Machine Learning Classifier Models Research Articles

CT-based intratumoral and peritumoral deep transfer learning features prediction of lymph node metastasis in non-small cell lung cancer.

The main metastatic route for lung cancer is lymph node metastasis, and studies have shown that non-small cell lung cancer (NSCLC) has a high risk of lymph node infiltration. This study aimed to compare the performance of handcrafted radiomics (HR) features and deep transfer learning (DTL) features in Computed Tomography (CT) of intratumoral and peritumoral regions in predicting the metastatic status of NSCLC lymph nodes in different machine learning classifier models. We retrospectively collected data of 199 patients with pathologically confirmed NSCLC. All patients were divided into training (n = 159) and validation (n = 40) cohorts, respectively. The best HR and DTL features in the intratumoral and peritumoral regions were extracted and selected, respectively. Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Light Gradient Boosting Machine (Light GBM), Multilayer Perceptron (MLP), and Logistic Regression (LR) models were constructed, and the performance of the models was evaluated. Among the five models in the training and validation cohorts, the LR classifier model performed best in terms of HR and DTL features. The AUCs of the training cohort were 0.841 (95% CI: 0.776-0.907) and 0.955 (95% CI: 0.926-0.983), and the AUCs of the validation cohort were 0.812 (95% CI: 0.677-0.948) and 0.893 (95% CI: 0.795-0.991), respectively. The DTL signature was superior to the handcrafted radiomics signature. Compared with the radiomics signature, the DTL signature constructed based on intratumoral and peritumoral areas in CT can better predict NSCLC lymph node metastasis.

Machine learning enabled detection of COVID-19 pneumonia using exhaled breath analysis: a proof-of-concept study

Detection of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) relies on real-time-reverse-transcriptase polymerase chain reaction (RT-PCR) on nasopharyngeal swabs. The false-negative rate of RT-PCR can be high when viral burden and infection is localized distally in the lower airways and lung parenchyma. An alternate safe, simple and accessible method for sampling the lower airways is needed to aid in the early and rapid diagnosis of COVID-19 pneumonia. In a prospective unblinded observational study, patients admitted with a positive RT-PCR and symptoms of SARS-CoV-2 infection were enrolled from three hospitals in Ontario, Canada. Healthy individuals or hospitalized patients with negative RT-PCR and without respiratory symptoms were enrolled into the control group. Breath samples were collected and analyzed by laser absorption spectroscopy (LAS) for volatile organic compounds (VOCs) and classified by machine learning (ML) approaches to identify unique LAS-spectra patterns (breathprints) for SARS-CoV-2. Of the 135 patients enrolled, 115 patients provided analyzable breath samples. Using LAS-breathprints to train ML classifier models resulted in an accuracy of 72.2%–81.7% in differentiating between SARS-CoV2 positive and negative groups. The performance was consistent across subgroups of different age, sex, body mass index, SARS-CoV-2 variants, time of disease onset and oxygen requirement. The overall performance was higher than compared to VOC-trained classifier model, which had an accuracy of 63%–74.7%. This study demonstrates that a ML-based breathprint model using LAS analysis of exhaled breath may be a valuable non-invasive method for studying the lower airways and detecting SARS-CoV-2 and other respiratory pathogens. The technology and the ML approach can be easily deployed in any setting with minimal training. This will greatly improve access and scalability to meet surge capacity; allow early and rapid detection to inform therapy; and offers great versatility in developing new classifier models quickly for future outbreaks.

A Machine Learning Ensemble Model for the Detection of Cyberbullying

The pervasive use of social media platforms, such as Facebook, Instagram, and X, has significantly amplified our electronic interconnectedness. Moreover, these platforms are now easily accessible from any location at any given time. However, the increased popularity of social media has also led to cyberbullying.It is imperative to address the need for finding, monitoring, and mitigating cyberbullying posts on social media platforms. Motivated by this necessity, we present this paper to contribute to developing an automated system for detecting binary labels of aggressive tweets.Our study has demonstrated remarkable performance compared to previous experiments on the same dataset. We employed the stacking ensemble machine learning method, utilizing four various feature extraction techniques to optimize performance within the stacking ensemble learning framework. Combining five machine learning algorithms,Decision Trees, Random Forest, Linear Support Vector Classification, Logistic Regression, and K-Nearest Neighbors into an ensemble method, we achieved superior results compared to traditional machine learning classifier models. The stacking classifier achieved a high accuracy rate of 94.00%, outperforming traditional machine learning models and surpassing the results of prior experiments that utilized the same dataset. The outcomes of our experiments showcased an accuracy rate of 0.94% in detection tweets as aggressive or non-aggressive.

Machine Learning based Forest Fire Prediction: A Comparative Approach

Wildfires are among the world's most pressing issues, and they are getting more prevalent as global warming and other environmental conditions deteriorate. These wildfires might be caused by humans or by natural causes. Wildfires are one of the factors contributing to the extinction of rare flora and wildlife that serve to maintain our planet's ecological balance. In this paper, a comparative analysis of various machine learning classifier models for predicting forest fires was undertaken using two separate datasets. The suggested system's processing is dependent on a few characteristics such as temperature, humidity, oxygen, and wind. Several machine learning classification techniques, including logistic regression, support vector classifier, decision tree, k neighbors and random forest, were used in this study. For further optimization of the model, K-fold cross validation method and hyperparameter tuning were implemented. The system reveals Support Vector Machine as the best strategy for the forest fire dataset, with 96.88% accuracy. Random Forest method was found to be the best for the Cortez and Morais dataset, with 90.24% accuracy.

An Ensemble-Based Multi-Classification Machine Learning Classifiers Approach to Detect Multiple Classes of Cyberbullying

The impact of communication through social media is currently considered a significant social issue. This issue can lead to inappropriate behavior using social media, which is referred to as cyberbullying. Automated systems are capable of efficiently identifying cyberbullying and performing sentiment analysis on social media platforms. This study focuses on enhancing a system to detect six types of cyberbullying tweets. Employing multi-classification algorithms on a cyberbullying dataset, our approach achieved high accuracy, particularly with the TF-IDF (bigram) feature extraction. Our experiment achieved high performance compared with that stated for previous experiments on the same dataset. Two ensemble machine learning methods, employing the N-gram with TF-IDF feature-extraction technique, demonstrated superior performance in classification. Three popular multi-classification algorithms: Decision Trees, Random Forest, and XGBoost, were combined into two varied ensemble methods separately. These ensemble classifiers demonstrated superior performance compared to traditional machine learning classifier models. The stacking classifier reached 90.71% accuracy and the voting classifier 90.44%. The results of the experiments showed that the framework can detect six different types of cyberbullying more efficiently, with an accuracy rate of 0.9071.

A Radiomics Approach to Identify Immunologically Active Tumor in Patients with Head and Neck Squamous Cell Carcinomas.

We recently developed a gene-expression-based HOT score to identify the hot/cold phenotype of head and neck squamous cell carcinomas (HNSCCs), which is associated with the response to immunotherapy. Our goal was to determine whether radiomic profiling from computed tomography (CT) scans can distinguish hot and cold HNSCC. We included 113 patients from The Cancer Genome Atlas (TCGA) and 20 patients from the Groupe Hospitalier Pitié-Salpêtrière (GHPS) with HNSCC, all with available pre-treatment CT scans. The hot/cold phenotype was computed for all patients using the HOT score. The IBEX software (version 4.11.9, accessed on 30 march 2020) was used to extract radiomic features from the delineated tumor region in both datasets, and the intraclass correlation coefficient (ICC) was computed to select robust features. Machine learning classifier models were trained and tested in the TCGA dataset and validated using the area under the receiver operator characteristic curve (AUC) in the GHPS cohort. A total of 144 radiomic features with an ICC >0.9 was selected. An XGBoost model including these selected features showed the best performance prediction of the hot/cold phenotype with AUC = 0.86 in the GHPS validation dataset. We identified a relevant radiomic model to capture the overall hot/cold phenotype of HNSCC. This non-invasive approach could help with the identification of patients with HNSCC who may benefit from immunotherapy.

The Investigative Study on the Performance Analysis of SMOTE employed Machine Learning Classifier Models to DDoS Attack Detection

Distributed Denial of Service (DDoS) attack, a severe attack on the network services during the contemporary era, is categorized under active attacks in security attacks. The impact of this attack on the organization or individual resources leads to massive loss in terms of finance, reputation. Therefore, detecting Distributed DDoS attacks is vital in ensuring the availability and integrity of online services of an organization. The work in this paper employed machine learning techniques, complemented by Synthetic Minority Over-sampling Technique (SMOTE), to tackle the inherent challenge of imbalanced DDoS attack dataset: CSE-CIC-2018 and to enhance computational efficiency while maintaining accuracy with a fraction of the original dataset. The emphasis of the this works is to comprehensively assess the performance of five prominent algorithms of machine learning - Naive Bayes, Random Forest, Logistic Regression, Decision Tree, and XGBoost - in the context of detection of DDoS attack. The overhead of oversampling is handled with the application of SMOTE oversampling and it has been addressed data imbalance issues, improving the algorithms' capability to identify attacks of DDoS effectively. The work of this paper finds and reveals distinct comparative advantages among the algorithms employed in the DDoS attack detection and provides actionable insights in choosing the most suitable algorithms of Machine learning for the detection of DDoS attack, provided emphasizing the significance of SMOTE to enhance the algorithms' performance in the presence of imbalanced data. Eventually, this paper offers invaluable guidance for organizations seeking to make safe their network against DDoS attacks while considering the crucial tradeoffs between accuracy and computational efficiency. The proposed work in this paper presented the results that Random Forest classifier ensured the better performance with F1-Score value 0.99, Mathews Correlation Coefficient (MCC) value 0.98 and accuracy value 0.99 relative to other classifiers employed.

Re-tear after arthroscopic rotator cuff tear surgery: risk analysis using machine learning

Postoperative rotator cuff re-tear after arthroscopic rotator cuff repair (ARCR) is still a major problem. Various risk factors such as age, gender, and tear size have been reported. Recently, magnetic resonance imaging (MRI)-based stump classification was reported as an index of rotator cuff fragility. Although stump type 3 is reported to have a high re-tear rate, there are few reports on the risk of postoperative re-tear based on this classification. Machine learning (ML), an artificial intelligence technique, allows for more flexible predictive models than conventional statistical methods and has been applied to predict clinical outcomes. In this study, we used ML to predict postoperative re-tear risk after ARCR. The retrospective case-control study included 353 patients who underwent surgical treatment for complete rotator cuff tear using the suture-bridge technique. Patients who initially presented with re-tears and traumatic tears were excluded. In study participants, after the initial tear repair, rotator cuff re-tears were diagnosed by MRI; Sugaya classification types IV and V were defined as re-tears. Age, gender, stump classification, tear size, Goutallier classification, presence of diabetes, and hyperlipidemia were used for ML parameters to predict the risk of re-tear. Using Python's Scikit-learn as an ML library, five different AI models (logistic regression, random forest, AdaBoost, CatBoost, LightGBM) were trained on the existing data, and the prediction models were applied to the test dataset. The performance of these ML models was measured by the area under the receiver operating characteristic curve (AUC). Additionally, key features affecting re-tear were evaluated. The AUC for logistic regression was 0.78, random forest 0.82, AdaBoost 0.78, CatBoost 0.83, and LightGBM 0.87, respectively for each model. LightGBM showed the highest score. The important factors for model prediction were age, stump classification, and tear size. The ML classifier model predicted re-tears after ARCR with high accuracy, and the AI model showed that the most important characteristics affecting re-tears were age and imaging findings, including stump classification. This model may be able to predict postoperative rotator cuff re-tears based on clinical features. Prognosis Study (Case-control study).

Machine Learning Prediction Models to Reduce Length of Stay at Ambulatory Surgery Centers Through Case Resequencing

The post-anesthesia care unit (PACU) length of stay is an important perioperative efficiency metric. The aim of this study was to develop machine learning models to predict ambulatory surgery patients at risk for prolonged PACU length of stay - using only pre-operatively identified factors - and then to simulate the effectiveness in reducing the need for after-hours PACU staffing. Several machine learning classifier models were built to predict prolonged PACU length of stay (defined as PACU stay ≥ 3 hours) on a training set. A case resequencing exercise was then performed on the test set, in which historic cases were re-sequenced based on the predicted risk for prolonged PACU length of stay. The frequency of patients remaining in the PACU after-hours (≥ 7:00 pm) were compared between the simulated operating days versus actual operating room days. There were 10,928 ambulatory surgical patients included in the analysis, of which 580 (5.31%) had a PACU length of stay ≥ 3 hours. XGBoost with SMOTE performed the best (AUC = 0.712). The case resequencing exercise utilizing the XGBoost model resulted in an over three-fold improvement in the number of days in which patients would be in the PACU past 7pm as compared with historic performance (41% versus 12%, P<0.0001). Predictive models using preoperative patient characteristics may allow for optimized case sequencing, which may mitigate the effects of prolonged PACU lengths of stay on after-hours staffing utilization.

Data driven machine learning models for short‐term load forecasting considering electrical vehicle load

AbstractElectric vehicles (EVs) are gaining popularity due to their fuel efficiency and ability to reduce greenhouse gas emissions. Significant penetration of EVs with unregulated charging, which can have a substantial impact on power networks. Accurate load predictions, including the charging of EVs are crucial for ensuring the cost‐effective and dependable operation of power systems. In order to estimate short‐term load in the presence of EV load, the XG (extreme gradient) boost algorithm is proposed and the effectiveness of the performances are checked against other models. A variety of distinct meteorological parameters and the electrical load pattern for the years 2017 and 2018 for Northeast India are used to train the machine learning classifier models. R‐squared value analyses are also performed to identify the most correlated input parameters that influence the results of various models. Analysis shows that temperature, cloud cover, heat index, dew point, wind chill, and perceived temperature are substantially connected with electricity consumption. The performance of XG boost is outperformed by comparing prediction results with decision tree, random forest, K nearest neighbors, and logistic regression. Three separate case studies were employed for verification using the precision, F1 score, sensitivity, specificity, and accuracy metrics. Our findings demonstrate that, in comparison to other forecasting models, XG Boost exhibits higher accuracy (83.84%, 81.51%, and 85.97%) and robust forecasts.

Framework for Formal Verification of Machine Learning Based Complex System‐of‐Systems

ABSTRACTA complex system is characterized by emergence of global properties which are very difficult, if not impossible, to anticipate just from complete knowledge of component behaviors. Emergence, hierarchical organization, and numerosity are some of the characteristics of complex systems. Recently, there has been an exponential increase on the adoption of various neural network‐based machine learning models to govern the functionality and behavior of systems. With this increasing system complexity, achieving confidence in systems becomes even more difficult. Further, ease of interconnectivity among systems is permeating numerous system‐of‐systems, wherein multiple independent systems are expected to interact and collaborate to achieve unparalleled levels of functionality. Traditional verification and validation approaches are often inadequate to bring in the nuances of potential emergent behavior in a system‐of‐systems, which may be positive or negative. This paper describes a novel approach towards application of machine learning based classifiers and formal methods for analyzing and evaluating emergent behavior of complex system‐of‐systems that comprise a hybrid of constituent systems governed by conventional models and machine learning models. The proposed approach involves developing a machine learning classifier model that learns on potential negative and positive emergent behaviors, and predicts the behavior exhibited. A formal verification model is then developed to assert negative emergent behavior. The approach is illustrated through the case of a swarm of autonomous UAVs flying in a formation, and dynamically changing the shape of the formation, to support varying mission scenarios. The effectiveness and performance of the approach are quantified.

Tri-Axial Accelerometer-Based Recognition of Daily Activities Causing Shortness of Breath in COPD Patients

<strong>Background:</strong> The 6 min walk test is widely used to measure the severity of COPD, but this test imposes an extensive burden on patients and medical staff. As an alternative to this test, the measurement of activity amount during daily activities that cause COPD patients to feel shortness of breath can be acquired using wearable devices, and the measurements between COPD patients and healthy people can be compared. Accordingly, based on machine learning, we first evaluated the accuracy of the accelerometers in recognizing such activities. <strong>Methods:</strong> Forty-six healthy participants wore tri-axial accelerometers on the wrist and hip, and performed nine activities: changing clothes, sitting, standing, lying in a supine position, brushing teeth, moving luggage, going up/down the stairs, running, and walking. Features were extracted from 10 s windows of 31 datasets and input into a machine-learning classifier model. <strong>Results:</strong> The wrist+hip classifier recognized the activities of changing clothes, standing, lying in a supine position, brushing teeth, moving luggage, going up/down the stairs, walking, and running. Sitting and other movements were not appropriately recognized. <strong>Conclusions:</strong> Sedentary movements such as sitting were difficult to recognize. However, changing clothes, standing, lying in a supine position, brushing teeth, moving luggage, going up/down the stairs, walking, and running were recognizable. These recognized activities tend to cause shortness of breath in COPD patients. Thus, the findings confirm the feasibility of an algorithm that could determine the severity of COPD by comparing activity amount for each of these activities with those of the healthy people of the same age.

Paediatric sleep apnea event prediction using nasal air pressure and machine learning.

Sleep-disordered breathing is an important health issue for children. The objective of this study was to develop a machine learning classifier model for the identification of sleep apnea events taken exclusively from nasal air pressure measurements acquired during overnight polysomnography for paediatric patients. A secondary objective of this study was to differentiate site of obstruction exclusively from hypopnea event data using the model. Computer vision classifiers were developed via transfer learning to either normal breathing while asleep, obstructive hypopnea, obstructive apnea or central apnea. A separate model was trained to identify site of obstruction as either adeno-tonsillar or tongue base. In addition, a survey of board-certified and board-eligible sleep physicians was completed to compare clinician versus model classification performance of sleep events, and indicated very good performance of our model relative to human raters. The nasal air pressure sample database available for modelling comprised 417 normal, 266 obstructive hypopnea, 122 obstructive apnea and 131 central apnea events derived from 28 paediatric patients. The four-way classifier achieved a mean prediction accuracy of 70.0% (95% confidence interval [67.1-72.9]). Clinician raters correctly identified sleep events from nasal air pressure tracings 53.8% of the time, whereas the local model was 77.5% accurate. The site of obstruction classifier achieved a mean prediction accuracy of 75.0% (95% confidence interval [68.7-81.3]). Machine learning applied to nasal air pressure tracings is feasible and may exceed the diagnostic performance of expert clinicians. Nasal air pressure tracings of obstructive hypopneas may "encode" information regarding the site of obstruction, which may only be discernable by machine learning.

Deficient butyrate-producing capacity in the gut microbiome is associated with bacterial network disturbances and fatigue symptoms in ME/CFS

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is characterized by unexplained debilitating fatigue, cognitive dysfunction, gastrointestinal disturbances, and orthostatic intolerance. Here, we report a multi-omic analysis of a geographically diverse cohort of 106 cases and 91 healthy controls that revealed differences in gut microbiome diversity, abundances, functional pathways, and interactions. Faecalibacterium prausnitzii and Eubacterium rectale, which are both recognized as abundant, health-promoting butyrate producers in the human gut, were reduced in ME/CFS. Functional metagenomics, qPCR, and metabolomics of fecal short-chain fatty acids confirmed a deficient microbial capacity for butyrate synthesis. Microbiome-based machine learning classifier models were robust to geographic variation and generalizable in a validation cohort. The abundance of Faecalibacterium prausnitzii was inversely associated with fatigue severity. These findings demonstrate the functional nature of gut dysbiosis and the underlying microbial network disturbance in ME/CFS, providing possible targets for disease classification and therapeutic trials.

A data-driven interpretable method to predict capacities of metal ion doped TiO2 anode materials for lithium-ion batteries using machine learning classifiers

Machine learning classifier models were built with the datasets of different ions doped into TiO2 materials to predict their charging and discharging performance.

Analysing Factors Influencing Women Unemployment Using a Random Forest Model

The unemployment crisis has been a persistent issue for both developed and developing countries, resulting in an economic indicator deficit. Women are at a disadvantage and continue to encounter significant obstacles to gaining employment. Nigeria, like many other developing countries with high unemployment rates, has a 33% unemployment rate. Consequently, there has been minimal research on the factors that affect women's unemployment. As a result, the purpose of this study investigates the factors that influence women's unemployment in Nigeria. Although the Random Forest model has been widely applied to classification issues, there is a gap in the literature's use of the random forest as a predictor for analyzing factors influencing women's unemployment. The random forest model was employed in this study because of its characteristics such as strong learning ability, robustness, and feasibility of the hypothesis space. As a result, the Random forest prediction model was benchmarked with seven different cutting-edge classical machine learning prediction models, which include the J48 pruned tree, Support Vector Machine, AdaBoost, Logistic Regression, Naive Bayes, Logistic Model Tree, Bagging, and Random Forest. The experimental results demonstrate that Random Forest outperformed the other seven machine learning classifier models using ten commonly used performance evaluation metrics. According to the study's findings, age groups, ethnicity, marital status, and religion were the essential factors affecting women's unemployment in Nigeria. Keywords: machine learning, National Demographic and Health Survey, random forest, influencing factors, women unemployment. DOI: https://doi.org/10.55463/hkjss.issn.1021-3619.60.38

Liver disease prediction using ensemble model

Despite the fact that medical science has progressed to the point where it is capable of identifying and prescribing the exact cure for known diseases, traditional medicinal practices are outdated and slow and it can take so long to identify a disease that it becomes fatal to patients who seek immediate attention. Because the liver plays such a crucial role in eliminating toxins from our bodies, this research focuses mostly on the prediction of liver disorders such as jaundice and various forms of hepatitis. As a result, early detection and treatment of liver illnesses are critical for successful diagnosis and recovery. For the purpose of demonstrating our work, we used examples of various liver illnesses. We used a disease dataset with a variety of symptoms as features to predict the disease in this paper. Before fitting the data into the Machine Learning model, various data pretreatment techniques and a filter-based feature selection strategy are employed on it. Three different supervised machine learning models, SOM, Back propagation and Ensemble model are used to train the data. The accuracy of these entire machine learning classifier models was evaluated and it was discovered that the ensemble Classifier produced the best results, with nearly 99.18 percent accuracy.

Residual Neural Network for the Accurate Recognition of Human Action and Compared with Bayesian Regression

Aim: In this research article, the aim is to analyze and compare the performance of Residual Neural Network and Bayesian Regression for accurate recognition of human actions. Materials and Methods: The proposed machine learning classifier model uses 80% of the UCF101 dataset for training and the remaining 20% for testing. For the SPSS analysis, the results of two classifiers are grouped with 20 samples in each group. The sample size is determined using a pretest with G-power, with a sample size of 80%, a confidence interval of 95%, and a significance level of 0.014 (p&lt;0.05). Result: The findings suggest that the novel residual neural network classifier and Bayesian regression classifier achieved accuracy rates of 95.63% and 93.97%, respectively, in identifying human activities accurately.The statistical significance value between residual neural networks and Bayesian regression has been calculated to be p=0.014 (independent sample t-test p&lt;0.05), indicating a statistically significant difference between the two classifiers.

Enhancing the Intrusion Detection Efficiency using a Partitioning-based Recursive Feature Elimination in Big Cloud Environment

In the era of cloud computing, the effectiveness of utilizing supervised machine-learning-based intrusion detection models for categorizing and detecting malicious network attacks depends on the preparation, extraction, and selection of the optimal subset of features from the dataset. Therefore, before beginning the training phase of the machine learning classifier models, it is required to remove redundant data, manage missing values, extract statistical features from the dataset, and choose the most valuable and appropriate attributes using the Python Jupyter Notebook. In this study, partitioning-based recursive feature elimination (PRFE) method was suggested to decrease the complexity space and training time for machine learning models while increasing the accuracy rate of detecting malicious attacks. On the information security and object technology cloud intrusion dataset (ISOT-CID), some of the most popular supervised machine learning classification techniques, including support vector machines (SVM) and decision trees (DT), have been assessed using the suggested PRFE technique. In comparison to some of the most popular filter and wrapper-based feature selection strategies, the results of the practical experiments demonstrated an improvement in accuracy, recall, F-score, and precision rate after using the PRFE technique on the ISOT-CID dataset. Additionally, the time required to train the machine-learning models was reduced.

Auto-detection of necessity for MRI-guided online adaptive replanning using a machine learning classifier.

MRI-guided adaptive radiation therapy (MRgART), particularly daily online adaptive replanning (OLAR) can substantially improve radiation therapy delivery, however, it can be labor-intensive and time-consuming. Currently, the decision to perform OLAR for a treatment fraction is determined subjectively. In this work, we develop a machine learning algorithm based on structural similarity index measure (SSIM) and change in entropy to quickly and objectively determine whether OLAR is necessary for a daily MRI set. A total of 109 daily MRI sets acquired on a 1.5T MR-Linac during MRgART for 22 pancreatic cancer patients each treated with five fractions were retrospectively analyzed. For each daily MRI set, OLAR and reposition (No-OLAR) plans were created and the superior plan with the daily fraction determined per clinical dose-volume criteria. SSIM and entropy maps were extracted from each daily MRI set, with respect to its reference (e.g., dry-run) MRI in the region enclosed by 50-100% isodose surfaces. A total of six common features were extracted from SSIM maps. Pearson's rank correlation coefficient was utilized to rule out redundant SSIM features. A t-test was used to determine significant SSIM features which were combined with the change in entropy to develop anensemble machine classifier with fivefold cross validation. The performance of the classifier was evaluated using the area under the curve (AUC) of the receiver operating characteristic curve. A machine learning classifier model using two SSIM features (mean and full width at half maximum) and change in entropy was determined to be able to significantly discriminate between No-OLAR and OLAR groups. The obtained machine learning ensemble classifier can predict OLAR necessity with a cross validated AUC of 0.93. Misclassification was found primarily for No-OLAR cases with dosimetric plan quality closely comparable to the corresponding OLAR plans, thus, are not a major practical concern. A machine learning classifier based on simple first-order image features, that is, SSIM features and change in entropy, was developed to determine when OLAR is necessary for a daily MRI set with practical acceptable prediction accuracy. This classifier may be implemented in the MRgART process to automatically and objectively determine if OLAR is required following daily MRI.