Combination Of Classifiers Research Articles

E-Ensemble: A Novel Ensemble Classifier for Encrypted Video Identification

In recent years, video identification within encrypted network traffic has gained popularity for many reasons. For example, a government may want to track what content is being watched by its citizens, or businesses may want to block certain content for productivity. Many such reasons advocate for the need to track users on the internet. However, with the introduction of the secure socket layer (SSL) and transport layer security (TLS), it has become difficult to analyze traffic. In addition, dynamic adaptive streaming over HTTP (DASH), which creates abnormalities due to the variable-bitrate (VBR) encoding, makes it difficult for researchers to identify videos in internet traffic. The default quality settings in browsers automatically adjust the quality of streaming videos depending on the network load. These auto-quality settings also increase the challenge in video detection. This paper presents a novel ensemble classifier, E-Ensemble, which overcomes the abnormalities in video identification in encrypted network traffic. To achieve this, three different classifiers are combined by using two different combinations of classifiers: the hard-level and soft-level combinations. To verify the performance of the proposed classifier, the classifiers were trained on a video dataset collected over one month and tested on a separate video dataset captured over 20 days at a different date and time. The soft-level combination of classifiers showed more stable results in handling abnormalities in the dataset than those of the hard-level combination. Furthermore, the soft-level classifier combination technique outperformed the hard-level combination with a high accuracy of 81.81%, even in the auto-quality mode.

Hierarchical, multiclass prediction of Alzheimer’s clinical diagnosis using imputed, multimodal NACC data

AbstractBackgroundEarly detection and diagnosis are critical for effective treatment of Alzheimer’s disease (AD). Statistical machine learning holds promise for disease prediction using large AD repositories, such as the National Alzheimer’s Coordinating Center (NACC) dataset. Past work has extensively explored binary classifications of AD diagnosis. Here, we extended this to address two major, remaining challenges: data imputation and multiclass classification (e.g., AD vs Mild Cognitive Impairment, MCI vs. Healthy Cognition, HC).MethodFor imputation, we tested single‐value (baseline), k‐NN, and Bayesian imputation on a subset of complete training data with artificially removed values (completely at random). Per modality of interest, a systematic search was conducted for the best input feature set to impute the missing values. The mean absolute deviation and mismatch rate between the imputed and true values were used to evaluate the methods on a held‐out test set. For classification, we then implemented both “flat” multiclass and hierarchical classification, an extension of multiclass classification that organizes classes into a tree, to predict clinical diagnosis using multimodal inputs. Several pipelines with combinations of imputation methods, classifiers, and input modalities were tested on different hierarchical classification strategies, e.g., HC vs [AD vs MCI], compared with the flat, multiclass baseline (AD vs MCI vs HC).ResultCompared to the baseline methods for both imputation and classification, the proposed alternatives performed better, as measured by the predetermined imputation and classification metrics evaluated on the test set. For imputation, behavioral and neuropsychiatric features were more easily and accurately imputed while ApoE genotype was more difficult to impute. For classification, using a combination of classifiers and input features in the hierarchy outperformed the baseline method and support the notion that select biomarkers are better able to characterize AD subpopulations within the hierarchy.ConclusionEstablishing a systematic imputation method tailored to AD biomarker modalities fills a critical gap in achieving more robust disease prediction. Building a hierarchical classifier parallels working through a differential diagnosis for AD, which offers greater translational value than binary classification of AD. In combination, these results form an important step toward developing reliable, holistic classifiers of AD disease status.

A Smart Diseases Diagnosis and Classification Strategy of Electronic Healthcare Application Using Novel Hybrid Artificial Intelligence Approaches

In today’s world, the healthcare industry faces difficulties like a scarcity of healthcare professionals, ageing, and rising healthcare costs. Also the classification and decision making process using the data generated via electronic health sensors is of major concern. In the fields of research and medical services, artificial intelligence (AI) is widely employed. However, correct estimate for various illnesses is a significant issue. The implementation of a new hybrid artificial intelligence (AI)-based classifier for helping prediction diagnosis in patients with chronic cancer conditions is examined in this work. Unknown qualities are predicted and given using the Hierarchical Red deer optimization (HRDO) based feature extraction, which is based on realworld cases. The Self-Systemized Generative Fuzzy Algorithm (SSGFA), which finds irregularities in patient data and predicts sickness, is used to create the hybrid classification design. This study’s simulation analysis included datasets for colon, lung, and brain cancer illnesses. The new combination of classifiers’ better performance resulted in total classification with increased accuracy, precision, recall, and F-measure, respectively. In terms of performance indicators, the suggested strategy is also compared to traditional methods. This demonstrates the suggested classification model’s ability to appropriately categorize various illnesses information for categorization.

Abstract A012: Using clinical characteristics and molecular markers to predict the risk of subsequent ipsilateral breast events after excision of DCIS

Abstract Purpose: To examine incremental values of estrogen receptor (ER) status, body mass index (BMI), menopausal status, and a multi-gene classifier over commonly used clinical factors (i.e. age, tumor grade, comedonecrosis, surgical margins, and treatment) in predicting risk of ipsilateral recurrence (IR) within five years after DCIS diagnosis. Methods: A derivation cohort consisted of participants in the discovery cohort, a retrospective multicenter cohort study in women undergoing surgical resection for DCIS between 01/01/1998 and 02/29/2016. The validation cohort provided cases meeting the same eligibility criteria as the discovery cohort. This analysis includes 313 participants with RNA-seq data who either developed IR 1-5 years after initial DCIS diagnosis or were free from subsequent breast events at least five years. Cox proportional hazards regression was used to estimate hazard ratios (HRs) of IR in 216 TBCRC participants (76 with IR). We developed the clinical score using clinical predictors (aforementioned clinical factors and ER) and their regression coefficients from the model with the maximum predictive accuracy (e.g. c-index) and the minimum number of predictors, and the summary score integrating the clinical score and multi-gene classifier. Predictive performance of clinical and summary scores was validated in 97 RAHBT patients (20 with IR). Results: In the discovery cohort, ER negativity, but not BMI or menopausal status, was independently associated with a higher IR risk (HR=2.06, 95% CI 1.18-3.58), and adding ER to the clinical factors-based model increased predictive accuracy (c-index 0.68 to 0.70). The clinical score-adjusted HR was 14.96 (95% CI 8.64-25.91) for multi-gene classifier. Summary scores were better in predicting IR risk than clinical scores (c-index 0.82 vs. 0.70). Compared with their low-risk counterparts, the HR was 4.79 (95% CI 2.55-8.98) in the clinical score-defined high-risk group and 29.02 (95% CI 14.23-59.15) in the summary score-defined high-risk group. In the validation cohort, model performance was improved using summary scores as compared to clinical scores (c-index 0.74 vs. 0.58). Conclusion: Combination of clinical factors and multigene classifier provided more accurate risk estimates of IR within five years after excision of DCIS than clinical factors alone. Citation Format: Graham A. Colditz, Ying Liu, Siri H. Strand, Lorraine King, Jeff Marks, Carlo Maley, Robert B. West, E. Shelley Hwang. Using clinical characteristics and molecular markers to predict the risk of subsequent ipsilateral breast events after excision of DCIS [abstract]. In: Proceedings of the AACR Special Conference on Rethinking DCIS: An Opportunity for Prevention?; 2022 Sep 8-11; Philadelphia, PA. Philadelphia (PA): AACR; Can Prev Res 2022;15(12 Suppl_1): Abstract nr A012.

Evidential Combination of Classifiers for Imbalanced Data

It remains an important research topic for the classification of imbalanced data. There exist some methods to solve this problem, such as hybrid-sampling, over-sampling, and under-sampling. Each method has its own advantage, and different methods generally provide some complementary knowledge. We want to combine these three methods at the decision level in an appropriate way for achieving as good as possible classification performance. Evidence theory is expert at representing and combining uncertain information. So a new method called an evidential combination of classifiers (ECC) is proposed for dealing with imbalanced data. The classification result generated by different strategies (i.e., hybrid-sampling, over-sampling, or under-sampling) may have different reliabilities for query patterns. A cautious reliability evaluation rule is developed for each classification result based on the close neighborhoods. After that, the classification result is revised with a new belief redistribution way according to the reliability evaluation, and the probability/belief of one class can be partially transferred to other classes as well as the total ignorance, which is defined by the whole frame of classes. By doing this, we can reduce the error risk of each classification method. Then, the revised classification results from different methods are combined by evidence theory to make the final class decision. The effectiveness of the ECC method has been demonstrated using several experiments, and it shows that ECC can effectively improve the classification performance comparing with other related methods.

A New Feature Extraction Method for EMG Signals

Surface Electromyography (sEMG) is an important tool for gesture recognition. Features and classification methods have to be carefully selected to be successful in the recognition of electromyografic signals. In most of the sEMG studies, time and frequency domain features have been extracted and classified with a single classifier. But neither one feature nor one classifier alone has achieved high classification accuracies. Using a feature and classifier combination would be a solution for this problem, and increase the accuracies. As a contribution to this field, a new time domain EMG feature is suggested and its classification performance is examined for feature and classifier combinations in this study. According to the results of this study, the new feature has high classification accuracy, and when it is used with AR and ST features, the average of the classification accuracy reaches 99.57% for multiple SVM classifier. Besides, the new feature+AR+ ST feature combination shows high classification accuracy for single classifier, and this eliminates the need for multiple classifiers.

The prognostic utility of serum thyrotropin in hospitalized Covid-19 patients: statistical and machine learning approaches.

To assess the prognostic value of serum TSH in Greek patients with COVID-19 and compare it with that of commonly used prognostic biomarkers. Retrospective study of 128 COVID-19 in patients with no history of thyroid disease. Serum TSH, albumin, CRP, ferritin, and D-dimers were measured at admission. Outcomes were classified as "favorable" (discharge from hospital) and "adverse" (intubation or in-hospital death of any cause). The prognostic performance of TSH and other indices was assessed using binary logistic regression, machine learning classifiers, and ROC curve analysis. Patients with adverse outcomes had significantly lower TSH compared to those with favorable outcomes (0.61 versus 1.09 mIU/L, p < 0.001). Binary logistic regression with sex, age, TSH, albumin, CRP, ferritin, and D-dimers as covariates showed that only albumin (p < 0.001) and TSH (p = 0.006) were significantly predictive of the outcome. Serum TSH below the optimal cut-off value of 0.5 mIU/L was associated with an odds ratio of 4.13 (95% C.I.: 1.41-12.05) for adverse outcome. Artificial neural network analysis showed that the prognostic importance of TSH was second only to that of albumin. However, the prognostic accuracy of low TSH was limited, with an AUC of 69.5%, compared to albumin's 86.9%. A Naïve Bayes classifier based on the combination of serum albumin and TSH levels achieved high prognostic accuracy (AUC 99.2%). Low serum TSH is independently associated with adverse outcome in hospitalized Greek patients with COVID-19 but its prognostic utility is limited. The integration of serum TSH into machine learning classifiers in combination with other biomarkers enables outcome prediction with high accuracy.

Epileptic Seizure Detection Using a Hybrid 1D CNN-Machine Learning Approach from EEG Data.

Electroencephalography (EEG) is a widely used technique for the detection of epileptic seizures. It can be recorded in a noninvasive manner to present the electrical activity of the brain. The visual inspection of nonlinear and highly complex EEG signals is both costly and time-consuming. Therefore, an effective automatic detection system is needed to assist in the long-term evaluation and treatment of patients. Traditional approaches based on machine learning require feature extraction, while deep learning approaches are time-consuming and require more layers for effective feature learning and processing of complex EEG waveforms. Deep learning-based approaches also have weak generalization ability. This paper proposes a solution based on the combination of convolution neural networks (CNN) and machine learning classifiers. It preprocesses the EEG signal using the Butterworth filter and performs feature extraction using CNN. From the extracted set of features, the approach selects only the relevant features using mutual information-based estimators to reduce the curse of dimensionality and improve classification accuracy. The selected features are then passed as input to different machine learning classifiers. The suggested solution is evaluated on the University of Bonn dataset and CHB-MIT datasets. Our model effectively predicts 2, 3, 4, and 5 classes with accuracy of 100%, 99%, 94.6%, and 94%, respectively, for the Bonn dataset and 98% for CHB-MIT datasets.

A Comparative Study of Local Descriptors and Classifiers for Facial Expression Recognition

Facial Expression Recognition (FER) is a growing area of research due to its numerous applications in market research, video gaming, healthcare, security, e-learning, and robotics. One of the most common frameworks for recognizing facial expressions is by extracting facial features from an image and classifying them as one of several prototypic expressions. Despite the recent advances, it is still a challenging task to develop robust facial expression descriptors. This study aimed to analyze the performances of various local descriptors and classifiers in the FER problem. Several experiments were conducted under different settings, such as varied extraction parameters, different numbers of expressions, and two datasets, to discover the best combinations of local descriptors and classifiers. Of all the considered descriptors, HOG (Histogram of Oriented Gradients) and ALDP (Angled Local Directional Patterns) were some of the most promising, while SVM (Support Vector Machines) and MLP (Multi-Layer Perceptron) were the best among the considered classifiers. The results obtained signify that conventional FER approaches are still comparable to state-of-the-art methods based on deep learning.

Combining sentiment analysis classifiers to explore multilingual news articles covering London 2012 and Rio 2016 Olympics

This study aims to present an approach for the challenges of working with Sentiment Analysis (SA) applied to news articles in a multilingual corpus. It looks at the use and combination of multiple algorithms to explore news articles published in English and Portuguese. It presents a methodology that starts by evaluating and combining four SA algorithms (SenticNet, SentiStrength, Vader and BERT, being BERT trained in two datasets) to improve the quality of outputs. A thorough review of the algorithms’ limitations is conducted using SHAP, an explainable AI tool, resulting in a list of issues that researchers must consider before using SA to interpret texts. We propose a combination of the three best classifiers (Vader, Amazon BERT and Sent140 BERT) to identify contradictory results, improving the quality of the positive, neutral and negative labels assigned to the texts. Challenges with translation are addressed, indicating possible solutions for non-English corpora. As a case study, the method is applied to the study of the media coverage of London 2012 and Rio 2016 Olympic legacies. The combination of different classifiers has proved to be efficient, revealing the unbalance between the media coverage of London 2012, much more positive, and Rio 2016, more negative.

Diagnostic implication of a circulating serum-based three-microRNA signature in hepatocellular carcinoma.

Owing to the diagnostic dilemma, the prognosis of hepatocellular carcinoma (HCC) remains impoverished, contributing to the globally high mortality rate. Currently, HCC diagnosis depends on the combination of imaging modalities and the measurement of serum alpha-fetoprotein (AFP) levels. Nevertheless, these conventional modalities exhibit poor performance in detecting HCC at early stages. Thus, there is a pressing need to identify novel circulating biomarkers to promote diagnostic accuracy and surveillance. Circulating miRNAs are emerging as promising diagnostic tools in screening various cancers, including HCC. However, because of heterogenous and, at times, contradictory reports, the universality of miRNAs in clinical settings remains elusive. Consequently, we proposed to explore the diagnostic potential of ten miRNAs selected on a candidate-based approach in HCC diagnosis. The expression of ten candidate miRNAs (Let-7a, miR-15a, miR-26a, miR-124, miR-126, miR-155, miR-219, miR-221, miR-222, and miR-340) was investigated in serum and tissue of 66 subjects, including 33 HCC patients and 33 healthy controls (HC), by rt-PCR. Receiver operating characteristic curve (ROC) analysis was used to determine the diagnostic accuracy of the prospective serum miRNA panel. To anticipate the potential biological roles of a three-miRNA signature, the target genes were evaluated using the Kyoto Encyclopedia of Genes and Genomes (KEGG) signaling pathway. The serum and tissue expression of miRNAs (Let-7a, miR-26a, miR-124, miR-155, miR-221, miR-222, and miR-340) were differentially expressed in HCC patients (p < 0.05). The ROC analysis revealed promising diagnostic performance of Let-7a (AUC = 0.801), miR-221 (AUC = 0.786), and miR-2 (AUC = 0.758) in discriminating HCC from HC. Furthermore, in a logistic regression equation, we identified a three-miRNA panel (Let-7a, miR-221, and miR-222; AUC = 0.932) with improved diagnostic efficiency in differentiating HCC from HC. Remarkably, the combination of AFP and a three-miRNA panel offered a higher accuracy of HCC diagnosis (AUC = 0.961) than AFP alone. The functional enrichment analysis demonstrated that target genes may contribute to pathways associated with HCC and cell-cycle regulation, indicating possible crosstalk of miRNAs with HCC development. To conclude, the combined classifier of a three-miRNA panel and AFP could be indispensable circulating biomarkers for HCC diagnosis. Furthermore, targeting predicted genes may provide new therapeutic clues for the treatment of aggressive HCC.

Identifying plastics with photoluminescence spectroscopy and machine learning

A quantitative understanding of the worldwide plastics distribution is required not only to assess the extent and possible impact of plastic litter on the environment but also to identify possible counter measures. A systematic collection of data characterizing amount and composition of plastics has to be based on two crucial components: (i) An experimental approach that is simple enough to be accessible worldwide and sensible enough to capture the diversity of plastics; (ii) An analysis pipeline that is able to extract the relevant parameters from the vast amount of experimental data. In this study, we demonstrate that such an approach could be realized by a combination of photoluminescence spectroscopy and a machine learning-based theoretical analysis. We show that appropriate combinations of classifiers with dimensional reduction algorithms are able to identify specific material properties from the spectroscopic data. The best combination is based on an unsupervised learning technique making our approach robust to alternations of the input data.

Research on the Application of Multi-Source Information Fusion in Multiple Gait Pattern Transition Recognition.

Multi-source information fusion technology is a kind of information processing technology which comprehensively processes and utilizes multi-source uncertain information. It is an effective scheme to solve complex pattern recognition and improve classification performance. This study aims to improve the accuracy and robustness of exoskeleton gait pattern transition recognition in complex environments. Based on the theory of multi-source information fusion, this paper explored a multi-source information fusion model for exoskeleton gait pattern transition recognition in terms of two aspects of multi-source information fusion strategy and multi-classifier fusion. For eight common gait pattern transitions (between level and stair walking and between level and ramp walking), we proposed a hybrid fusion strategy of multi-source information at the feature level and decision level. We first selected an optimal feature subset through correlation feature extraction and feature selection algorithm, followed by the feature fusion through the classifier. We then studied the construction of a multi-classifier fusion model with a focus on the selection of base classifier and multi-classifier fusion algorithm. By analyzing the classification performance and robustness of the multi-classifier fusion model integrating multiple classifier combinations with a number of multi-classifier fusion algorithms, we finally constructed a multi-classifier fusion model based on D-S evidence theory and the combination of three SVM classifiers with different kernel functions (linear, RBF, polynomial). Such multi-source information fusion model improved the anti-interference and fault tolerance of the model through the hybrid fusion strategy of feature level and decision level and had higher accuracy and robustness in the gait pattern transition recognition, whose average recognition accuracy for eight gait pattern transitions reached 99.70%, which increased by 0.15% compared with the highest average recognition accuracy of the single classifier. Moreover, the average recognition accuracy in the absence of different feature data reached 97.47% with good robustness.

New Classification Method for Independent Data Sources Using Pawlak Conflict Model and Decision Trees.

The research concerns data collected in independent sets-more specifically, in local decision tables. A possible approach to managing these data is to build local classifiers based on each table individually. In the literature, many approaches toward combining the final prediction results of independent classifiers can be found, but insufficient efforts have been made on the study of tables' cooperation and coalitions' formation. The importance of such an approach was expected on two levels. First, the impact on the quality of classification-the ability to build combined classifiers for coalitions of tables should allow for the learning of more generalized concepts. In turn, this should have an impact on the quality of classification of new objects. Second, combining tables into coalitions will result in reduced computational complexity-a reduced number of classifiers will be built. The paper proposes a new method for creating coalitions of local tables and generating an aggregated classifier for each coalition. Coalitions are generated by determining certain characteristics of attribute values occurring in local tables and applying the Pawlak conflict analysis model. In the study, the classification and regression trees with Gini index are built based on the aggregated table for one coalition. The system bears a hierarchical structure, as in the next stage the decisions generated by the classifiers for coalitions are aggregated using majority voting. The classification quality of the proposed system was compared with an approach that does not use local data cooperation and coalition creation. The structure of the system is parallel and decision trees are built independently for local tables. In the paper, it was shown that the proposed approach provides a significant improvement in classification quality and execution time. The Wilcoxon test confirmed that differences in accuracy rate of the results obtained for the proposed method and results obtained without coalitions are significant, with a p level = 0.005. The average accuracy rate values obtained for the proposed approach and the approach without coalitions are, respectively: 0.847 and 0.812; so the difference is quite large. Moreover, the algorithm implementing the proposed approach performed up to 21-times faster than the algorithm implementing the approach without using coalitions.

Ensemble feature selection and classification methods for machine learning-based coronary artery disease diagnosis

Coronary artery disease (CAD) is a condition in which the heart is not fed sufficiently as a result of the accumulation of fatty matter. As reported by the World Health Organization, around 32% of the total deaths in the world are caused by CAD, and it is estimated that approximately 23.6 million people will die from this disease in 2030. CAD develops over time, and the diagnosis of this disease is difficult until a blockage or a heart attack occurs. In order to bypass the side effects and high costs of the current methods, researchers have proposed to diagnose CADs with computer-aided systems, which analyze some physical and biochemical values at a lower cost. In this study, for the CAD diagnosis, (i) seven different computational feature selection (FS) methods, one domain knowledge-based FS method, and different classification algorithms have been evaluated; (ii) an exhaustive ensemble FS method and a probabilistic ensemble FS method have been proposed. The proposed approach is tested on three publicly available CAD data sets using six different classification algorithms and four different variants of voting algorithms. The performance metrics have been comparatively evaluated with numerous combinations of classifiers and FS methods. The multi-layer perceptron classifier obtained satisfactory results on three data sets. Performance evaluations show that the proposed approach resulted in 91.78%, 85.55%, and 85.47% accuracy for the Z-Alizadeh Sani, Statlog, and Cleveland data sets, respectively.

A REVIEW PAPER ON OCR USING CONVOLUTIONAL NEURAL NETWORKS

this paper presents a literature review on OCR for different languages using convolutional neural network techniques. Optical Character Recognition is the process of converting an input text image into a machine encoded format. Different methods are used in OCR for different languages. The main steps of optical character recognition are pre-processing, segmentation and recognition. Recognizing handwritten text is harder than recognizing printed text. Convolutional Neural Network has shown remarkable improvement in recognizing characters of different languages. The novelty of the OCR is its robustness to image quality, image contrast, font style and font size. Common machine learning methods usually apply a combination of feature extractor and trainable classifier. The use of CNN leads to significant improvements across different machine-learning classification algorithms.

Semantic and Acoustic Markers in Schizophrenia-Spectrum Disorders: A Combinatory Machine Learning Approach.

Speech is a promising marker to aid diagnosis of schizophrenia-spectrum disorders, as it reflects symptoms like thought disorder and negative symptoms. Previous approaches made use of different domains of speech for diagnostic classification, including features like coherence (semantic) and form (acoustic). However, an examination of the added value of each domain when combined is lacking as of yet. Here, we investigate the acoustic and semantic domains separately and combined. Using semi-structured interviews, speech of 94 subjects with schizophrenia-spectrum disorders (SSD) and 73 healthy controls (HC) was recorded. Acoustic features were extracted using a standardized feature-set, and transcribed interviews were used to calculate semantic word similarity using word2vec. Random forest classifiers were trained for each domain. A third classifier was used to combine features from both domains; 10-fold cross-validation was used for each model. The acoustic random forest classifier achieved 81% accuracy classifying SSD and HC, while the semantic domain classifier reached an accuracy of 80%. Joining features from the two domains, the combined classifier reached 85% accuracy, significantly improving on separate domain classifiers. For the combined classifier, top features were fragmented speech from the acoustic domain and variance of similarity from the semantic domain. Both semantic and acoustic analyses of speech achieved ~80% accuracy in classifying SSD from HC. We replicate earlier findings per domain, additionally showing that combining these features significantly improves classification performance. Feature importance and accuracy in combined classification indicate that the domains measure different, complementing aspects of speech.

A Deep Learning Classifier Based on Pre-Radiation Computed Tomography and Clinical Parameters to Predict Pathological Complete Response after Neoadjuvant Chemoradiation in Esophageal Cancer

<h3>Purpose/Objective(s)</h3> Neoadjuvant chemoradiation (NCRT) followed by surgery is the standard treatment for resectable esophageal cancer. More than 30% patients achieve pathological complete response (pCR) after NCRT, who may avoid the followed surgery. However, there is no reliable method in predicting pCR yet. Artificial intelligence, especially deep learning, has made great progress in many fields including treatment response prediction. Therefore, we built up a deep learning classifier based on pre-radiation computed tomography and clinical parameters to predict pCR after NCRT for esophageal cancer. <h3>Materials/Methods</h3> Between 2009 and 2021, consecutive patients with esophageal cancer received NCRT and complete resection were retrospectively analyzed. Pathological response assessed on surgical specimen was collected. Patients were randomly assigned to the training set, validation set, and testing set as 7: 1: 2. We built a binary classification neural network based on 3D Resnet. Pre-radiation computed tomography (CT) was fed as input to build the imaging classifier. The filtered clinical parameters including gender, tumor location, clinical stage, pathological type, sequence of chemoradiation, chemotherapy regimen and radiotherapy technique were then added by encoded as fully connected layer to build the combined classifier. Area under the receiver operating characteristic curve (AUC) was calculated to evaluate the prediction performance and the optimal cut-off point was determined by Youden index. <h3>Results</h3> Totally 279 patients were enrolled, of whom 93 achieved pCR (33.3%). The performances of imaging classifier were AUC=0.989 (95% CI 0.937-0.986) with the sensitivity of 98.6% and specificity of 98.5% in the training set, and AUC=0.649 (95% CI 0.481-0.660) with the sensitivity of 66.7% and specificity of 58.9% in the testing set, respectively. After the addition of clinical parameters, the combined classifier showed AUC=0.855 (95% CI 0.797-0.986) with the sensitivity of 82.4% and specificity of 74.3% in the training set, and AUC=0.731 (95%CI 0.631-0.819) with the sensitivity of 76.6% and specificity of 65.6% in the testing set, respectively. <h3>Conclusion</h3> The combined deep learning classifier can accurately predict pCR after NCRT for esophageal cancer. Besides, addition of necessary clinical parameters can remedy the overfitting of imaging classifier. Prospective exploration based on larger data sets is needed to further improve the accuracy and generalization.

An Efficient Method for Feature Extraction and Selection in Power Quality Recognition

—Since decade, many time-frequency analysis methods with combination of classifiers have been studied in literature for recognition of power quality (PQ) events. In these studies, feature extraction and selection have a vital role to enhance PQ classification accuracy and to reduce computational complexity for PQ recognition. This paper presents a new method to detect and classify PQ disturbances based on modified Stockwell Transform (ST) for extraction of features and Hybrid Grey Wolf Optimization (HGWO) for feature selection along with K Nearest Neighbor (KNN) classifier. Simulation of the proposed method using MATLAB is carried out through a wide range of eighteen synthetic PQ events to validate the effectiveness of the selected features. Further, an experiment is extended for six classes of real PQ events acquired from self excited induction generator (SEIG) system in a laboratory experimental setup. Proposed method is also employed on those real time data to study the classification accuracy performance. In these experiments, an impressive accuracy of 99.94% and 99.3% for synthetic and real time PQ event data, respectively are reported. Hence, it is observed from result analysis, this proposed method can be utilized for recognition of PQ events in real time power system.

A combined signature of glycolysis and immune landscape predicts prognosis and therapeutic response in prostate cancer.

Prostate cancer (PCa) is a common malignancy that poses a major threat to the health of men. Prostate-specific antigen (PSA) and its derivatives, as FDA-approved detection assays, are insufficient to serve as optimal markers for patient prognosis and clinical decision-making. It is widely acknowledged that aberrant glycolytic metabolism in PCa is related to tumor progression and acidifies the tumor microenvironment (TME). Considering the non-negligible impacts of glycolysis and immune functions on PCa, we developed a combined classifier in prostate cancer. The Glycolysis Score containing 19 genes and TME Score including three immune cells were created, using the univariate and multivariate Cox proportional hazards model, log-rank test, least absolute shrinkage and selection operator (LASSO) regression analysis and the bootstrap approach. Combining the glycolysis and immunological landscape, the Glycolysis-TME Classifier was then constructed. It was observed that the classifier was more accurate in predicting the prognosis of patients than the current biomarkers. Notably, there were significant differences in metabolic activity, signaling pathways, mutational landscape, immunotherapeutic response, and drug sensitivity among the Glycolysishigh/TMElow, Mixed group and Glycolysislow/TMEhigh identified by this classifier. Overall, due to the significant prognostic value and potential therapeutic guidance of the Glycolysis-TME Classifier, we anticipate that this classifier will be clinically beneficial in the management of patients with PCa.