospectively analyzing 218 glycemic traces, is also performed. Results: All four tested models performed better than the existing techniques. LASSO regression with extended feature-set including quadratic terms (LASSO $_Q$ ) produced the best results. In silico, LASSO $_Q$ reduced the error in estimating the optimal bolus to only 0.86 U (1.45 U of SF and 1.36–1.44 U of literature methods), as well as hypoglycemia incidence (from 44.41% of SF and 44.60–45.01% of literature methods, to 35.93%). Results are confirmed by the retrospective application to real data. Conclusion: New models to improve MIB calculation accounting for CGM- $\Delta$ G and easy-to-measure features can be developed within a machine learning framework. Particularly, in this paper, a new LASSO $_Q$ model was developed, which ensures better glycemic control than SF and other literature methods. Significance: MIB dosage with the proposed LASSO $_Q$ model can potentially reduce the risk of adverse events in T1D therapy.

We developed a radiomics model to differentiate hematologic marrow diseases and compared the performance with radiologists' readings and a quantitative measurement. Patients were retrospectively analyzed from the diseased (n = 254) and control groups (n = 230). A sagittal T1-weighted lumbar spinal MR image was normalized by an intervertebral disk, and bone marrow was segmented. A hundred features were extracted, and final features were selected using Principle Component Analysis (PCA) and least absolute shrinkage and selection operator (LASSO). Finally, Random forest (RF) and logistic regression (LR) models were trained. Two radiologists with different levels of experience analyzed the images for the presence of bone marrow diseases, independently. The area under the receiver operating characteristic curves (AUC) and decision curve analysis (DCA) was evaluated. Among the subjects, 363 cases were assigned as a training set and 121 as a validation set. The combination of LASSO and RF produced the best results. With the validation set, the sensitivity (SE) was 87.3%, specificity (SP) was 86.2% and AUC was 0.928 ($p

Objective: To identify a multi-gene prognostic factor in patients with lung adenocarcinoma (LUAD). Materials and methods Prognosis-related genes were screened in the TCGA-LUAD cohort. Then, patients in this cohort were randomly separated into training set and test set. Least absolute shrinkage and selection operator (LASSO) regression was performed to the penalized the Cox proportional hazards regression (CPH) model on the training set, and a prognostication combination based on the result of LASSO analysis was developed. By performing Kaplan-Meier curve analysis, univariate and multivariable CPH model on the overall survival (OS) as well as recurrence free survival (RFS), the prognostication performance of the multigene combination were evaluated. Moreover, we constructed a nomogram and performed decision curve analysis to evaluate the clinical application of the multigene combination. Results We obtained 99 prognosis-related genes and screened out a 4-gene combination (including CIDEC, ZFP3, DKK1, and USP4) according to the LASSO analysis. The results of survival analyses suggested that patients in the 4-gene combination low-risk group had better OS and RFS than those in the 4-gene combination high-risk group. The 4-gene mentioned was demonstrated to be independent prognostic factor of patients with LUAD in the training set(OS, HR=11.962, P

Background Extensive literature has shown an association of Adverse Childhood Experiences (ACEs) with adverse health outcomes; however, its ability to predict events or stratify risks is less known. Individuals with mental illness and ACE exposure have been shown to visit emergency departments (ED) more often than those in the general population. This study thus examined the ability of the ACEs checklist to predict ED visits within the subsequent year among children and adolescents presenting to mental health clinics with pre-existing mental health issues. Methods The study analyzed linked data (n = 6100) from two databases provided by Alberta Health Services (AHS). The Regional Access and Intake System (RAIS 2016–2018) database provided data on the predictors (ACE items, age, sex, residence, mental health program type, and primary diagnosis) regarding children and adolescents (aged 0–17 years) accessing addiction and mental health services within Calgary Zone, and the National Ambulatory Care Reporting System (NACRS 2016–2019) database provided data on ED visits. A 25% random sample of the data was reserved for validation purposes. Two Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression models, each employing a different method to tune the shrinkage parameter lambda (namely cross-validated and adaptive) and performing 10-fold cross-validation for a set of 100 lambdas in each model were examined. Results The adaptive LASSO model had a slightly better fit in the validation dataset than the cross-validated model; however, it still demonstrated poor discrimination (AUC 0.60, sensitivity 37.8%, PPV 49.6%) and poor calibration (over-triaged in low-risk and under-triaged in high-risk subgroups). The model’s poor performance was evident from an out-of-sample deviance ratio of − 0.044. Conclusion The ACEs checklist did not perform well in predicting ED visits among children and adolescents with existing mental health concerns. The diverse causes of ED visits may have hindered accurate predictions, requiring more advanced statistical procedures. Future studies exploring other machine learning approaches and including a more extensive set of childhood adversities and other important predictors may produce better predictions. Furthermore, despite highly significant associations being observed, ACEs may not be deterministic in predicting health-related events at the individual level, such as general ED use.

Ciguatera poisoning is a globally occurring seafood disease caused by the ingestion of marine products contaminated with dinoflagellate produced neurotoxins. Persistent forms of ciguatera, which prove to be highly debilitating, are poorly studied and represent a significant medical issue. The present study aims to better understand chronic ciguatera manifestations and identify potential predictive factors for their duration. Medical files of 49 patients were analyzed, and the post-hospitalization evolution of the disease assessed through a follow-up questionnaire. A rigorous logistic lasso regression model was applied to select significant predictors from a list of 37 patient characteristics potentially predictive of having chronic symptoms. Missing data were handled by complete case analysis, and a survival analysis was implemented. All models used standardized variables, and multiple comparisons in the survival analyses were handled by Bonferroni correction. Among all studied variables, five significant predictors of having symptoms lasting ≥3 months were identified: age, tobacco consumption, acute bradycardia, laboratory measures of urea, and neutrophils. This exploratory, hypothesis-generating study contributes to the development of ciguatera epidemiology by narrowing the list from 37 possible predictors to a list of five predictors that seem worth further investigation as candidate risk factors in more targeted studies of ciguatera symptom duration.

BackgroundTimely diagnosis of ischemic stroke (IS) in the acute phase is extremely vital to achieve proper treatment and good prognosis. In this study, we developed a novel prediction model based on the easily obtained information at initial inspection to assist in the early identification of IS.MethodsA total of 627 patients with IS and other intracranial hemorrhagic diseases from March 2017 to June 2018 were retrospectively enrolled in the derivation cohort. Based on their demographic information and initial laboratory examination results, the prediction model was constructed. The least absolute shrinkage and selection operator algorithm was used to select the important variables to form a laboratory panel. Combined with the demographic variables, multivariate logistic regression was performed for modeling, and the model was encapsulated within a visual and operable smartphone application. The performance of the model was evaluated on an independent validation cohort, formed by 304 prospectively enrolled patients from June 2018 to May 2019, by means of the area under the curve (AUC) and calibration.ResultsThe prediction model showed good discrimination (AUC = 0.916, cut-off = 0.577), calibration, and clinical availability. The performance was reconfirmed in the more complex emergency department. It was encapsulated as the Stroke Diagnosis Aid app for smartphones. The user can obtain the identification result by entering the values of the variables in the graphical user interface of the application.ConclusionThe prediction model based on laboratory and demographic variables could serve as a favorable supplementary tool to facilitate complex, time-critical acute stroke identification.

BACKGROUND Hepatitis B virus, together with hepatitis C virus, has been recognized as the leading causes of hepatocellular carcinoma (HCC). Long non-coding RNAs (lncRNAs) have been suggested in increasing studies to be the potential prognostic factors for HCC. However, the role of combined application of lncRNAs in estimating overall survival (OS) for hepatitis virus positive HCC (VHCC) is uncertain. AIM To construct an lncRNA signature related to the OS of VHCC patients to enhance the accuracy of prognosis prediction. METHODS The expression patterns of lncRNAs, as well as related clinical data were collected from 149 VHCC patients from The Cancer Genome Atlas database. The R package was adopted to obtain the differentially expressed lncRNAs (DElncRNAs). LncRNAs significantly associated with OS were screened by means of univariate Cox regression analysis, so as to construct a least absolute shrinkage and selection operator (LASSO) model. Subsequently, the constructed lncRNA signature was developed and validated. Afterwards, the prognostic nomogram was established, which combined the as-established lncRNA signature as well as the clinical features. Meanwhile, subgroup analysis stratified by the virus type was also performed. Finally, the above-mentioned lncRNAs were enriched to corresponding pathways according to the markedly co-expressed genes. RESULTS A total of 1420 DElncRNAs were identified, among which 406 were significant in univariate Cox regression analysis. LASSO regression confirmed 8 out of the 406 lncRNAs, including AC005722.2, AC107959.3, AL353803.1, AL589182.1, AP000844.2, AP002478.1, FLJ36000, and NPSR1-AS1. Then, the prognostic risk score was calculated. Our results displayed a significant association between the risk model and the OS of VHCC [hazard ratio = 1.94, 95% confidence interval (CI): 1.61-2.34, log-rank P = 2e-10]. The inference tree suggested that the established lncRNA signature was useful in the risk stratification of VHCC. Furthermore, a nomogram was plotted, and the concordance index of internal validation was 0.763 (95%CI: 0.700-0.826). Moreover, the subgroup analysis regarding etiology confirmed this risk model. In addition, the Wnt signaling pathway, angiogenesis, the p53 pathway, and the PI3 kinase pathway were the remarkably enriched pathways. CONCLUSION An eight-lncRNA signature has been established to predict the prognosis for VHCC, which contributes to providing a novel foundation for the targeted therapy of VHCC.

PurposeOur study aimed to establish and validate prognostic nomograms based on gross tumor volume (GTV) and cervical nodal volume (CNV) for nasopharyngeal carcinoma (NPC) patients treated with two cycles of concurrent chemoradiotherapy (CCRT).MethodsFrom 2012 to 2015, 620 eligible patients who received radical treatment at the Cancer Hospital of Shantou University Medical College were recruited for a nomogram study. Variables were determined in a training set of 463 patients from 2012 to 2014 by X-tile analysis, univariate and multivariate Cox proportional hazard analyses, and the least absolute shrinkage and selection operator (LASSO). Another cohort of 157 patients in 2015 was validated with bootstrap resampling. The concordance index (C-index) and calibration curves were applied to assess its predictive discriminative and accuracy ability, while decision curve analysis (DCA), X-tile analysis and Kaplan–Meier curve for clinical application.ResultsIndependent prognostic variables for overall survival (OS) were age, GTV, CNV, cranial nerve, positive cervical lymph node laterality below the caudal border of cricoid cartilage (LNBC), and were selected for the nomogram. Optimal prognostic factors including Karnofsky performance status (KPS), age, GTV, CNV, LNBC were incorporated in the nomogram for progression-free survival (PFS). In the training set, the C-index of our nomograms for OS and PFS were 0.755 (95% CI, 0.704 to 0.807) and 0.698 (95% CI, 0.652 to 0.744). The calibration curve showed good agreement between nomogram-predicted and actual survival. DCA indicated that our nomograms were of clinical benefit.ConclusionOur nomograms are capable of effective prognostic prediction for patients with NPC.

We aimed to develop and validate a model for predicting mortality in patients with angina across the spectrum of dysglycemia. A total of 1479 patients admitted for coronary angiography due to angina were enrolled. All-cause mortality served as the primary endpoint. The models were validated with five-fold cross validation to predict long-term mortality. The features selected by least absolute shrinkage and selection operator (LASSO) were age, heart rate, plasma glucose levels at 30 min and 120 min during an oral glucose tolerance test (OGTT), the use of angiotensin II receptor blockers, the use of diuretics, and smoking history. This best performing model was built using a random survival forest with selected features. It had a good discriminative ability (Harrell’s C-index: 0.829) and acceptable calibration (Brier score: 0.08) for predicting long-term mortality. Among patients with obstructive coronary artery disease confirmed by angiography, our model outperformed the Global Registry of Acute Coronary Events discharge score for mortality prediction (Harrell’s C-index: 0.829 vs. 0.739, p &lt, 0.001). In conclusion, we developed a machine learning model to predict long-term mortality among patients with angina. With the integration of OGTT, the model could help to identify a high risk of mortality across the spectrum of dysglycemia.

The prognosis of acute myeloid leukemia (AML) is closely related to immune response changes. Further exploration of the pathobiology of AML focusing on immune-related genes would contribute to the development of more advanced evaluation and treatment strategies. In this study, we established a novel immune-17 signature based on transcriptome data from The Cancer Genome Atlas (TCGA) and The Genotype-Tissue Expression (GTEx) databases. We found that immune biology processes and transcriptional dysregulations are critical factors in the development of AML through enrichment analyses. We also formulated a prognostic model to predict the overall survival of AML patients by using LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis. Furthermore, we incorporated the immune-17 signature to improve the prognostic accuracy of the ELN2017 risk stratification system. We concluded that the immune-17 signature represents a novel useful model for evaluating AML survival outcomes and may be implemented to optimize treatment selection in the next future.

This study aimed to investigate the value of amplitude of low-frequency fluctuation (ALFF)-based histogram analysis in the diagnosis of Parkinson’s disease (PD) and to investigate the regions of the most important discriminative features and their contribution to classification discrimination. Patients with PD (n = 59) and healthy controls (HCs; n = 41) were identified and divided into a primary set (80 cases, including 48 patients with PD and 32 HCs) and a validation set (20 cases, including 11 patients with PD and nine HCs). The Automated Anatomical Labeling (AAL) 116 atlas was used to extract the histogram features of the regions of interest in the brain. Machine learning methods were used in the primary set for data dimensionality reduction, feature selection, model construction, and model performance evaluation. The model performance was further validated in the validation set. After feature data dimension reduction and feature selection, 23 of a total of 1,276 features were entered in the model. The brain regions of the selected features included the frontal, temporal, parietal, occipital, and limbic lobes, as well as the cerebellum and the thalamus. In the primary set, the area under the curve (AUC) of the model was 0.974, the sensitivity was 93.8%, the specificity was 90.6%, and the accuracy was 93.8%. In the validation set, the AUC, sensitivity, specificity, and accuracy were 0.980, 90.9%, 88.9%, and 90.0%, respectively. ALFF-based histogram analysis can be used to classify patients with PD and HCs and to effectively identify abnormal brain function regions in PD patients.

Background Dietary pattern analysis is a promising approach to understanding the complex relationship between diet and health. While many statistical methods exist, the literature predominantly focuses on classical methods such as dietary quality scores, principal component analysis, factor analysis, clustering analysis, and reduced rank regression. There are some emerging methods that have rarely or never been reviewed or discussed adequately. Methods This paper presents a landscape review of the existing statistical methods used to derive dietary patterns, especially the finite mixture model, treelet transform, data mining, least absolute shrinkage and selection operator and compositional data analysis, in terms of their underlying concepts, advantages and disadvantages, and available software and packages for implementation. Results While all statistical methods for dietary pattern analysis have unique features and serve distinct purposes, emerging methods warrant more attention. However, future research is needed to evaluate these emerging methods’ performance in terms of reproducibility, validity, and ability to predict different outcomes. Conclusion Selection of the most appropriate method mainly depends on the research questions. As an evolving subject, there is always scope for deriving dietary patterns through new analytic methodologies.

BackgroundThis study aimed to assess the prognostic value of various diagnostic immunohistochemical (IHC) markers and develop an IHC-based classifier to predict the disease-free survival (DFS) of patients with bladder cancer undergoing radical cystectomy.MethodsIHC was performed on tumor specimens from 366 patients with transitional cell bladder cancer. The least absolute shrinkage and selection operator (LASSO) Cox regression model was used to develop a multi-marker classifier for predicting DFS of patients with bladder cancer. The Kaplan–Meier estimate was performed to assess DFS, and unadjusted and adjusted Cox regression models were used to identify independent risk factors to predict DFS of patients with bladder cancer.ResultsBased on the LASSO Cox regression model, nine prognostic markers were identified in the training cohort. Patients were stratified into low- and high-risk groups using the IHC-based classifier. In the training cohort, the 10-year DFS was significantly better in low-risk patients (71%) compared with high-risk patients (18%) (p < 0.001); in the validation cohort, the 10-year DFS was 86% for the low-risk group and 20% for the high-risk group (p < 0.001). Multivariable Cox regression analyses showed that the high-risk group based on the classifier was associated with poorer DFS adjusted by clinicopathological characteristics. Finally, a nomogram comprising the classifier and clinicopathological factors was developed for clinical application.ConclusionThe nine-IHC-based classifier is a reliable prognostic tool, which can eventually guide clinical decision making regarding treatment strategy and follow-up scheduling of bladder cancer.

IntroductionThe insufficient understanding and misdiagnosis of clinically diagnosed pulmonary tuberculosis (PTB) without an aetiological evidence is a major problem in the diagnosis of tuberculosis (TB). This study aims to confirm the value of Long non-coding RNA (lncRNA) n344917 in the diagnosis of PTB and construct a rapid, accurate, and universal prediction model.MethodsA total of 536 patients were prospectively and consecutively recruited, including clinically diagnosed PTB, PTB with an aetiological evidence and non-TB disease controls, who were admitted to West China hospital from Dec 2014 to Dec 2017. The expression levels of lncRNA n344917 of all patients were analyzed using reverse transcriptase quantitative real-time PCR. Then, the laboratory findings, electronic health record (EHR) information and expression levels of n344917 were used to construct a prediction model through the Least Absolute Shrinkage and Selection Operator algorithm and multivariate logistic regression.ResultsThe factors of n344917, age, CT calcification, cough, TBIGRA, low-grade fever and weight loss were included in the prediction model. It had good discrimination (area under the curve = 0.88, cutoff = 0.657, sensitivity = 88.98%, specificity = 86.43%, positive predictive value = 85.61%, and negative predictive value = 89.63%), consistency and clinical availability. It also showed a good replicability in the validation cohort. Finally, it was encapsulated as an open-source and free web-based application for clinical use and is available online at https://ziruinptb.shinyapps.io/shiny/.ConclusionCombining the novel potential molecular biomarker n344917, laboratory and EHR variables, this web-based prediction model could serve as a user-friendly, accurate platform to improve the clinical diagnosis of PTB.

Introduction The immune status of the tumor microenvironment is extremely complex. One single immune feature cannot reflect the integral immune status, and its prognostic value was limited. We postulated that the immune signature based on multiple immuno-features could markedly improve the prediction of post-chemoradiotherapeutic survival in inoperable locally advanced non-small-cell lung cancer (LA-NSCLC) patients. Methods In this study, 100 patients who were diagnosed as having inoperable LA-NSCLC between January 2005 and January 2016 were analyzed. A five immune features-based signature was then constructed using the nested repeat 10-fold cross validation with least absolute shrinkage and selection operator (LASSO) Cox regression model. Nomograms were then established for predicting prognosis. Results The immune signature combining five immuno-features was significantly associated with overall survival (OS) and progression-free survival (PFS) (P = 0.002 and P = 0.014, respectively) in patients with inoperable LA-NSCLC, and at a cutoff of -0.05 stratified patients into two groups with 5-year OS rates of 39.8 and 8.8%, and 2-year PFS rates of 22.2 and 5.5% for the high- and low-immune signature groups, respectively. Integrating immune signature, we proposed predictive nomograms that were better than the traditional TNM staging system in terms of discriminating ability (OS: 0.692 vs. 0.588; PFS: 0.672 vs. 0.586, respectively) or net weight classification (OS: 32.96%; PFS: 9.22%), suggesting that the immune signature plays a significant role in improving the prognostic value. Conclusion Multiple immune features-based immune signature could effectively predict recurrence and survival of inoperable LA-NSCLC patients and complemented the prognostic value of the TNM staging system.

Background Cervical cancer (CC) represents the fourth most frequently diagnosed malignancy affecting women all over the world. However, effective prognostic biomarkers are still limited for accurately identifying high-risk patients. Here, we provided a combination machine learning algorithm-based signature to predict the prognosis of cervical squamous cell carcinoma (CSCC). Methods and materials After utilizing RNA sequencing (RNA-seq) data from 36 formalin-fixed and paraffin-embedded (FFPE) samples, the most significant modules were highlighted by the weighted gene co-expression network analysis (WGCNA). A candidate genes-based prognostic classifier was constructed by the least absolute shrinkage and selection operator (LASSO) and then validated in an independent validation set. Finally, based on the multivariate analysis, a nomogram including the FIGO stage, therapy outcome, and risk score level was built to predict progression-free survival (PFS) probability. Results A mRNA-based signature was developed to classify patients into high- and low-risk groups with significantly different PFS and overall survival (OS) rate (training set: p Conclusion The mRNA-based biomarker is a powerful and independent prognostic factor. Furthermore, the nomogram comprising our prognostic classifier is a promising predictor in identifying the progression risk of CSCC patients.

Background Gastric cancer (GC) is one of the most frequently diagnosed gastrointestinal cancers throughout the world. Novel prognostic biomarkers are required to predict the prognosis of GC. Aim To identify a multi-long noncoding RNA (lncRNA) prognostic model for GC. Methods Transcriptome data and clinical data were downloaded from The Cancer Genome Atlas. COX and least absolute shrinkage and selection operator regression analyses were performed to screen for prognosis associated lncRNAs. Receiver operating characteristic curve and Kaplan-Meier survival analyses were applied to evaluate the effectiveness of the model. Results The prediction model was established based on the expression of AC007991.4, AC079385.3, and AL109615.2 Based on the model, GC patients were divided into "high risk" and "low risk" groups to compare the differences in survival. The model was re-evaluated with the clinical data of our center. Conclusion The 3-lncRNA combination model is an independent prognostic factor for GC.

Robust early prediction of clinical outcomes in Parkinson&rsquo, s disease (PD) is paramount for implementing appropriate management interventions. We propose a method that uses the baseline MRI, measuring diffusion parameters from multiple parcellated brain regions, to predict the 2-year clinical outcome in Parkinson&rsquo, s disease. Diffusion tensor imaging was obtained from 82 patients (males/females = 45/37, mean age: 60.9 &plusmn, 7.3 years, baseline and after 23.7 &plusmn, 0.7 months) using a 3T MR scanner, which was normalized and parcellated according to the Automated Anatomical Labelling template. All patients were diagnosed with probable Parkinson&rsquo, s disease by the National Institute of Neurological Disorders and Stroke criteria. Clinical outcome was graded using disease severity (Unified Parkinson&rsquo, s Disease Rating Scale and Modified Hoehn and Yahr staging), drug administration (levodopa equivalent daily dose), and quality of life (39-item PD Questionnaire). Selection and regularization of diffusion parameters, the mean diffusivity and fractional anisotropy, were performed using least absolute shrinkage and selection operator (LASSO) between baseline diffusion index and clinical outcome over 2 years. Identified features were entered into a stepwise multivariate regression model, followed by a leave-one-out/5-fold cross validation and additional blind validation using an independent dataset. The predicted Unified Parkinson&rsquo, s Disease Rating Scale for each individual was consistent with the observed values at blind validation (adjusted R2 0.76) by using 13 features, such as mean diffusivity in lingual, nodule lobule of cerebellum vermis and fractional anisotropy in rolandic operculum, and quadrangular lobule of cerebellum. We conclude that baseline diffusion MRI is potentially capable of predicting 2-year clinical outcomes in patients with Parkinson&rsquo, s disease on an individual basis.

Background Coronary heart disease (CHD) is a common cardiovascular disease with high morbidity and mortality in China. The CHD risk prediction model has a great value in early prevention and diagnosis. Methods In this study, CHD risk prediction models among rural residents in Xinxiang County were constructed using Random Forest (RF), Support Vector Machine (SVM), and the least absolute shrinkage and selection operator (LASSO) regression algorithms with identified 16 influencing factors. Results Results demonstrated that the CHD model using the RF classifier performed best both on the training set and test set, with the highest area under the curve (AUC = 1 and 0.9711), accuracy (one and 0.9389), sensitivity (one and 0.8725), specificity (one and 0.9771), precision (one and 0.9563), F1-score (one and 0.9125), and Matthews correlation coefficient (MCC = one and 0.8678), followed by the SVM (AUC = 0.9860 and 0.9589) and the LASSO classifier (AUC = 0.9733 and 0.9587). Besides, the RF model also had an increase in the net reclassification index (NRI) and integrated discrimination improvement (IDI) values, and achieved a greater net benefit in the decision curve analysis (DCA) compared with the SVM and LASSO models. Conclusion The CHD risk prediction model constructed by the RF algorithm in this study is conducive to the early diagnosis of CHD in rural residents of Xinxiang County, Henan Province.

Esophageal cancer (ESCA) carries a poor prognosis among gastrointestinal malignancies. The present study developed a signature based on mRNAs and long non-coding RNAs (lncRNAs) to predict prognosis in ESCA by using The Cancer Genome Atlas database. By using least absolute shrinkage and selection operator penalized regression, a set of RNAs (three mRNAs and two lncRNAs) was identified and used to build a risk score system of ESCA prognosis, which was used to stratify patients having considerable diverse survival in the training set [hazard ratio (HR), 3.932; 95% CI, 1.555–9.944; P

Multidimensional computerized adaptive testing (MCAT) is one of the widely discussed topics in psychometrics. Within the context of item replenishment in MCAT, it is important to identify the item-trait pattern for each replenished item, which indicates the set of the latent traits that are measured by each replenished item in the item pool. We propose a pattern recognition method based on the least absolute shrinkage and selection operator (LASSO) to detect the optimal item-trait patterns of the replenished items via an MCAT test. Simulation studies are conducted to investigate the performance of the proposed method in pattern recognition accuracy under different conditions across various latent trait correlation, item discrimination, test lengths, and item selection criteria in the test. Results show that the proposed method can accurately and efficiently identify the item-trait patterns of the replenished items in both the two-dimensional and three-dimensional item pools.

IntroductionDespite the decline in infant and under-five mortality rates since the last decade, Ghana did not meet the millennium development goal (MDG) 4 target. To implement effective interventions that could fast-track progress towards achieving the sustainable development goal 3 in 2030, factors contributing to the decline in child mortality throughout the MDG period and which factor(s) has/have been consistent in affecting child survival in the last decade need to be understood.MethodsThis study used Demographic and Health Surveys (DHS) from 2003, 2008 and 2014 and data from World Bank Development Indicators (2000–2018). We employed modified Poisson with robust SE and multivariate decomposition approach to assess risk factors of child mortality using DHS data from 2003, 2008 and 2014. Penalised regression was used assess the effect of 25 country-level contextual factors on child survival.ResultsThe risk of infant mortality is approximately five times higher among mothers who had multiple births compared with mothers who had single birth over the last decade (adjusted relative risk 4.6, 95% CI 3.2 to 6.6, pConclusionsThis study found that multiple births and shorter birth spacing are associated with increased risk of infant and under-five deaths over the last decade. Increased in FLFP, and the proportion of children sleeping under bed-net are associated with reduced risk of both infants and under-five deaths.

Background Tardive dyskinesia (TD) is a serious, often irreversible movement disorder caused by prolonged exposure to antipsychotics; identifying patients at risk for TD is critical to preventing it. Predictive models for the occurrence of TD can improve patient monitoring and inform implementation of counteractive interventions. This study aims to identify risk factors associated with TD and to develop a model using a retrospective data analysis to predict the incidence of TD among patients taking antipsychotic medications. Methods Adult patients with schizophrenia, major depressive disorder, or bipolar disorder taking oral antipsychotics were identified in a Medicaid claims database (covering six US states from 1997 to 2016) and divided into cohorts based on whether they developed TD within 1 year after the first observed claim for antipsychotics. Patient characteristics between cohorts were compared, and univariate Cox analyses were used to identify potential TD risk factors. A cross-validated version of the least absolute shrinkage and selection operator regression method was used to develop a parsimonious multivariable Cox proportional hazards model to predict diagnosis of TD. Results A total of 189,415 eligible patients were identified. Potential TD risk factors were identified based on the cohort analysis within a sample of 151,280 patients with at least 1 year of continuous eligibility. The prediction model had a clinically meaningful concordance of 70% and was well calibrated (P = 0.32 for Hosmer–Lemeshow goodness-of-fit test). Age (hazard ratio [HR] = 1.04, P

Background Hepatocellular carcinoma (HCC) is one of the most aggressive malignant tumors, with a poor long-term prognosis worldwide. The functional deregulations of global transcriptome were associated with the genesis and development of HCC, but lacks systematic research and validation. Methods A total of 519 postoperative HCC patients were included. We built an interactive and visual competing endogenous RNA network. The prognostic signature was established with the least absolute shrinkage and selection operator algorithm. Multivariate Cox regression analysis was used to screen for independent prognostic factors for HCC overall survival. Results In the training set, we identified a four-gene signature (PBK, CBX2, CLSPN, and CPEB3) and effectively predicted the overall survival. The survival times of patients in the high-score group were worse than those in the low-score group (p = 0.0004), and death was also more likely in the high-score group (HR 2.444, p

There are many prognostic gene signature models in clear cell renal cell carcinoma (ccRCC). However, different results from various methods and samples are hard to contribute to clinical practice. It is necessary to develop a robust gene signature for improving clinical practice in ccRCC. A method was proposed to integrate least absolute shrinkage and selection operator and multiple Cox regression to obtain mRNA and microRNA signature from the cancer genomic atlas database for predicting prognosis of ccRCC. The gene signature model consisted by 5 mRNAs and 1 microRNA was identified. Prognosis index (PI) model was constructed from RNA expression and median value of PI is used to classified patients into high- and low-risk groups. The results showed that high-risk patients showed significantly decrease survival comparison with low-risk groups [hazard ratio (HR) =7.13, 95% confidence interval = 3.71–13.70, P