Split Set Research Articles

Artificial intelligence facial recognition of obstructive sleep apnea: a Bayesian meta-analysis.

Conventional obstructive sleep apnea (OSA) diagnosis via polysomnography can be costly and inaccessible. Recent advances in artificial intelligence (AI) have enabled the use of craniofacial photographs to diagnose OSA. This meta-analysis aims to clarify the diagnostic accuracy of this innovative approach. Two blinded reviewers searched PubMed, Embase, Scopus, Web of Science, and IEEE Xplore databases, then selected and graded the risk of bias of observational studies of adults (≥ 18 years) comparing the diagnostic performance of AI algorithms using craniofacial photographs, versus conventional OSA diagnostic criteria (i.e. apnea-hypopnea index [AHI]). Studies were excluded if they detected apneic events without diagnosing OSA. AI models evaluated with a random split test set or k-fold cross-validation were included in a Bayesian bivariate meta-analysis. From 5,147 records, 6 studies were included, containing 10 AI models trained/tested on 1,417/983 participants. The risk of bias was low. AI trained on craniofacial photographs achieved a pooled 84.9% sensitivity (95% credible interval [95% CrI]: 77.1-90.7%) and 71.2% specificity (95% CrI: 60.7-81.4%). Bayesian meta-regression identified deep learning (convolutional neural networks) as the most accurate AI algorithm (91.1% sensitivity, 79.2% specificity) comparable to home sleep apnea tests. AHI cutoffs, OSA prevalence, feature engineering, input data, camera type and informativeness of Bayesian prior did not alter diagnostic accuracy. There was no substantial publication bias. AI trained on craniofacial photographs have high diagnostic accuracy and should be considered as a low-cost OSA screening tool. Future work focused on deep learning using smartphone images could improve the feasibility of this approach in primary care.

Synthesis of tamsulosin precursor through palladium nanocatalytic hydrogenolysis: Crystal structure, DFT analysis, and ADME studies

The present work describes a potential method for the synthesis of industrially important Tamsulosine intermediate under milder reaction conditions. The major advantages include i) nano catalytic system and ii) ammonium formate is used as greener alternative during hydrogenolysis with an overall yield of 98 %. Density functional theory (DFT) studies in combination with B3LYP exchange-correlation functional method with split valence basis set 6–311 G is used to understand molecular structure of intermediate is also described. The crystal structure of tamsulosine precursor is elucidated by using single crystal x-ray crystallography is also reported. Therefore, this study is expected to bridge the need to optimize a greener alternative process in this hydrogenolysis reaction to synthesize tamsulosin precursor. ADME studies reveals that better oral bioavailability, aqueous solubility, Gastrointerstinal absorption and option “yes” in Lipinski, Ghose, Veber, Egan and Muegge rules.

Data-adaptive binary classifiers in high dimensions using random partitioning

Classification in high dimensions has been highlighted for the past two decades since Fisher's linear discriminant analysis (LDA) is not optimal in a smaller sample size n comparing the number of covariates p, i.e. p>n, which is mostly due to the singularity of the sample covariance matrix. Rather than modifying how to estimate the sample covariance and sample mean vector in constructing a classifier, we build two types of high-dimensional classifiers using data splitting, i.e. single data splitting (SDS) and multiple data splitting (MDS). Moreover, we introduce a weighted version of MDS classifier that improves classification performance as illustrated in numerical studies. Each of the split data sets has a smaller size of covariates compared to the sample size so that LDA is applicable, and classification results can be combined with respect to minimizing the misclassification rate. We present theoretical justification backing up our proposed methods by comparing misclassification rates with LDA in high dimension. We also conduct a wide range of simulations and analyse four microarray data sets, which demonstrates that our proposed methods outperform some existing methods or at least yield comparable performances.

Predictive etiological classification of acute ischemic stroke through interpretable machine learning algorithms: a multicenter, prospective cohort study

BackgroundThe prognosis, recurrence rates, and secondary prevention strategies varied significantly among different subtypes of acute ischemic stroke (AIS). Machine learning (ML) techniques can uncover intricate, non-linear relationships within medical data, enabling the identification of factors associated with etiological classification. However, there is currently a lack of research utilizing ML algorithms for predicting AIS etiology.ObjectiveWe aimed to use interpretable ML algorithms to develop AIS etiology prediction models, identify critical factors in etiology classification, and enhance existing clinical categorization.MethodsThis study involved patients with the Third China National Stroke Registry (CNSR-III). Nine models, which included Natural Gradient Boosting (NGBoost), Categorical Boosting (CatBoost), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Light Gradient Boosting Machine (LGBM), Gradient Boosting Decision Tree (GBDT), Adaptive Boosting (AdaBoost), Support Vector Machine (SVM), and logistic regression (LR), were employed to predict large artery atherosclerosis (LAA), small vessel occlusion (SVO), and cardioembolism (CE) using an 80:20 randomly split training and test set. We designed an SFS-XGB with 10-fold cross-validation for feature selection. The primary evaluation metrics for the models included the area under the receiver operating characteristic curve (AUC) for discrimination and the Brier score (or calibration plots) for calibration.ResultsA total of 5,213 patients were included, comprising 2,471 (47.4%) with LAA, 2,153 (41.3%) with SVO, and 589 (11.3%) with CE. In both LAA and SVO models, the AUC values of the ML models were significantly higher than that of the LR model (P < 0.001). The optimal model for predicting SVO (AUC [RF model] = 0.932) outperformed the optimal LAA model (AUC [NGB model] = 0.917) and the optimal CE model (AUC [LGBM model] = 0.846). Each model displayed relatively satisfactory calibration. Further analysis showed that the optimal CE model could identify potential CE patients in the undetermined etiology (SUE) group, accounting for 1,900 out of 4,156 (45.7%).ConclusionsThe ML algorithm effectively classified patients with LAA, SVO, and CE, demonstrating superior classification performance compared to the LR model. The optimal ML model can identify potential CE patients among SUE patients. These newly identified predictive factors may complement the existing etiological classification system, enabling clinicians to promptly categorize stroke patients’ etiology and initiate optimal strategies for secondary prevention.

Development and validation of a population-based risk algorithm for premature mortality in Canada: the Premature Mortality Population Risk Tool (PreMPoRT)

IntroductionTo develop and validate the Premature Mortality Population Risk Tool (PreMPoRT), a population-based risk algorithm that predicts the 5-year incidence of premature mortality among the Canadian adult population.MethodsRetrospective cohort analysis...

Immune-based Machine learning Prediction of Diagnosis and Illness State in Schizophrenia and Bipolar Disorder

BackgroundSchizophrenia and bipolar disorder frequently face significant delay in diagnosis, leading to being missed or misdiagnosed in early stages. Both disorders have also been associated with trait and state immune abnormalities. Recent machine learning-based studies have shown encouraging results using diagnostic biomarkers in predictive models, but few have focused on immune-based markers. Our main objective was to develop supervised machine learning models to predict diagnosis and illness state in schizophrenia and bipolar disorder using only a panel of peripheral kynurenine metabolites and cytokines. MethodsThe cross-sectional I-GIVE cohort included hospitalized acute bipolar patients (n = 205), stable bipolar outpatients (n = 116), hospitalized acute schizophrenia patients (n = 111), stable schizophrenia outpatients (n = 75) and healthy controls (n = 185). Serum kynurenine metabolites, namely tryptophan (TRP), kynurenine (KYN), kynurenic acid (KA), quinaldic acid (QUINA), xanthurenic acid (XA), quinolinic acid (QUINO) and picolinic acid (PICO) were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS), while V-plex Human Cytokine Assays were used to measure cytokines (interleukin-6 (IL-6), IL-8, IL-17, IL-12/IL23-P40, tumor necrosis factor-alpha (TNF-ɑ), interferon-gamma (IFN-γ)). Supervised machine learning models were performed using JMP Pro 17.0.0. We compared a primary analysis using nested cross-validation to a split set as sensitivity analysis. Post-hoc, we re-ran the models using only the significant features to obtain the key markers. ResultsThe models yielded a good Area Under the Curve (AUC) (0.804, Positive Prediction Value (PPV) = 86.95; Negative Prediction Value (NPV) = 54.61) for distinguishing all patients from controls. This implies that a positive test is highly accurate in identifying the patients, but a negative test is inconclusive. Both schizophrenia patients and bipolar patients could each be separated from controls with a good accuracy (SCZ AUC 0.824; BD AUC 0.802). Overall, increased levels of IL-6, TNF-ɑ and PICO and decreased levels of IFN-γ and QUINO were predictive for an individual being classified as a patient. Classification of acute versus stable patients reached a fair AUC of 0.713. The differentiation between schizophrenia and bipolar disorder yielded a poor AUC of 0.627. ConclusionsThis study highlights the potential of using immune-based measures to build predictive classification models in schizophrenia and bipolar disorder, with IL-6, TNF-ɑ, IFN-γ, QUINO and PICO as key candidates. While machine learning models successfully distinguished schizophrenia and bipolar disorder from controls, the challenges in differentiating schizophrenic from bipolar patients likely reflect shared immunological pathways by the both disorders and confounding by a larger state-specific effect. Larger multi-centric studies and multi-domain models are needed to enhance reliability and translation into clinic.

Multicellular target QSAR models for predicting of novel inhibitors against pancreatic cancer by Monte Carlo approach

Pancreatic cancer is one of the most fatal malignant tumors, with a poor prognosis. Nowadays, computational methods for designing and predicting anticancer agents have gained increased attention. In this work, a QSAR study is conducted to further investigate the activity of 5797 agents against pancreatic cancer cell lines. QSAR models are a system of classification approaches that are used to construct and validate the predictive potential of created models. This approach demonstrates the prediction in the active (1) and inactive (−1) format. Here, the classification QSAR technique is employed to construct and evaluate models. The classification QSAR model has been constructed by the Monte Carlo method based on the representation of the molecular structure by SMILES (simplified molecular input line entry systems) using the CORAL software. Four random splits of data by making four sets such as active training, passive training, calibration, and validation sets are assessed. The numerical values of sensitivity, specificity, and accuracy for the validation set of splits 1 to 4 are in the range of 0.7726–0.7901, 0.9007–0.9220, and 0.8548–0.8730 respectively. The Matthews correlation coefficient (MCC) of these models for external validation sets is found to be in the range of 0.6832 to 0.7215.

Phylogenetic network-assisted rooting of unrooted gene trees

Gene trees inferred from molecular sequence alignments are typically unrooted, and determining the most credible rooting edge is a classical problem in computational biology. One approach to solve this problem is unrooted reconciliation, where the rooting edge is postulated based on the split of the root from a given species tree. In this paper, we propose a novel variant of the gene tree rooting problem, where the gene tree root is inferred using a phylogenetic network of the species present in the gene tree. To obtain the best rooting, unrooted reconciliation can be applied, where the unrooted gene tree is jointly reconciled with a set of splits inferred from the network. However, the exponential size of the set induced by display trees of the network makes this approach computationally prohibitive. To address this, we propose a broader and easier-to-control set of splits based on the structural properties of the network. We then derive exact mathematical formulas for the rooting problem and propose two general rooting algorithms to handle cases where the input network does not meet the initial requirements. Our experimental study based on simulated gene trees and networks demonstrates that our algorithms infer gene tree rootings correctly or with a small error in most cases.

Split Detour Monophonic Sets in Graph

A subset T ⊆ V is a detourmonophonic set of G if each node (vertex) x in G contained in an p-q detourmonophonic path where p, q ∈ T.. The number of points in a minimum detourmonophonic set of G is called as the detourmonophonic number of G, dm(G). A subset T ⊆ V of a connected graph G is said to be a split detourmonophonic set of G if the set T of vertices is either T = V or T is detoumonophonic set and V – T induces a subgraph in which is disconnected. The minimum split detourmonophonic set is split detourmonophonic set with minimum cardinality and it is called a split detourmonophonic number, denoted by dms(G). For certain standard graphs, defined new parameter was identified. Some of the realization results on defined new parameters were established.

Machine-Learning-Based Classification Model to Address Diagnostic Challenges in Transbronchial Lung Biopsy.

When obtaining specimens from pulmonary nodules in TBLB, distinguishing between benign samples and mis-sampling from a tumor presents a challenge. Our objective is to develop a machine-learning-based classifier for TBLB specimens. Three pathologists assessed six pathological findings, including interface bronchitis/bronchiolitis (IB/B), plasma cell infiltration (PLC), eosinophil infiltration (Eo), lymphoid aggregation (Ly), fibroelastosis (FE), and organizing pneumonia (OP), as potential histologic markers to distinguish between benign and malignant conditions. A total of 251 TBLB cases with defined benign and malignant outcomes based on clinical follow-up were collected and a gradient-boosted decision-tree-based machine learning model (XGBoost) was trained and tested on randomly split training and test sets. Five pathological changes showed independent, mild-to-moderate associations (AUC ranging from 0.58 to 0.75) with benign conditions, with IB/B being the strongest predictor. On the other hand, FE emerged to be the sole indicator of malignant conditions with a mild association (AUC = 0.66). Our model was trained on 200 cases and tested on 51 cases, achieving an AUC of 0.78 for the binary classification of benign vs. malignant on the test set. The machine-learning model developed has the potential to distinguish between benign and malignant conditions in TBLB samples excluding the presence or absence of tumor cells, thereby improving diagnostic accuracy and reducing the burden of repeated sampling procedures for patients.

A study protocol for a predictive model to assess population-based avoidable hospitalization risk: Avoidable Hospitalization Population Risk Prediction Tool (AvHPoRT)

IntroductionAvoidable hospitalizations are considered preventable given effective and timely primary care management and are an important indicator of health system performance. The ability to predict avoidable hospitalizations at the population level represents a significant advantage for health system decision-makers that could facilitate proactive intervention for ambulatory care-sensitive conditions (ACSCs). The aim of this study is to develop and validate the Avoidable Hospitalization Population Risk Tool (AvHPoRT) that will predict the 5-year risk of first avoidable hospitalization for seven ACSCs using self-reported, routinely collected population health survey data.Methods and analysisThe derivation cohort will consist of respondents to the first 3 cycles (2000/01, 2003/04, 2005/06) of the Canadian Community Health Survey (CCHS) who are 18–74 years of age at survey administration and a hold-out data set will be used for external validation. Outcome information on avoidable hospitalizations for 5 years following the CCHS interview will be assessed through data linkage to the Discharge Database (1999/2000–2017/2018) for an estimated sample size of 394,600. Candidate predictor variables will include demographic characteristics, socioeconomic status, self-perceived health measures, health behaviors, chronic conditions, and area-based measures. Sex-specific algorithms will be developed using Weibull accelerated failure time survival models. The model will be validated both using split set cross-validation and external temporal validation split using cycles 2000–2006 compared to 2007–2012. We will assess measures of overall predictive performance (Nagelkerke R2), calibration (calibration plots), and discrimination (Harrell’s concordance statistic). Development of the model will be informed by the Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) statement.Ethics and disseminationThis study was approved by the University of Toronto Research Ethics Board. The predictive algorithm and findings from this work will be disseminated at scientific meetings and in peer-reviewed publications.

Predictors of nonpulmonary vein triggers for atrial fibrillation: A clinical risk score

BackgroundTargeting non-pulmonary vein triggers (NPVTs) after pulmonary vein isolation may reduce atrial fibrillation (AF) recurrence. Isoproterenol infusion and cardioversion of spontaneous or induced AF can provoke NPVTs but typically require vasopressor support and increased procedural time. ObjectiveThe purpose of this study was to identify risk factors for the presence of NPVTs and create a risk score to identify higher-risk subgroups. MethodsUsing the AF ablation registry at the Hospital of the University of Pennsylvania, we included consecutive patients who underwent AF ablation between January 2021 and December 2022. We excluded patients who did not receive NPVT provocation testing after failing to demonstrate spontaneous NPVTs. NPVTs were defined as non-pulmonary vein ectopic beats triggering AF or focal atrial tachycardia. We used risk factors associated with NPVTs with P <.1 in multivariable logistic regression model to create a risk score in a randomly split derivation set (80%) and tested its predictive accuracy in the validation set (20%). ResultsIn 1530 AF ablations included, NPVTs were observed in 235 (15.4%). In the derivation set, female sex (odds ratio [OR] 1.40; 95% confidence interval [CI] 0.96–2.03; P = .080), sinus node dysfunction (OR 1.67; 95% CI 0.98–2.87; P = .060), previous AF ablation (OR 2.50; 95% CI 1.70–3.65; P <.001), and left atrial scar (OR 2.90; 95% CI 1.94–4.36; P <.001) were risk factors associated with NPVTs. The risk score created from these risk factors (PRE2SSS2 score; [PRE]vious ablation: 2 points, female [S]ex: 1 point, [S]inus node dysfunction: 1 point, left atrial [S]car: 2 points) had good predictive accuracy in the validation cohort (area under the receiver operating characteristic curve 0.728; 95% CI 0.648–0.807). ConclusionA risk score incorporating predictors for NPVTs may allow provocation of triggers to be performed in patients with greatest expected yield.

Impact of a Divided Phenobarbital Load and Taper Compared With Lorazepam Symptom Triggered Therapy in Hospitalized Patients.

Benzodiazepines are the preferred treatment for alcohol withdrawal. Phenobarbital is an alternative in the setting of prescriber expertise or benzodiazepine contraindication. To evaluate the efficacy and safety of a phenobarbital dosing strategy aimed at treating a spectrum of alcohol withdrawal symptoms across various patient populations. Retrospective review of patients admitted with concerns of alcohol withdrawal between May 2018 and November 2022. Patients were separated into a before-after cohort of lorazepam or phenobarbital. The primary outcome was hospital length of stay (LOS). Secondary outcomes were intensive care unit (ICU) LOS, escalation of respiratory support, increased level of care (LOC), and incidence of delirium tremens and/or seizures. Two hundred and seventy-seven patients received lorazepam and 198 received phenobarbital. Hospital LOS was longer in the phenobarbital cohort compared with the lorazepam cohort (6.9 vs 9.3 days). There was no difference in ICU LOS. Level of care increases were fewer in the phenobarbital cohort (4 events vs 19 events). There were higher rates of non-invasive respiratory interventions in the lorazepam cohort and higher rates of mechanical ventilation in the phenobarbital cohort. Utilization of phenobarbital was attributed to a reduction in delirium tremens and seizures. This study is novel because of the broad application of a phenobarbital order set across multiple levels of care and patient admission diagnoses. A risk targeted split load intravenous phenobarbital order set can safely be administered to patients with fewer escalations of care, seizures, delirium tremens, and respiratory care escalation.

Longitudinal clinical data improve survival prediction after hematopoietic cell transplantation using machine learning

AbstractSerial prognostic evaluation after allogeneic hematopoietic cell transplantation (allo-HCT) might help identify patients at high risk of lethal organ dysfunction. Current prediction algorithms based on models that do not incorporate changes to patients’ clinical condition after allo-HCT have limited predictive ability. We developed and validated a robust risk-prediction algorithm to predict short- and long-term survival after allo-HCT in pediatric patients that includes baseline biological variables and changes in the patients’ clinical status after allo-HCT. The model was developed using clinical data from children and young adults treated at a single academic quaternary-care referral center. The model was created using a randomly split training data set (70% of the cohort), internally validated (remaining 30% of the cohort) and then externally validated on patient data from another tertiary-care referral center. Repeated clinical measurements performed from 30 days before allo-HCT to 30 days afterwards were extracted from the electronic medical record and incorporated into the model to predict survival at 100 days, 1 year, and 2 years after allo-HCT. Naïve-Bayes machine learning models incorporating longitudinal data were significantly better than models constructed from baseline variables alone at predicting whether patients would be alive or deceased at the given time points. This proof-of-concept study demonstrates that unlike traditional prognostic tools that use fixed variables for risk assessment, incorporating dynamic variability using clinical and laboratory data improves the prediction of mortality in patients undergoing allo-HCT.

SIMPD: an algorithm for generating simulated time splits for validating machine learning approaches

Time-split cross-validation is broadly recognized as the gold standard for validating predictive models intended for use in medicinal chemistry projects. Unfortunately this type of data is not broadly available outside of large pharmaceutical research organizations. Here we introduce the SIMPD (simulated medicinal chemistry project data) algorithm to split public data sets into training and test sets that mimic the differences observed in real-world medicinal chemistry project data sets. SIMPD uses a multi-objective genetic algorithm with objectives derived from an extensive analysis of the differences between early and late compounds in more than 130 lead-optimization projects run within the Novartis Institutes for BioMedical Research. Applying SIMPD to the real-world data sets produced training/test splits which more accurately reflect the differences in properties and machine-learning performance observed for temporal splits than other standard approaches like random or neighbor splits. We applied the SIMPD algorithm to bioactivity data extracted from ChEMBL and created 99 public data sets which can be used for validating machine-learning models intended for use in the setting of a medicinal chemistry project. The SIMPD code and simulated data sets are available under open-source/open-data licenses at github.com/rinikerlab/molecular_time_series.

Deep learning myocardial infarction segmentation framework from cardiac magnetic resonance images

Segmentation of myocardial infarction (MI) is a crucial task in the field of heart disease theranostics. Cardiac magnetic resonance imaging (MRI) is a well-known non-invasive imaging technique that provides comprehensive insights into the structure and function of the heart. However, manually interpreting myocardial infarction from multiple MRI frames is time-consuming, labor-intensive, and prone to errors. This study aims to develop an end-to-end deep learning framework that can automatically segment myocardial infarction (MI) and persistent microvascular obstruction (MVO) among the normal tissues of left ventricle (LV), normal myocardium (Myo), and the remaining normal foreground (BG). The proposed framework includes various stages, such as cardiac MR image collection, preprocessing via three enhancement techniques, splitting and training set augmentation, selection of the most suitable artificial intelligence (AI) segmentation model, and performance evaluation and comparison. For the multi-class segmentation process, we adopt and develope four AI state-of-the-art models: U-Net, U-Net_VGG16, SegNet, and ResUnet, which are well-regarded for their effectiveness in image segmentation across different computer vision domains. The publicly available benchmark EMIDEC MRI dataset is utilized for training and evaluating the proposed segmentation framework. The ResU-Net achieved the top performance compared to other AI models, recording an overall accuracy (Acc) of 88.48%, recall (Re) of 85.24%, precision (Pre) of 85.46%, F1-score of 85.35, and MIoU of 84.23%. Comparing with the original dataset (without preprocessing), the CLAHE preprocessing improves the ResU-Net segmentation performance by 2.19% and 3.08% in terms of average F1-score and MIoU for all classes (LV, Myo, MI, and MVO). Therefore, the proposed AI segmentation framework demonstrates its potential for effectively performing multi-class segmentation of cardiac diseases from MRI images.

Способ и устройство определения несущей способности фрикционных анкеров

In underground mining, surface reinforcement is a mandatory part of the works. In addition to frame supports, rock bolts and in some cases a combination of the two are now used. The Split Set® friction bolts, referred to as the FRS, i.e. Friction Rock Stabilizers, in the foreign literature, are actively used. More than 11 companies in Russia are currently officially engaged in production of the friction bolts. At the same time, there is a significant difference in the properties of steel used in production of the bolt shafts. Foreign manufacturers recommend the use of steel with the yield strength of 350–400 MPa. This parameter determines the main characteristic of the FRS, i.e. the load-bearing capacity, which can be assessed only after the friction bolt is introduced in the borehole. Until recently, no way has been available to control this parameter of the standard rock bolt. This is explained by the specific feature of the FRS design, i.e. the absence of projecting parts. The load-bearing capacity is determined using specially prepared rock bolts. The purpose of the research was to develop a method and equipment for determining the bearing capacity of the standard FRS bolt using a certified rock bolt tester. The "expanding collet" method of gripping the bolt shaft was adopted. The design of the loading device was based on interpretation of the structural analysis method for mechanical devices. The parametric synthesis of the device elements was carried out on the basis of experimental trials. Experimental and industrial trials of the device confirmed the possibility of determining the load bearing capacity of any FRS bolt. The implementation of the device makes it possible to test the bearing capacity without disturbing the integrity of the excavation and the rock bolt itself.

A multi-day waste collection and transportation problem with selective collection and split delivery

This paper addresses a multi-day waste collection and transportation problem with selective collection and split delivery (MDWCTP-SCSD). Rather than emptying waste according to a fixed schedule, garbage trucks only visit community waste collection sites whose waste level reaches a predetermined threshold, reducing collection costs but increasing overflow risk. Collection sites likely to overflow the next day if unvisited today are also selected for preventive collection. Split delivery is permitted. The MDWCTP-SCSD is formulated as a mixed integer linear programming model. To solve this NP-hard problem, a heuristic combining variable neighborhood search and simulated annealing is proposed. In addition to classical operators, three new operators are developed to handle split delivery. Experiments demonstrate the proposed algorithm is competitive in solving two split delivery vehicle routing benchmark sets; outperforms CPLEX in solving small-scale instances within an hour; and quickly solves large-scale instances with high quality. Noteworthy findings indicate fleet costs and size can be reduced by approximately 50% if only visiting sites whose waste level reaches 80%.

Research on Apple Origins Classification Optimization Based on Least-Angle Regression in Instance Selection

Machine learning is used widely in near-infrared spectroscopy (NIRS) for fruit qualification. However, the directly split training set used contains redundant samples, and errors may be introduced into the model. Euclidean distance-based and K-nearest neighbor-based instance selection (IS) methods are widely used to remove useless samples because of their accessibility. However, they either have high accuracy and low compression or vice versa. To compress the sample size while improving the accuracy, the least-angle regression (LAR) method was proposed for classification instance selection, and a discrimination experiment was conducted on a total of four origins of 952 apples. The sample sets were split into the raw training set and testing set; the optimal training samples were selected using the LAR-based instance selection (LARIS) method, and the four other selection methods were compared. The results showed that 26.9% of the raw training samples were selected using LARIS, and the model based on these training samples had the highest accuracy. Thus, the apple origin classification model based on LARIS can achieve the goal of high accuracy and compression and provide experimental support for the least-angle regression algorithm in classification instance selection.

Combinatorial Use of Machine Learning and Logistic Regression for Predicting Carotid Plaque Risk Among 5.4 Million Adults With Fatty Liver Disease Receiving Health Check-Ups: Population-Based Cross-Sectional Study.

Carotid plaque can progress into stroke, myocardial infarction, etc, which are major global causes of death. Evidence shows a significant increase in carotid plaque incidence among patients with fatty liver disease. However, unlike the high detection rate of fatty liver disease, screening for carotid plaque in the asymptomatic population is not yet prevalent due to cost-effectiveness reasons, resulting in a large number of patients with undetected carotid plaques, especially among those with fatty liver disease. This study aimed to combine the advantages of machine learning (ML) and logistic regression to develop a straightforward prediction model among the population with fatty liver disease to identify individuals at risk of carotid plaque. Our study included 5,420,640 participants with fatty liver from Meinian Health Care Center. We used random forest, elastic net (EN), and extreme gradient boosting ML algorithms to select important features from potential predictors. Features acknowledged by all 3 models were enrolled in logistic regression analysis to develop a carotid plaque prediction model. Model performance was evaluated based on the area under the receiver operating characteristic curve, calibration curve, Brier score, and decision curve analysis both in a randomly split internal validation data set, and an external validation data set comprising 32,682 participants from MJ Health Check-up Center. Risk cutoff points for carotid plaque were determined based on the Youden index, predicted probability distribution, and prevalence rate of the internal validation data set to classify participants into high-, intermediate-, and low-risk groups. This risk classification was further validated in the external validation data set. Among the participants, 26.23% (1,421,970/5,420,640) were diagnosed with carotid plaque in the development data set, and 21.64% (7074/32,682) were diagnosed in the external validation data set. A total of 6 features, including age, systolic blood pressure, low-density lipoprotein cholesterol (LDL-C), total cholesterol, fasting blood glucose, and hepatic steatosis index (HSI) were collectively selected by all 3 ML models out of 27 predictors. After eliminating the issue of collinearity between features, the logistic regression model established with the 5 independent predictors reached an area under the curve of 0.831 in the internal validation data set and 0.801 in the external validation data set, and showed good calibration capability graphically. Its predictive performance was comprehensively competitive compared with the single use of either logistic regression or ML algorithms. Optimal predicted probability cutoff points of 25% and 65% were determined for classifying individuals into low-, intermediate-, and high-risk categories for carotid plaque. The combination of ML and logistic regression yielded a practical carotid plaque prediction model, and was of great public health implications in the early identification and risk assessment of carotid plaque among individuals with fatty liver.