| Titre : | Prediction of heart disease using deep learning |
| Auteurs : | AKILA BEN ZEID, Auteur ; Fatima Zohra Torki, Directeur de thèse |
| Type de document : | Mémoire magistere |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2025 |
| Format : | 1 vol. (77 p.) / ill.couv.ill.encoul / 30cm |
| Langues: | Anglais |
| Langues originales: | Anglais |
| Résumé : |
Heart disease is a serious condition and is considered the leading cause of death in many countries. Without early detection, the condition of patients affected by heart disease can worsen until it becomes too late for effective treatment. One diagnostic method that can aid in the early detection of heart disease is electrocardiography (ECG), due to its low cost. However, interpreting ECG signals is challenging and time-consuming. Recent advances in deep learning offer a potential solution by assisting both doctors and patients in the early detection of heart disease through the classification of ECG records.In this work, we discussed heart disease, ECG and its limitations, as well as machine learning and deep learning approaches used in this field, each with its own strengths. In our project, we proposed an LSTM Autoencoder model to support early prediction of the disease. The model was trained on an ECG dataset and achieved promising results. |
| Sommaire : |
Contents General Introduction ......1 Chapter 1: Heart Diseases and Electrocardiography 1.1. Introduction .....4 1.2. Cardiovascular diseases (Heart diseases) .........4 1.3. Types of cardiovascular diseases .... 4 Coronary heart disease (CAD) ..5 Definition ......5 Cause .....5 Symptoms ........6 Diagnostic .....6 Treatments ....7 Stroke .......8 Definition .......8 Cause .............8 Symptoms .........8 Diagnostic ...............9 Treatments ...............9 Peripheral arterial disease (PAD) ............... 10 Definition ............. 10 Cause ......... 10 Symptoms ......... 10 Diagnostic ............... 11 Treatments ......... 12 Aortic disease ......... 12 Definition ........... 12 Cause .......... 13 Symptoms ...... 14 Diagnostic .......... 15 Treatments ........... 15 1.4. Complications of heart disease .... 16 1.5. Prevention of heart disease ........ 17 1.6. Electrocardiography (ECG) .......18 ECG Signal Components ................................................................................................. 18 ECG Rhythms ..... 19 Regular Rhythms ..... 19 Irregular Rhythms ..... 20 Electrocardiogram Lead Setup 22 Electrode Placement ....... 23 ECG Limitations...... 24 1.7. Conclusion........ 25 Chapter 2: Machine and Deep learning approaches 2.1. Introduction ......... 27 2.2. Machine Learning ... 27 Types of Machine Learning ... 28 Supervised learning....... 28 Unsupervised learning .... 29 Semi-supervised learning . 30 Self-supervised learning ... 30 2.3. Deep learning ... 31 Convolutional Neural Networks (CNN) ....... 32 Recurrent Neural Networks (RNN) ....................... 33 Long Short-Term Memory (LSTM) ....... 33 Gated Recurrent Unit (GRU) ... 34 Autoencoder ...... 35 LSTM-Autoencoder (LSTM-AE)......... 35 2.4. Related Works ..... 36 CNN-FWS: A Model for the Diagnosis of Normal and Abnormal ECG with Feature Adaptive 36 Dynamic prediction of cardiovascular disease using improved LSTM .... 37 A CNN based model for heart disease detection ........ 38 2.5. Conclusion........ 39 Chapter 3: Design of the approach 3.1. Introduction ........ 41 3.2. General workflow ........ 41 3.3. Dataset..... 41 Dataset structure ... 42 3.4. Preprocessing ......... 42 3.5. Proposed model ............. 43 Attention Mechanism......... 44 Activation Functions ....... 44 Learning rate ...... 44 Optimization Algorithm ...... 44 Loss Function ......... 45 Prediction ............. 46 Evaluation metrics ........... 46 3.6. Conclusion... 46 Chapter 4: Implementation and results 4.1. Introduction .......... 48 4.2. Implementation frameworks, tools and libraries ............... 48 4.3. Implementation phases ......... 50 Loading dataset ......... 50 Preprocessing ........ 51 Data Filtering ....... 51 Selection of input features and output target ........ 51 Data augmentation ...... 52 Data Reshaping ...... 52 LSTM Autoencoder Model .53 4.4. Results ..... 55 C onfusion matrix ... 57 Comparison with other related works ........ 57 4.5. Interface Design ....... 58 4.6. Model Integration and System Deployment................... 61 Model Integration ... 61 System Deployment ..... 62 4.7. Conclusion....... 63 General Conclusion . 64 Bibliographies ...... |
| Type de document : | Mémoire master |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| MINF/957 | Mémoire master | bibliothèque sciences exactes | Consultable |
