Automated heart disease prediction using improved explainable learning-based technique

Neural Computing and Applications, Год журнала: 2024, Номер 36(26), С. 16289 - 16318

Опубликована: Май 25, 2024

Язык: Английский

---

Unsupervised Learning for Heart Disease Prediction: Clustering-Based Approach

ITM Web of Conferences, Год журнала: 2025, Номер 74, С. 01005 - 01005

Опубликована: Янв. 1, 2025

This paper on the prediction of heart disease addresses application unsupervised machine learning algorithms, digs up latent pattern risk in data patients for early diagnosis, and intervenes. We have compared models K-Means Clustering, DBSCAN, Agglomerative Gaussian Mixture Model, Spectral wherein brought out best result that happened to be 84 percent with groups formed using nuanced indicators. For such insights, project embeds an HTML web-based interface where healthcare professionals alike can easily read predictions. approach advances predictive accuracy, yet brings medical profession incredibly powerful tool a more personalized type care. Providers would then ability identify ahead time high-risk people monitor their care carefully. It, however, opens possibility health analytics shows how this applied role detection targeted treatment, thereby contributing better patient outcomes proactivity managing risks.

Язык: Английский

---

Optimized Feature Selection and Deep Neural Networks to Improve Heart Disease Prediction

Deleted Journal, Год журнала: 2025, Номер unknown

Опубликована: Апрель 16, 2025

Heart disease remains a significant health threat due to its high mortality rate and increasing prevalence. Early prediction using basic physical markers from routine exams is crucial for timely diagnosis intervention. However, manual analysis of large datasets can be labor-intensive error-prone. Our goal rapidly reliably anticipate cardiac variety body signs. This research presents unique model heart prediction. We provide system predicting that blends the deep convolutional neural network with feature selection technique based on LinearSVC. integrated method selects subset characteristics are strongly linked disease. feed these features into conventual we constructed. Also improve speed predictor avoid gradient varnishing or explosion, network's hyperparameters were tuned random search algorithm. The proposed was evaluated UCI MIT datasets. number indicators, such as accuracy, recall, precision, F1 score. results demonstrate our attains accuracy rates 98.16%, 98.2%, 95.38%, 97.84% in dataset, an average MCC score 90%. These affirm efficacy reliability predict

Язык: Английский

---

A smart CardioSenseNet framework with advanced data processing models for precise heart disease detection

Computers in Biology and Medicine, Год журнала: 2024, Номер 185, С. 109473 - 109473

Опубликована: Дек. 3, 2024

Язык: Английский

---

DECNet: Left Atrial Pulmonary Vein Class Imbalance Classification Network

Deleted Journal, Год журнала: 2024, Номер unknown

Опубликована: Авг. 20, 2024

In clinical practice, the anatomical classification of pulmonary veins plays a crucial role in preoperative assessment atrial fibrillation radiofrequency ablation surgery. Accurate vein anatomy assists physicians selecting appropriate mapping electrodes and avoids causing arterial hypertension. Due to diverse subtly different classifications veins, as well imbalance data distribution, deep learning models often exhibit poor expression capability extracting features, leading misjudgments affecting accuracy. Therefore, order solve problem unbalanced left this paper proposes network integrating multi-scale feature-enhanced attention dual-feature extraction classifiers, called DECNet. The utilizes information guide reinforcement generating channel weights spatial enhance features. classifier assigns fixed number channels each category, equally evaluating all categories, thus alleviating bias overfitting caused by imbalance. By combining two, features is strengthened, achieving accurate morphology providing support for subsequent treatment. proposed method evaluated on datasets provided People's Hospital Liaoning Province publicly available DermaMNIST dataset, average accuracies 78.81% 83.44%, respectively, demonstrating effectiveness approach.

Язык: Английский

---

Myocardial Infarction Diagnosis: Pattern Analysis of ECG Report Images Using Machine Learning Techniques

Опубликована: Апрель 26, 2024

The ECG machine data is utilized to diagnose cardiac conditions, specifically focusing on identifying myocardial infarction rates by analyzing pattern variations within report images. Variations in the output of electrodes 2 and 3 are noted as indicative a heart attack. authors employ various image processing techniques like thresholding, contrast enhancement learning methods SVM, GBC, k-neighbors process these patterns, aiming enhance accuracy. After extracting four features, most effective classifiers employed, with Gradient Boosting Classifier (GBC) set features exhibiting highest accuracy at 76.60%. This paper emphasizes preprocessing crucial for obtaining structured refined data, facilitating better feature selection extraction from graph It underscores distinctive aid rate prediction. evaluates several machines classifiers, highlighting their efficiency simplifying expediting diagnosis process. Furthermore, research suggests that incorporating additional could potentially improve

Язык: Английский

---

Advancing Patient Care with an Intelligent and Personalized Medication Engagement System

Information, Год журнала: 2024, Номер 15(10), С. 609 - 609

Опубликована: Окт. 4, 2024

Therapeutic efficacy is affected by adherence failure as also demonstrated WHO clinical studies that 50–70% of patients follow a treatment plan properly. Patients’ to prescribed drugs the main reason for morbidity and mortality more cost healthcare services. Adherence medication could be improved with use patient engagement systems. Such systems can include patient’s preferences beliefs in plans, resulting responsive customized treatments. However, one key limitation existing their generic applications. We propose personalized framework using AI methods such Reinforcement Learning (RL) Deep (DL). The proposed Personalized Medication Engagement System (PMES) has two major components. first component PMES based on an RL agent, which trained reports later utilized engage patient. after training, identify each patterns responsiveness observing learning response signs then optimize individual. second system DL used monitor process. additional feature it cloud-based anywhere remotely. Moreover, separately, while part given plan. Thus, advantage work two-fold, i.e., improves minimizes errors.

Язык: Английский

---