Supervised Machine Learning Models for Liver Disease Risk Prediction

Computers, Journal Year: 2023, Volume and Issue: 12(1), P. 19 - 19

Published: Jan. 13, 2023

The liver constitutes the largest gland in human body and performs many different functions. It processes what a person eats drinks converts food into nutrients that need to be absorbed by body. In addition, it filters out harmful substances from blood helps tackle infections. Exposure viruses or dangerous chemicals can damage liver. When this organ is damaged, disease develop. Liver refers any condition causes may affect its function. serious threatens life requires urgent medical attention. Early prediction of using machine learning (ML) techniques will point interest study. Specifically, content research work, various ML models Ensemble methods were evaluated compared terms Accuracy, Precision, Recall, F-measure area under curve (AUC) order predict occurrence. experimental results showed Voting classifier outperforms other with an accuracy, recall, 80.1%, precision 80.4%, AUC equal 88.4% after SMOTE 10-fold cross-validation.

Language: Английский

---

Efficient Data-Driven Machine Learning Models for Cardiovascular Diseases Risk Prediction

Sensors, Journal Year: 2023, Volume and Issue: 23(3), P. 1161 - 1161

Published: Jan. 19, 2023

Cardiovascular diseases (CVDs) are now the leading cause of death, as quality life and human habits have changed significantly. CVDs accompanied by various complications, including all pathological changes involving heart and/or blood vessels. The list includes hypertension, coronary disease, failure, angina, myocardial infarction stroke. Hence, prevention early diagnosis could limit onset or progression disease. Nowadays, machine learning (ML) techniques gained a significant role in disease prediction an essential tool medicine. In this study, supervised ML-based methodology is presented through which we aim to design efficient models for CVD manifestation, highlighting SMOTE technique's superiority. Detailed analysis understanding risk factors shown explore their importance contribution prediction. These fed input features plethora ML models, trained tested identify most appropriate our objective under binary classification problem with uniform class probability distribution. Various were evaluated after use non-use Synthetic Minority Oversampling Technique (SMOTE), comparing them terms Accuracy, Recall, Precision Area Under Curve (AUC). experiment results showed that Stacking ensemble model 10-fold cross-validation prevailed over other ones achieving Accuracy 87.8%, Recall 88.3%, 88% AUC equal 98.2%.

Language: Английский

---

Long-Term Coronary Artery Disease Risk Prediction with Machine Learning Models

Sensors, Journal Year: 2023, Volume and Issue: 23(3), P. 1193 - 1193

Published: Jan. 20, 2023

The heart is the most vital organ of human body; thus, its improper functioning has a significant impact on life. Coronary artery disease (CAD) coronary arteries through which nourished and oxygenated. It due to formation atherosclerotic plaques wall epicardial arteries, resulting in narrowing their lumen obstruction blood flow them. can be delayed or even prevented with lifestyle changes medical intervention. Long-term risk prediction will area interest this work. In specific research paper, we experimented various machine learning (ML) models after use non-use synthetic minority oversampling technique (SMOTE), evaluating comparing them terms accuracy, precision, recall an under curve (AUC). results showed that stacking ensemble model SMOTE 10-fold cross-validation prevailed over other models, achieving accuracy 90.9 %, precision 96.7%, 87.6% AUC equal 96.1%.

Language: Английский

---

Stacked neural nets for increased accuracy on classification on lung cancer

Measurement Sensors, Journal Year: 2024, Volume and Issue: 32, P. 101052 - 101052

Published: Feb. 15, 2024

Lung cancer is regarded as one of the most lethal diseases endangering human survival. It difficult to detect lung in its early stages, because ambiguity regions medical images. Healthcare business automating itself with use image recognition and data analytics, much computing sector has completely automated. This article proposes a novel architecture, Stacked Neural Network (SNN), for detection classification using CT scan data. The goal proposed technique investigate accuracy levels different Networks (NN) determine stage cancer. effective processing images, classifying detecting nodules, extracting features, predicting deep learning. First, areas are extracted techniques. SNN used segmentation process. Various neural network techniques utilised process once features retrieved from segmented pictures. suggested methods' performances assessed F1-Measure, accuracy, precision, recall metrics. 96% shown testing findings, which comparatively greater than other methods currently use. Proposed algorithm clearly supported real-world clinical practice.

Language: Английский

---

Optimizing double-layered convolutional neural networks for efficient lung cancer classification through hyperparameter optimization and advanced image pre-processing techniques

BMC Medical Informatics and Decision Making, Journal Year: 2024, Volume and Issue: 24(1)

Published: May 27, 2024

Abstract Lung cancer remains a leading cause of cancer-related mortality globally, with prognosis significantly dependent on early-stage detection. Traditional diagnostic methods, though effective, often face challenges regarding accuracy, early detection, and scalability, being invasive, time-consuming, prone to ambiguous interpretations. This study proposes an advanced machine learning model designed enhance lung stage classification using CT scan images, aiming overcome these limitations by offering faster, non-invasive, reliable tool. Utilizing the IQ-OTHNCCD dataset, comprising scans from various stages healthy individuals, we performed extensive preprocessing including resizing, normalization, Gaussian blurring. A Convolutional Neural Network (CNN) was then trained this preprocessed data, class imbalance addressed Synthetic Minority Over-sampling Technique (SMOTE). The model’s performance evaluated through metrics such as precision, recall, F1-score, ROC curve analysis. results demonstrated accuracy 99.64%, F1-score values exceeding 98% across all categories. SMOTE enhanced ability classify underrepresented classes, contributing robustness These findings underscore potential in transforming diagnostics, providing high classification, which could facilitate detection tailored treatment strategies, ultimately improving patient outcomes.

Language: Английский

---

Performance of machine learning algorithms for lung cancer prediction: a comparative approach

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Aug. 9, 2024

Due to the excessive growth of PM 2.5 in aerosol, cases lung cancer are increasing rapidly and most severe among other types as highest mortality rate. In cases, is detected with least symptoms at its later stage. Hence, clinical records may play a vital role diagnose this disease correct stage for suitable medication cure it. To detect an accurate prediction method needed which significantly reliable. digital record era advancement computing algorithms including machine learning techniques opens opportunity ease process. Various be applied over realistic data but predictive power yet comprehended results. This paper envisages compare twelve potential eleven along two major habits patients predict positive case accurately. The result has been found based on classification heat map correlation. K-Nearest Neighbor Model Bernoulli Naive Bayes significant methods early prediction.

Language: Английский

---

An Optimized Predictive Machine Learning Model for Lung Cancer Diagnosis

Biomedical & Pharmacology Journal, Journal Year: 2025, Volume and Issue: 18(December Spl Edition), P. 85 - 98

Published: Jan. 20, 2025

Lung cancer is one of the leading causes death worldwide. Increasing patient survival rates requires early detection. Traditional methods diagnosis often result in late-stage detection, necessitating development more advanced and accurate predictive models. This paper has proposed a methodology for lung prediction using machine learning Synthetic minority over-sampling technique (SMOTE) used before classification to resolve problem class imbalance. Bayesian optimization enhance model’s performance. Performance three classifiers adaptive boosting (AdaBoost), random forest (RF), extreme gradient (XGBoost) evaluated both with without hyperparmater optimization. Optimized models RF, AdaBoost XGBoost achieved accuracies 96.11%, 95.74% 95.92% respectively. Results demonstrate effectiveness combining classifiers, SMOTE, hyperparameter tuning improving accuracy.

Language: Английский

---

XML-GBM lung: An explainable machine learning-based application for the diagnosis of lung cancer

Journal of Pathology Informatics, Journal Year: 2023, Volume and Issue: 14, P. 100307 - 100307

Published: Jan. 1, 2023

Lung cancer has been the leading cause of cancer-related deaths worldwide. Early detection and diagnosis lung can greatly improve chances survival for patients. Machine learning increasingly used in medical sector cancer, but lack interpretability these models remains a significant challenge. Explainable machine (XML) is new approach that aims to provide transparency models. The entire experiment performed dataset obtained from Kaggle. outcome predictive model with ROS (Random Oversampling) class balancing technique comprehend most relevant clinical features contributed prediction using explainable termed SHAP (SHapley Additive exPlanation). results show robustness GBM's capacity detect 98.76% accuracy, 98.79% precision, recall, F-Measure, 0.16% error rate, respectively. Finally, mobile app developed incorporating best efficacy our approach.

Language: Английский

---

Machine Learning‐Based Lung Cancer Detection Using Multiview Image Registration and Fusion

Journal of Sensors, Journal Year: 2023, Volume and Issue: 2023(1)

Published: Jan. 1, 2023

The exact lung cancer identification is a critical problem that has attracted the researchers’ attention. practice of multiview single image and segmentation been widely used for last 2 years to improve disease. utilization machine learning (ML) deep (DL) techniques can significantly expedite process detection stage classification, enabling researchers study larger number patients in shorter time frame at reduced cost applying approach herein, multiresolution rigid registration mechanism applied enhance further. Techniques like principle component averaging discrete wavelet transform are verified fusion development. To review performance suggested technique, database resource initiative‐based lungs consortium tested this paper which includes 4,682 computed tomography scan images 61 with nodules sizes from 3 30 mm. According finding, outperformed results our model obtained terms feature mutual information, peak signal‐to‐noise ratio, were recorded 0.80 19.25, respectively. Moreover, stages (STG‐1, STG‐2, STG‐3, STG‐4) also assessed by using ResNet‐18 convolutional neural network classifier. With only 1.8 FP/scan, achieved accuracy sensitivity 98.2% 96.4%, study’s findings show proposed strategy outperforms existing models significantly. Therefore, have potential be implemented clinical settings provide support doctors early diagnosis cancer, while minimizing occurrence false positives scans.

Language: Английский

---

Multi-Layered Non-Local Bayes Model for Lung Cancer Early Diagnosis Prediction with the Internet of Medical Things

Bioengineering, Journal Year: 2023, Volume and Issue: 10(2), P. 138 - 138

Published: Jan. 20, 2023

The Internet of Things (IoT) has been influential in predicting major diseases current practice. deep learning (DL) technique is vital monitoring and controlling the functioning healthcare system ensuring an effective decision-making process. In this study, we aimed to develop a framework implementing IoT DL identify lung cancer. accurate efficient prediction disease challenging task. proposed model deploys process with multi-layered non-local Bayes (NL Bayes) manage early diagnosis. Medical (IoMT) could be useful determining factors that enable sorting quality values through use sensors image processing techniques. We studied by analyzing its results regard specific attributes such as accuracy, quality, efficiency. overcome problems existing practical computational comparison provided low error rate (2%, 5%) increase number instance values. experimental led us conclude can make predictions based on images high sensitivity better precision compared other results. achieved expected accuracy (81%, 95%), specificity (80%, 98%), 99%). This adequate for real-time health systems cancer

Language: Английский

---