Inverse attribute‐based optimization with an application in assortment optimization

Background Safety issues for fluoroquinolones have been provided by regulatory agencies. This study was conducted to identify signals of fluoroquinolones reported in the Korea Adverse Event Reporting System (KAERS) using tree-based machine learning (ML) methods. Research design and methods All adverse events (AEs) associated with the target drugs reported in the KAERS from 2013 to 2017 were matched with drug label information. A dataset containing label-positive and -negative AEs was arbitrarily divided into training and test sets. Decision tree, random forest (RF), bagging, and gradient boosting machine (GBM) were fitted on the training set with hyperparameters tuned using five-fold cross-validation and applied to the test set. The ML method with the highest area under the curve (AUC) scores was selected as the final ML model. Results Bagging was selected as the final ML model for gemifloxacin (AUC score: 1) and levofloxacin (AUC: 0.9987). RF was selected in ciprofloxacin, moxifloxacin, and ofloxacin (AUC scores: 0.9859, 0.9974, and 0.9999 respectively). We found that the final ML methods detected additional signals that were not detected using the disproportionality analysis (DPA) methods. Conclusions The bagging-or-RF-based ML methods performed better than DPA and detected novel AE signals previously unidentified using the DPA methods.

This paper presents a two-stage stochastic programming model for retail assortment optimisation. Decisions related to what products to stock and in what quantity are taken in the first stage while decisions related to what products to use to satisfy the primary versus the secondary demand are taken in the second stage. Such a model is very useful for online retailers and catalogue merchants who witness a gap between an order arrival and fulfilment. A case study illustrating the modelling approach and its benefits as compared to other existing assortment optimisation techniques is also presented.

Background Predicting the prognosis of patients with acute myocardial infarction (AMI) combined with diabetes mellitus (DM) is crucial due to high in-hospital mortality rates. This study aims to develop and validate a mortality risk prediction model for these patients by interpretable machine learning (ML) methods. Methods Data were sourced from the Medical Information Mart for Intensive Care IV (MIMIC-IV, version 2.2). Predictors were selected by Least absolute shrinkage and selection operator (LASSO) regression and checked for multicollinearity with Spearman’s correlation. Patients were randomly assigned to training and validation sets in an 8:2 ratio. Seven ML algorithms were used to construct models in the training set. Model performance was evaluated in the validation set using metrics such as area under the curve (AUC) with 95% confidence interval (CI), calibration curves, precision, recall, F1 score, accuracy, negative predictive value (NPV), and positive predictive value (PPV). The significance of differences in predictive performance among models was assessed utilising the permutation test, and 10-fold cross-validation further validated the model’s performance. SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) were applied to interpret the models. Results The study included 2,828 patients with AMI combined with DM. Nineteen predictors were identified through LASSO regression and Spearman’s correlation. The Random Forest (RF) model was demonstrated the best performance, with an AUC of 0.823 (95% CI: 0.774–0.872), high precision (0.867), accuracy (0.873), and PPV (0.867). The RF model showed significant differences (p < 0.05) compared to the K-Nearest Neighbours and Decision Tree models. Calibration curves indicated that the RF model’s predicted risk aligned well with actual outcomes. 10-fold cross-validation confirmed the superior performance of RF model, with an average AUC of 0.828 (95% CI: 0.800–0.842). Significant Variables in RF model indicated that the top eight significant predictors were urine output, maximum anion gap, maximum urea nitrogen, age, minimum pH, maximum international normalised ratio (INR), mean respiratory rate, and mean systolic blood pressure. Conclusion This study demonstrates the potential of ML methods, particularly the RF model, in predicting in-hospital mortality risk for AMI patients with DM. The SHAP and LIME methods enhance the interpretability of ML models.

BackgroundDespite the large neuroanatomic measurements obtained from brain MRI segmentation, we do not have yet a definitive neuroanatomic or vascular feature/even that can precisely delineate a cognitively normal subject from cognitively impaired and Alzheimer’s disease (AD). Increasing lifespan of population accompanied with increased prevalence of cardiovascular risks across globe demands for precise early detection of AD and cognitive status (impaired or normal) in the aging individuals. Here, we have employed machine learning (ML) method on neuroanatomic volume and white matter hyperintensity (WMH) volume obtained from MRI analyses of data from National Alzheimer’s Coordinating Center (NACC) to identify participants as cognitively normal (CN), cognitively impaired with etiological diagnosis as AD and cognitively impaired (CI) subjects.MethodBrain segmentation outputs for neuroanatomic‐volume and White matter hyperintensity (WMH) was obtained from NACC for 1290 unique subjects with total of 3684 longitudinal‐MRI. Supervised ML was performed using the MRI parameters for identifying the subjects as AD, CI, and CN. 50% of the MRI were used as training set while another 50% was used as test. Various combinations of neuroanatomic volume with WMH quantity were tested for accuracy of delineating the AD, CI and CN groups using random forest method. Furthermore, various ML methods such as XGB classifiers, classification tree, bagging classifier and simple classification were tested for accuracy using the optimized neuroanatomic volume and WMH.ResultRandom forest method converged with a combination of 5‐MRI features, intracranial, gray‐matter, cerebral CSF, hippocampus, and White‐matter hyperintensity volume with highest accuracy (83.3%) in identifying AD, CI, and CN subjects (Fig. 1). Inclusion of normal appearing white‐matter volume during the ML optimization of MRI features led to significant reduction in the accuracy. Simple random forest ML method showed highest accuracy using the optimized MRI features (Fig. 2).ConclusionAll the neuroanatomic events obtained from brain MRI may not be specific in providing diagnostic features for evaluating brain health. Here we show that a combination of 5 MRI features that includes microvascular pathology and neuroanatomic volume may provide a specific and efficient non‐invasive imaging biomarker to evaluate brain health in aging population.

To determine the feasibility of predicting the rate of an axillary lymph node pathological complete response (apCR) using nomogram and machine learning methods. A total of 247 patients with early breast cancer (eBC), who underwent neoadjuvant therapy (NAT) were included retrospectively. We compared pre- and post-NAT ultrasound information and calculated the maximum diameter change of the primary lesion (MDCPL): [(pre-NAT maximum diameter of primary lesion - post-NAT maximum diameter of preoperative primary lesion)/pre-NAT maximum diameter of primary lesion] and described the lymph node score (LNS) (1): unclear border (2), irregular morphology (3), absence of hilum (4), visible vascularity (5), cortical thickness, and (6) aspect ratio <2. Each description counted as 1 point. Logistic regression analyses were used to assess apCR independent predictors to create nomogram. The area under the curve (AUC) of the receiver operating characteristic curve as well as calibration curves were employed to assess the nomogram's performance. In machine learning, data were trained and validated by random forest (RF) following Pycharm software and five-fold cross-validation analysis. The mean age of enrolled patients was 50.4 ± 10.2 years. MDCPL (odds ratio [OR], 1.013; 95% confidence interval [CI], 1.002-1.024; p=0.018), LNS changes (pre-NAT LNS - post-NAT LNS; OR, 2.790; 95% CI, 1.190-6.544; p=0.018), N stage (OR, 0.496; 95% CI, 0.269-0.915; p=0.025), and HER2 status (OR, 2.244; 95% CI, 1.147-4.392; p=0.018) were independent predictors of apCR. The AUCs of the nomogram were 0.74 (95% CI, 0.68-0.81) and 0.76 (95% CI, 0.63-0.90) for training and validation sets, respectively. In RF model, the maximum diameter of the primary lesion, axillary lymph node, and LNS in each cycle, estrogen receptor status, progesterone receptor status, HER2, Ki67, and T and N stages were included in the training set. The final validation set had an AUC value of 0.85 (95% CI, 0.74-0.87). Both nomogram and machine learning methods can predict apCR well. Nomogram is simple and practical, and shows high operability. Machine learning makes better use of a patient's clinicopathological information. These prediction models can assist surgeons in deciding on a reasonable strategy for axillary surgery.

BACKGROUNDTumor deposits (TDs) are not equivalent to lymph node (LN) metastasis (LNM) but have become independent adverse prognostic factors in patients with rectal cancer (RC). Although preoperatively differentiating TDs and LNMs is helpful in designing individualized treatment strategies and achieving improved prognoses, it is a challenging task.AIMTo establish a computed tomography (CT)-based radiomics model for preoperatively differentiating TDs from LNM in patients with RC.METHODSThis study retrospectively enrolled 219 patients with RC [TDs+LNM- (n = 89); LNM+ TDs- (n = 115); TDs+LNM+ (n = 15)] from a single center between September 2016 and September 2021. Single-positive patients (i.e., TDs+LNM- and LNM+TDs-) were classified into the training (n = 163) and validation (n = 41) sets. We extracted numerous features from the enhanced CT (region 1: The main tumor; region 2: The largest peritumoral nodule). After deleting redundant features, three feature selection methods and three machine learning methods were used to select the best-performing classifier as the radiomics model (Rad-score). After validating Rad-score, its performance was further evaluated in the field of diagnosing double-positive patients (i.e., TDs+LNM+) by outlining all peritumoral nodules with diameter (short-axis) > 3 mm.RESULTSRad-score 1 (radiomics signature of the main tumor) had an area under the curve (AUC) of 0.768 on the training dataset and 0.700 on the validation dataset. Rad-score 2 (radiomics signature of the largest peritumoral nodule) had a higher AUC (training set: 0.940; validation set: 0.918) than Rad-score 1. Clinical factors, including age, gender, location of RC, tumor markers, and radiological features of the largest peritumoral nodule, were excluded by logistic regression. Thus, the combined model was comprised of Rad-scores of 1 and 2. Considering that the combined model had similar AUCs with Rad-score 2 (P = 0.134 in the training set and 0.594 in the validation set), Rad-score 2 was used as the final model. For the diagnosis of double-positive patients in the mixed group [TDs+LNM+ (n = 15); single-positive (n = 15)], Rad-score 2 demonstrated moderate performance (sensitivity, 73.3%; specificity, 66.6%; and accuracy, 70.0%).CONCLUSIONRadiomics analysis based on the largest peritumoral nodule can be helpful in preoperatively differentiating between TDs and LNM.

The paper presents the results of experiments to assess the machine learning methods applicability for solving the problem of identifying argumentative connections in scientific communication texts. Argumentative connection is understood as a relationship that connects the premise and the conclusion of a typical reasoning or an argument used by the author to persuade the readers. To assess the quality, the characteristics of accuracy, completeness and F-measure were used obtained when solving the problem of recognizing argumentative connections between adjacent text fragments of two types: sentences and clauses. The basis of the experiment was a Russian-language corpus of texts from the field of scientific communication with arguments marked up by linguistic experts. For markup, the ArgNetBank Studio tool was used, which allows creating collections of texts with detailed argumentation markup. Data sets for machine learning were built on the basis of labeled texts, in which the ratio of pairs of text fragments (sentences or clauses) connected and non-connected by argumentative relationships was 1 to 3. To improve the quality of model training, the sets were balanced in two ways. In the first case, a balance was achieved due to the fact that an equal number of pairs of both types were selected from each text; in the second, pairs were duplicated. Using the obtained data sets, experiments were carried out on linking text fragments using different types of machine learning methods. The range of changes in quality assessments when recognizing related fragments depending on their share in the training and test collections was experimentally determined. It has been established that, within the framework of the existing imbalance in real collections, the values of quality assessments can vary within 40–50%. The novelty of the work lies in the study of the range of possible discrepancies in quality assessments when applying different machine learning methods on balanced and unbalanced training and test collections in Russian-language material.

This paper deals with stability analysis in multi-objective combinatorial optimization problems. The stability radius of an efficient solution is defined as the maximal adjustment of the problem parameters such that this solution remains efficient. An algorithm based on inverse optimization is proposed to compute it. The adjustment is limited to the coefficients of the objective functions and measured by the Chebyshev norm. This approach is applied to randomly generated instances of the bi-objective knapsack problem and computational results are reported. Several illustrative examples are analyzed.

Understanding and modeling of agent’s risk/reward preferences is a central problem in various applications of risk management including investment science and portfolio theory in particular. One of the approaches is to axiomatically define a set of performance measures and to use a benchmark to identify a particular measure from that set by either inverse optimization or functional dominance. For example, such a benchmark could be the rate of return of an existing attractive financial instrument. This work introduces deviation and drawdown measures that incorporate rates of return of indicated financial instruments (benchmarks). For discrete distributions and discrete sample paths, portfolio problems with such measures are reduced to linear programs and solved based on historical data in cases of a single benchmark and three benchmarks used simultaneously. The optimal portfolios and corresponding benchmarks have similar expected/cumulative rates of return in sample and out of sample, but the former are considerably less volatile.

BackgroundThe brain tumor is the growth of abnormal cells inside the brain. These cells can be grown into malignant or benign tumors. Segmentation of tumor from MRI images using image processing techniques started decades back. Image processing based brain tumor segmentation can be divided in to three categories conventional image processing methods, Machine Learning methods and Deep Learning methods. Conventional methods lacks the accuracy in segmentation due to complex spatial variation of tumor. Machine Learning methods stand as a good alternative to conventional methods. Methods like SVM, KNN, Fuzzy and a combination of either of these provide good accuracy with reasonable processing speed. The difficulty in processing the various feature extraction methods and maintain accuracy as per the medical standards still exist as a limitation for machine learning methods. In Deep Learning features are extracted automatically in various stages of the network and maintain accuracy as per the medical standards. Huge database requirement and high computational time is still poses a problem for deep learning. To overcome the limitations specified above we propose an unsupervised dual autoencoder with latent space optimization here. The model require only normal MRI images for its training thus reducing the huge tumor database requirement. With a set of normal class data, an autoencoder can reproduce the feature vector into an output layer. This trained autoencoder works well with normal data while it fails to reproduce an anomaly to the output layer. But a classical autoencoder suffer due to poor latent space optimization. The Latent space loss of classical autoencoder is reduced using an auxiliary encoder along with the feature optimization based on singular value decomposition (SVD). The patches used for training are not traditional square patches but we took both horizontal and vertical patches to keep both local and global appearance features on the training set. An Autoencoder is applied separately for learning both horizontal and vertical patches. While training a logistic sigmoid transfer function is used for both encoder and decoder parts. SGD optimizer is used for optimization with an initial learning rate of .001 and the maximum epochs used are 4000. The network is trained in MATLAB 2018a with a processor capacity of 3.7 GHz with NVIDIA GPU and 16 GB of RAM.ResultsThe results are obtained using a patch size of 16 × 64, 64 × 16 for horizontal and vertical patches respectively. In Glioma images tumor is not grown from a point rather it spreads randomly. Region filling and connectivity operations are performed to get the final tumor segmentation. Overall the method segments Meningioma better than Gliomas. Three evaluation metrics are considered to measure the performance of the proposed system such as Dice Similarity Coefficient, Positive Predictive Value, and Sensitivity.ConclusionAn unsupervised method for the segmentation of brain tumor from MRI images is proposed here. The proposed dual autoencoder with SVD based feature optimization reduce the latent space loss in the classical autoencoder. The proposed method have advantages in computational efficiency, no need of huge database requirement and better accuracy than machine learning methods. The method is compared Machine Learning methods Like SVM, KNN and supervised deep learning methods like CNN and commentable results are obtained.

Sepsis remains a crucial global health issue characterized by high mortality rates and a lack of specific treatments. This study aimed to elucidate the molecular mechanisms underlying sepsis and to identify potential therapeutic targets and compounds. High-throughput sequencing data from the GEO database (GSE26440 as the training set and GSE13904 and GSE32707 as the validation sets), weighted gene co-expression network analysis (WGCNA), Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis, alongside a combination of PPI and machine learning methods (LASSO and SVM) were utilized. WGCNA identified the black module as positively correlated, and the green module as negatively correlated with sepsis. Further intersections of these module genes with age-related genes yielded 57 sepsis-related genes. GO and KEGG pathway enrichment analysis, PPI, LASSO, and SVM selected six hub aging-related genes: BCL6, FOS, ETS1, ETS2, MAPK14, and MYC. A diagnostic model was constructed based on these six core genes, presenting commendable performance in both the training and validation sets. Notably, ETS1 demonstrated significant differential expression between mild and severe sepsis, indicating its potential as a biomarker of severity. Furthermore, immune infiltration analysis of these six core genes revealed their correlation with most immune cells and immune-related pathways. Additionally, compounds were identified in the traditional Chinese medicine Danshen, which upon further analysis, revealed 354 potential target proteins. GO and KEGG enrichment analysis of these targets indicated a primary enrichment in inflammation and immune-related pathways. A Venn diagram intersects these target proteins, and our aforementioned six core genes yielded three common genes, suggesting the potential efficacy of Danshen in sepsis treatment through these genes. This study highlights the pivotal roles of age-related genes in the molecular mechanisms of sepsis, offers potential biomarkers, and identifies promising therapeutic compounds, laying a robust foundation for future studies on the treatment of sepsis.

Researchers are interested in genomic level cancer classification because it offers a thorough, precise, and reliable diagnosis for various cancer types. The gene expression data set is made up of hundreds of samples, each containing thousands of genes. As a result, accurately and effectively classifying such samples is difficult. To analyze gene expression patterns efficiently, recent studies have used artificial intelligence, data mining, and statistical techniques. Furthermore, machine learning methods have been extensively used to construct robust and accurate classification frameworks. In this paper, we examined the effectiveness of machine learning-based classification techniques such as Support Vector Machine, Logistic Regression, Naïve Bayes, Decision Tree, K-Nearest Neighbors, and Random Forest methods. In a total of 72 patients, 7129 distinct gene expressions and their degrees were evaluated to determine their relevance. Features have been normalized and PCA was used to decrease the dataset's dimensionality and complexity. The dataset was divided into training and test sets, with 90% of the occurrences in the training set and 10% in the test set. And finally, SMOTE was employed to oversample the minority class. In this study, gene expression data was utilized to differentiate Acute Myeloid Leukemia (AML) from Acute Lymphoblastic Leukemia (ALL). To evaluate the performance of the six machine learning models we used, we examined their associated confusion matrices and calculated nine performance evaluation metrics for each model's prediction on the test dataset. Results demonstrate that SVM, Logistic Regression, and Naïve Bayes outperformed the other ML techniques with an accuracy of 100%.

BackgroundThis study aimed to construct predictive models for the risk of sepsis in patients with Acute pancreatitis (AP) using machine learning methods and compared optimal one with the logistic regression (LR) model and scoring systems.MethodsIn this retrospective cohort study, data were collected from the Medical Information Mart for Intensive Care III (MIMIC III) database between 2001 and 2012 and the MIMIC IV database between 2008 and 2019. Patients were randomly divided into training and test sets (8:2). The least absolute shrinkage and selection operator (LASSO) regression plus 5-fold cross-validation were used to screen and confirm the predictive factors. Based on the selected predictive factors, 6 machine learning models were constructed, including support vector machine (SVM), K-nearest neighbour (KNN), multi-layer perceptron (MLP), LR, gradient boosting decision tree (GBDT) and adaptive enhancement algorithm (AdaBoost). The models and scoring systems were evaluated and compared using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and the area under the curve (AUC).ResultsA total of 1, 672 patients were eligible for participation. In the training set, 261 AP patients (19.51%) were diagnosed with sepsis. The predictive factors for the risk of sepsis in AP patients included age, insurance, vasopressors, mechanical ventilation, Glasgow Coma Scale (GCS), heart rate, respiratory rate, temperature, SpO2, platelet, red blood cell distribution width (RDW), International Normalized Ratio (INR), and blood urea nitrogen (BUN). The AUC of the GBDT model for sepsis prediction in the AP patients in the testing set was 0.985. The GBDT model showed better performance in sepsis prediction than the LR, systemic inflammatory response syndrome (SIRS) score, bedside index for severity in acute pancreatitis (BISAP) score, sequential organ failure assessment (SOFA) score, quick-SOFA (qSOFA), and simplified acute physiology score II (SAPS II).ConclusionThe present findings suggest that compared to the classical LR model and SOFA, qSOFA, SAPS II, SIRS, and BISAP scores, the machine learning model-GBDT model had a better performance in predicting sepsis in the AP patients, which is a useful tool for early identification of high-risk patients and timely clinical interventions.

COVID-19 is a pandemic that causes a high rate of spread and Acute Respiratory Distress Syndrome (ARDS). Severe pneumonia in infected individuals has resulted in too many patients being admitted to the Intensive Care Unit (ICU). This has placed unprecedented pressure on health systems by exceeding capacities. It is essential to detect the prognosis of this disease so that the health systems can remain active and the conditions of the patients who need to be hospitalized in the ICU do not become critical. In this study, COVID-19 prognosis was detected by using ICU admission (COVID-19 SEVERITY) and COVID-19 related death (COVID- 19 MORTALITY) datasets with Machine Learning (ML) methods. The missing data of the datasets were filled with K-Nearest Neighbor (KNN), and Min-Max normalization was performed. Datasets were divided three times into training and test sets, and the data were balanced with the Synthetic Minority Oversampling Technique (SMOTE). Then, classification was carried out using Ensemble Learning (EL) methods. For COVID-19 SEVERITY and COVID-19 MORTALITY, 89.54% and 97.25% accuracy were achieved with the Adaboost classifier, respectively. Successful and rapid COVID-19 prognosis detection with ML methods will help to use the ICU more efficiently and relieve the pressure on health systems.

Immune checkpoint inhibitor-related pneumonitis (ICIP) is a common and potentially life-threatening adverse event with non-specific symptoms. It is of significance to identify high-risk population of ICIP. However, existing prediction models for ICIP are often limited by their reliance on clinically inaccessible variables and homogeneous methodologies, hindering their clinical utility. This study aimed to develop a clinical risk-prediction model for ICIP in patients with gastrointestinal (GI) cancer based on four machine learning (ML) methods. We conducted a retrospective analysis of data from GI cancer patients who received immune checkpoint inhibitors (ICIs) between 2018 and 2022 in Beijing Cancer Hospital. For each patient, 36 clinical indicators associated with pneumonia risk were gathered. The dataset was split into training and testing sets in a ratio of 7:3. Variable selection was first performed using Least Absolute Shrinkage and Selection Operator (LASSO) regression. Subsequently, four ML algorithms: logistic regression (LR), random forest (RF), Support vector machine (SVM), and Adaptive Boosting (AdaBoost), were employed to develop and validate ICIP prediction models. The models' performance was assessed using sensitivity, specificity, precision, F1-score, and the area under the receiver operating characteristic curve (AUC) value. The optimal cutoff point for the best model was determined and a web-based tool was developed based on it. We collected medical data from 1,101 GI cancer patients. Ten predictive variables were identified as significant: gender, age, treatment line, smoking index, drinking history, lung metastasis, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, hemoglobin, and albumin. After constructing and comparing four ML models, the RF model demonstrated best performance with an AUC of 0.899. The web-based tool for ICIP risk prediction is available at https://healthy.aistarfish.com/business/pneumonia-prediction/#/home. We analyzed 36 clinical predictors of ICIP in 1,101 patients treated with ICIs, and 10 variables were included. The smoking index, albumin and hemoglobin emerged as novel predictors specific to GI cancers. Among the models constructed using four ML methods, the RF model showed the best performance. Additionally, a web-based tool was developed to facilitate the early clinical identification of populations at high risk of ICIP. Future directions include external validation of the model to enhance clinical usability.