Recent Advances in Computer-Aided Medical Diagnosis Using Machine Learning Algorithms With Optimization Techniques

Background Peripheral Arterial Disease(PAD) is a prevalent cardiovascular condition affecting millions worldwide, and the advent of machine learning(ML) techniques offers promising opportunities to enhance risk prediction and personalise patient management in this complex disease. This systematic review and meta-analysis aimed to evaluate the performance of ML models in predicting adverse outcomes in PAD patients compared to traditional statistical approaches. Methods A comprehensive search of major databases(MEDLINE,EMBASE,CINAHL,Cochrane CENTRAL) was conducted for studies published between 2000 and 2024. Two reviewers independently screened studies, extracted data, and assessed risk of bias using the PROBAST tool. The quality of included studies was also assessed using criteria outlined by Qiao et al. A modified Hierarchical Summary Receiver Operating Characteristic(HSROC) analysis was performed to compare the predictive performance of various ML models and traditional regression methods. Results Thirteen studies met the inclusion criteria. Gradient boosted models demonstrated the highest predictive performance with a diagnostic odds ratio(DOR) of 36.593(95% CI:24.44-54.79), sensitivity of 0.853, and specificity of 0.863. The best performing ML models for outcomes within 2 years, showed a DOR of 34.169(95% CI:22.840-51.117). Traditional regression models consistently underperformed compared to ML approaches, with the lowest DOR of 3.326(95% CI:1.814-6.097). The quality assessment revealed a mix of methodological rigor, with 54% of studies rated as low risk of bias and 46% as unclear. Conclusion Advanced ML techniques demonstrate superior predictive power for adverse outcomes in PAD patients compared to traditional regression methods.

e17570 Background: Ovarian cancer(OC) is the leading cause of death from gynecologic malignancy. Current challenges include lack of diagnostic tools, predictive biomarkers, and identifying appropriate surgical candidates. Machine learning(ML) is an emerging field that can make accurate projections by making inferences on data and may play a crucial role in OC.The objective of the current study was to review the literature on application of ML in OC and report the most commonly used algorithms and their performance in comparison to existing prediction tools and traditional regression models. Methods: This is a systematic review of published literature from January 1985 to March 2021 on the use of ML in OC. An extensive search of electronic library databases was conducted. Four independent reviewers screened the articles initially by title then full text. Quality was assessed using the MINORS criteria. P-values were generated using the Pearson’s Chi-squared(x2) test to compare performance of ML models with traditional statistics. No p-values were reported if only one study was available. Results: Among 4,295 articles screened, 88 studies on ML in OC were included. The mean age of OC patients was 54.7 years(11-90) and the most common stages at diagnosis were:Stage III (39.9%) and IV (34%). Applications of ML were in clinical datasets(33%, n = 29), preoperative diagnostics(30.7%, n = 27), serum biomarkers (21.6%, n = 19), genomics (12.5%, n = 11), and prediction of cytoreductive outcomes (2.3%, n = 2). The most commonly applied algorithms were Support Vector Machine [SVM](28%, n = 33)and Neural Networks[NN] (25.28%). Over the past decades, the number of publications on ML in OC increased three-fold from 20(1994-2010) to 67 (2011–2021). Only 9 (10%) studies compared ML techniques with existing prediction tools, or traditional regression models. Among 29 clinical dataset studies, 4 compared ML with traditional logistic regression(LR). Two studies reported better performance with ML compared to LR but not significant(accuracy: 0.88 vs 0.84, p = 0.15), one study performed comparably(accuracy: 0.1 vs 0.1) while one study performed worse(accuracy: 0.1 vs 0.97). Only one preoperative diagnostic study compared ML techniques with LR. SVM classifiers outperformed LR in classifying ovarian masses as benign or malignant(sensitivity: 0.88 vs. 0.70). One serum biomarker study compared LR with ML algorithms; LR performed better using two biomarkers for predicting OC(accuracy: 0.97 vs. 0.94). Among five studies reporting overall survival outcomes, only one study compared survival ML techniques using NN with LR and showed that NN classifiers outperformed LR in predicting overall survival(AUC: 0.72 vs. 0.62). Conclusions: This is the first systematic review exploring the literature on ML algorithms in OC. Most ML models outperformed traditional models. However, larger datasets would be required to validate findings.

The traditional methods of cancer diagnosis and cancer-type recognition have quite a large number of limitations in terms of speed and accuracy. However, recent studies on cancer diagnosis are focused on molecular level identification so as to improve the capability of diagnosis process. By statistically analyzing the heart cancer datasets using a set of protocols and algorithms, gene expression profiles are efficiently analyzed. Various machine learning classifiers are used to classify the selected data. Cross-validation was performed to avoid overfitting and different ratios of training, and testing data was used to conclude the best optimization technique and classification algorithm for the heart cancer datasets. The data is optimized using optimization techniques like particle swarm optimization (PSO), grey wolf optimization (GWO), and hybrid particle swarm optimization with grey wolf optimizer (HPSOGWO). Results show an improvement in the prediction accuracy of heart cancer by the hybrid algorithm as compared to PSO and GWO algorithms.

Machine learning (ML) models provide more choices to patients with diabetes mellitus (DM) to more properly manage blood glucose (BG) levels. However, because of numerous types of ML algorithms, choosing an appropriate model is vitally important. In a systematic review and network meta-analysis, this study aimed to comprehensively assess the performance of ML models in predicting BG levels. In addition, we assessed ML models used to detect and predict adverse BG (hypoglycemia) events by calculating pooled estimates of sensitivity and specificity. PubMed, Embase, Web of Science, and Institute of Electrical and Electronics Engineers Explore databases were systematically searched for studies on predicting BG levels and predicting or detecting adverse BG events using ML models, from inception to November 2022. Studies that assessed the performance of different ML models in predicting or detecting BG levels or adverse BG events of patients with DM were included. Studies with no derivation or performance metrics of ML models were excluded. The Quality Assessment of Diagnostic Accuracy Studies tool was applied to assess the quality of included studies. Primary outcomes were the relative ranking of ML models for predicting BG levels in different prediction horizons (PHs) and pooled estimates of the sensitivity and specificity of ML models in detecting or predicting adverse BG events. In total, 46 eligible studies were included for meta-analysis. Regarding ML models for predicting BG levels, the means of the absolute root mean square error (RMSE) in a PH of 15, 30, 45, and 60 minutes were 18.88 (SD 19.71), 21.40 (SD 12.56), 21.27 (SD 5.17), and 30.01 (SD 7.23) mg/dL, respectively. The neural network model (NNM) showed the highest relative performance in different PHs. Furthermore, the pooled estimates of the positive likelihood ratio and the negative likelihood ratio of ML models were 8.3 (95% CI 5.7-12.0) and 0.31 (95% CI 0.22-0.44), respectively, for predicting hypoglycemia and 2.4 (95% CI 1.6-3.7) and 0.37 (95% CI 0.29-0.46), respectively, for detecting hypoglycemia. Statistically significant high heterogeneity was detected in all subgroups, with different sources of heterogeneity. For predicting precise BG levels, the RMSE increases with a rise in the PH, and the NNM shows the highest relative performance among all the ML models. Meanwhile, current ML models have sufficient ability to predict adverse BG events, while their ability to detect adverse BG events needs to be enhanced. PROSPERO CRD42022375250; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=375250.

Understanding the Performance of Machine Learning Models to Predict Credit Default: A Novel Approach for Supervisory Evaluation

Cervical cancer is a highly prevalent malignancy affecting women worldwide, ranking as the seventh most common cancer globally. This study aims to systematically review and analyze cervical cancer survival predictions using machine learning (ML) algorithms. A comprehensive search was conducted across Scopus and PubMed databases in February 2024. Extracted articles were screened using Hubmeta software, with duplicates and non-relevant studies excluded. The final selection, comprising 24 articles, focused on survival predictions through ML techniques. These studies, published mostly post-2019, included datasets ranging from 75 to 9,462 cervical cancer patients and up to 91,294 squamous cell samples. The most commonly applied ML models were Random Forest (RF), Neural Networks (NN), Support Vector Machines (SVM), Ensemble and Hybrid Learning, and Deep Learning (DL). The area under the curve (AUC) for these models ranged from 0.84 to 0.9875, demonstrating their strong predictive capabilities. Clinical patient records were the primary data source. Meta-analysis was performed on the extracted data using GraphPad Prism for descriptive statistics and One-Way ANOVA. No significant differences were found between group means, as evidenced by an R-squared value of 0.1459. This result indicates that the independent variable (year of study) explained only 14.59% of the variance in ML model performance. The study found that the use of ML models has increased over time, particularly with Convolutional Neural Networks (CNNs) such as the ResNet50 model, which demonstrated superior accuracy metrics, including over 90% accuracy for the ResNet152 variant. These findings suggest that integrating multi-dimensional data with ML models holds significant potential for improving survival predictions in cervical cancer patients. Future research is recommended to develop tailored ML algorithms with even higher predictive accuracy for cervical cancer survival.

The development of machine learning (ML) approaches in materials science offers the opportunity to exploit existing engineering and developmental alloy datasets, such as Oak Ridge National Laboratory (ORNL)’s consistently measured creep-rupture dataset for alumina-forming austenitic (AFA) alloys, to accelerate their further development. As a first step toward achieving ML insights for improved alloy design, the potential sources of uncertainty and their impacts on ML output are examined. It is observed that the selection of algorithms and features as well as data sampling significantly affects the performance of ML models, either positively or negatively. The performance of various ML models in predicting the creep properties of AFA alloys is compared, with further evaluation by assessment of a small set of new developmental AFA alloys that were not part of the training dataset. The present study demonstrates that uncertainty quantification (UQ) is essential in materials science for evaluating the performance of ML algorithms with specifically selected feature sets and obtaining a comprehensive understanding of their limitations and the resultant capability of effective prediction in complex materials systems.

To enhance the performance of machine learning (ML) models for the post-embolization recanalization of cerebral aneurysms, we evaluated the impact of hemodynamic feature derivation and selection method on six ML algorithms. We utilized computational fluid dynamics (CFD) to simulate hemodynamics in 66 cerebral aneurysms from 65 patients, including 57 stable and nine recanalized aneurysms. We derived a total of 107 features for each aneurysm, encompassing four clinical features, 12 morphological features, and 91 hemodynamic features. To investigate the influence of feature derivation and selection methods on the ML models, we employed two derivation methods, simplified and fully derived, in combination with four selection methods: all features, statistically significant analysis, stepwise multivariate logistic regression analysis (stepwise-LR), and recursive feature elimination (RFE). Model performance was assessed using the area under the receiver operating characteristic curve (AUROC) and precision-recall curve (AUPRC) on both the training and testing datasets. The AUROC values on the testing dataset exhibited a wide-ranging spectrum, spanning from 0.373 to 0.863. Fully derived features and the RFE selection method demonstrated superior performance in intra-model comparisons. The multi-layer perceptron (MLP) model, trained with RFE-selected fully derived features, achieved the best performance on the testing dataset, with an AUROC value of 0.863 (95% CI: 0.684- 1.000). Our study demonstrated the importance of feature derivation and selection in determining the performance of ML models. This enabled the development of accurate decision-making models without the need to invade the patient.

Performance evaluation of machine learning techniques in predicting cumulative absolute velocity

Biogas plant operators often face huge challenges in the monitoring, controlling and optimisation of the anaerobic digestion (AD) process, as it is very sensitive to surrounding changes, which often leads to process failure and adversely affects biogas production. Conventional implemented methods and mechanistic models are impractical and find it difficult to model the nonlinear and intricate interactions of the AD process. Thus, the development of machine learning (ML) algorithms has attracted considerable interest in the areas of process optimization, real-time monitoring, perturbation detection and parameter prediction. This paper provides a comprehensive and up-to-date overview of different machine learning algorithms, including artificial neural network (ANN), fuzzy logic (FL), adaptive network-based fuzzy inference system (ANFIS), support vector machine (SVM), genetic algorithm (GA) and particle swarm optimization (PSO) in terms of working mechanism, structure, advantages and disadvantages, as well as their prediction performances in modelling the biogas production. A few recent case studies of their applications and limitations are also critically reviewed and compared, providing useful information and recommendation in the selection and application of different ML algorithms. This review shows that the prediction efficiency of different ML algorithms is greatly impacted by variations in the reactor configurations, operating conditions, influent characteristics, selection of input parameters and network architectures. It is recommended to incorporate mixed liquor volatile suspended solids (MLVSS) concentration of the anaerobic digester (ranging from 16,500 to 46,700mg/L) as one of the input parameters to improve the prediction efficiency of ML modelling. This review also shows that the combination of different ML algorithms (i.e. hybrid GA-ANN model) could yield better accuracy with higher R2 (0.9986) than conventional algorithms and could improve the optimization model of AD. Besides, future works could be focused on the incorporation of an integrated digital twin system coupled with ML techniques into the existing Supervisory Control and Data Acquisition (SCADA) system of any biogas plant to detect any operational abnormalities and prevent digester upsets.

Optimizing asphalt mix design through predicting effective asphalt content and absorbed asphalt content using machine learning

In recent years, infections and damage caused by malware have increased at exponential rates. At the same time, machine learning (ML) techniques have shown tremendous promise in many domains, often out performing human efforts by learning from large amounts of data. Results in the open literature suggest that ML is able to provide similar results for malware detection, achieving greater than 99% classification accuracy [49]. However, the same detection rates when applied in deployed settings have not been achieved. Malware is distinct from many other domains in which ML has shown success in that (1) it purposefully tries to hide, leading to noisy labels and (2) often its behavior is similar to benign software only differing in intent, among other complicating factors. This report details the reasons for the difficultly of detecting novel malware by ML methods and offers solutions to improve the detection of novel malware. We propose to detect malware by detecting behaviors commonly exhibited by malware such as DLL injection, and process hollowing. This is based on the assumption that there is a set of behaviors that are common to most malware samples and detecting them will generalize to novel malware. Additionally, detected behaviors point analysts toward appropriate handling and mitigation strategies, which is not the case with a binary benign/malicious classification. A behavior labeling method was developed and was used to label an existing malware dataset. Results show that detecting malicious behaviors is much more difficult than simply classifying malware and goodware?achieving 80% accuracy compared to reported 99% accuracy from classifying malware and goodware. This drop is due to several reasons which are detailed in the report. We also propose to evaluate the performance of detecting novel malware by holding out a malware family for testing and training on the other families. Traditional ML evaluation will shuffle the data and then split the data into training and testing. Our method addresses the use-case when novel malware families are encountered and they require more than just a malicious or benign designation. Our results suggest that this type of evaluation is much more difficult than traditional methods and provides more realistic results, albeit, significantly worse. For our behavior detection, accuracy decreases from 80% to 68% across all behaviors when holding out a malware family from training. We show that the degradation in performance is because each malware family has distinct characteristics resulting in high extrapolations by an ML model. Here, an ML model should return an "I do not know" response and request further analysis from an analyst. We run a number of experiments that compare novel malware families to the training data using different feature representations including a genomics-inspired distance measure and features extracted by deep learning. Generally, held-out families are significantly different from the training data, resulting in unpredictable results. This has been observed generally in the ML community [22, 9]. We empirically demonstrate this in the domain of malware detection. In an attempt to improve the detection of malware behaviors, we examine the impact that additional synthetic data has on the performance of an ML model in detecting behaviors in novel malware families. We find that while synthetic data does improve the performance of ML models, often simpler methods perform better than more complicated ones. Two generative modeling techniques were examined to produce synthetic malware samples such that the behaviors present are able to be specified externally. The difficulty is due to finer grained analysis of the executable and modifying the problem from a binary classification problem to a multi-label problem. The addition of synthetic data increases the overall accuracy from 68% to 70%. While far less accurate than measures presented in academic analyses, we believe that this is more representative of real-world performance and allows models to be properly placed within a malware detection system. We suggest that in highly dynamic environments ML pipelines should determine whether an ML model is competent in the area of new data and should involve mechanisms to improve over time with a human in the loop.

ObjectiveEnhanced technology in computer and internet has driven scale and quality of data to be improved in various areas including healthcare sectors. Machine Learning (ML) has played a pivotal role in efficiently analyzing those big data, but a general misunderstanding of ML algorithms still exists in applying them (e.g., ML techniques can settle a problem of small sample size, or deep learning is the ML algorithm). This paper reviewed the research of diagnosing mental illness using ML algorithm and suggests how ML techniques can be employed and worked in practice.MethodsResearches about mental illness diagnostic using ML techniques were carefully reviewed. Five traditional ML algorithms-Support Vector Machines (SVM), Gradient Boosting Machine (GBM), Random Forest, Naïve Bayes, and K-Nearest Neighborhood (KNN)-frequently used for mental health area researches were systematically organized and summarized.ResultsBased on literature review, it turned out that Support Vector Machines (SVM), Gradient Boosting Machine (GBM), Random Forest, Naïve Bayes, and K-Nearest Neighborhood (KNN) were frequently employed in mental health area, but many researchers did not clarify the reason for using their ML algorithm though every ML algorithm has its own advantages. In addition, there were several studies to apply ML algorithms without fully understanding the data characteristics.ConclusionResearchers using ML algorithms should be aware of the properties of their ML algorithms and the limitation of the results they obtained under restricted data conditions. This paper provides useful information of the properties and limitation of each ML algorithm in the practice of mental health.

SummaryAccurate prediction of the solubility of chemical substances in solvents remains a challenge. The sparsity of high-quality solubility data is recognized as the biggest hurdle in the development of robust data-driven methods for practical use. Nonetheless, the effects of the quality and quantity of data on aqueous solubility predictions have not yet been scrutinized. In this study, the roles of the size and the quality of data sets on the performances of the solubility prediction models are unraveled, and the concepts of actual and observed performances are introduced. In an effort to curtail the gap between actual and observed performances, a quality-oriented data selection method, which evaluates the quality of data and extracts the most accurate part of it through statistical validation, is designed. Applying this method on the largest publicly available solubility database and using a consensus machine learning approach, a top-performing solubility prediction model is achieved.

The machine learning (ML) and Internet of things (IoT) technologies are increasingly focussed on decision tree classification algorithm. Its use is expanding through numerous fields, solving incredibly complex problems. DTCA is also being used in medical health data using computer-aided diagnosis to identify chronic kidney diseases like cancer and diabetes. Deep learning is a class of machine learning that utilizes neural networks to solve problems and learn unsupervised from unstructured or unlabelled data. The DL used to deep stacked auto-encoder and softmax classifier methods is applied for CKD. In this work, based an R Studio and Python Colab software using random forest, SVM, C5.0, decision tree classification algorithm, C4.5, ANN, neuro-fuzzy systems, classification and clustering, CNN, RNN, MLP is used to predict multiple machine and deep learning techniques, discover an early diagnosis of CKD patients. In this work, classify the chronic kidney disease various stages are identified.KeywordsDecision tree classification algorithmComputational decision support systemChronic kidney diseaseDeep and machine learning algorithmsIoT and CKD stages