| Titre : | Big Data analytics using Artificial Intelligence techniques in medical PHM |
| Auteurs : | Abir Belaala, Auteur ; Sadek Labib Terrissa, Auteur |
| Type de document : | Thése doctorat |
| Année de publication : | 2021 |
| Format : | 1 vol. (113 p.) / couv. ill. en coul / 30 cm |
| Langues: | Anglais |
| Résumé : |
Today, With the development of information technology, the concept of smart healthcare became a trending research area. Smart healthcare uses a new generation of information technologies such as big data, cloud computing, and artificial intelligence(AI). These new techniques helps to transform the traditional medical system to be more intelligent, efficient, convenient, and personalized. Computer-aided diagnosis (CAD) has become one of the major research subjects in medical computing and clinical diagnosis. However, how to efficiently and effectively make accurate diagnosis remains a challenging problem in data-driven models. In this thesis, we are interested in improving the performance of computer-aided diagnostic systems in the medical field by increasing the quality of medical data and the analytical techniques. To this end, several contributions have been proposed. First, we proposed an extension of Prognostic and Health Management (PHM) approaches in order to exploit its potential by adapting advanced industrial diagnostic models to medical diagnostics. Secondly, we focused on improving computer-assisted diagnosis, particularly in the dermatology field, using AI techniques as well as those of Big data. The proposed methods and the results obtained were validated by an extensive comparative analysis using benchmarks and private medical data. |
| Sommaire : |
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i R´esum´e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii Publications of the author . . . . . . . . . . . . . . . . . . . . . . . . . . iii List of figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi List of tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii List of algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii List of abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv I General introduction 1 1 Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 Problem statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 4 Dissertation plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 II Preliminaries and Basic Concepts 7 1 Big data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.1 Features of Big Data . . . . . . . . . . . . . . . . . . . . . . . . 8 1.1.1 Volume . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.1.2 Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.1.3 Variety . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.2 Big data process in healthcare . . . . . . . . . . . . . . . . . . 10 1.2.1 Big data generation . . . . . . . . . . . . . . . . . . . 11 1.2.2 Big data storage . . . . . . . . . . . . . . . . . . . . 13 1.2.3 Big data analysis: . . . . . . . . . . . . . . . . . . . . 17 2 Machine learning and Deep learning . . . . . . . . . . . . . . . . . . . . 19 2.1 Machine learning categories . . . . . . . . . . . . . . . . . . . . 19 2.1.1 Supervised learning . . . . . . . . . . . . . . . . . . . . 19 2.1.2 Unsupervised learning . . . . . . . . . . . . . . . . . . 20 viiCONTENTS viii 2.1.3 Reinforcement learning . . . . . . . . . . . . . . . . . . 21 2.2 Deep learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.3 Transfer learning . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.4 Machine learning process . . . . . . . . . . . . . . . . . . . . . 25 2.4.1 Data preprocessing . . . . . . . . . . . . . . . . . . . 25 2.4.2 Feature selection . . . . . . . . . . . . . . . . . . . . . 27 2.4.3 Choosing a model . . . . . . . . . . . . . . . . . . . . 28 2.4.4 Model evaluation . . . . . . . . . . . . . . . . . . . . 30 3 Medical PHM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.1 Engineering PHM VS medical PHM . . . . . . . . . . . . . . . . 33 3.2 M-PHM analytics . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.2.1 Computer Aided Detection (Descriptive analysis) . . . 35 3.2.2 Computer Aided Diagnosis (Diagnostic analysis) . . . 36 3.2.3 Computer Aided Prognostic (Predictive analysis) . . . 36 3.2.4 Computer Aided Decision making (prescriptive analysis) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 III Retargeting PHM tools: from industrial to medical field 39 1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2 Adapted PHM tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.1 Data description . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.2 Data preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.2.1 Missing data . . . . . . . . . . . . . . . . . . . . . . . 42 2.2.2 Scaling data . . . . . . . . . . . . . . . . . . . . . . . . 43 2.2.3 Feature selection . . . . . . . . . . . . . . . . . . . . . 43 2.3 Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3 Experimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 IV Computer Aided Diagnosis for Spitzoid lesions classification using Artificial Intelligence techniques 49 1 Medical overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 2 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 3 Proposed method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.1 Data description . . . . . . . . . . . . . . . . . . . . . . . . . . 53CONTENTS ix 3.2 Pre-Processing Phase . . . . . . . . . . . . . . . . . . . . . . . . 56 3.3 The Feature Selection Phase . . . . . . . . . . . . . . . . . . . . 57 3.4 The Classification Phase . . . . . . . . . . . . . . . . . . . . . . 59 4 Performance metrics and experimentation . . . . . . . . . . . . . . . . 62 4.1 Performance Metrics . . . . . . . . . . . . . . . . . . . . . . . . 62 4.2 Data Sampling Results . . . . . . . . . . . . . . . . . . . . . . . 63 4.3 Feature Selection Results . . . . . . . . . . . . . . . . . . . . . . 66 5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 V Toward efficient Automatic Hyperparameters selection using Big Data tools to improve Skin Lesions classification 73 1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 2 Related work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 2.1 Architecture selection . . . . . . . . . . . . . . . . . . . . . . . 75 2.2 Data preparation . . . . . . . . . . . . . . . . . . . . . . . . . . 76 2.3 Model’s hyperparameter optimization . . . . . . . . . . . . . . . 76 3 Proposed method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 3.1 Data description . . . . . . . . . . . . . . . . . . . . . . . . . . 77 3.2 Data preparation . . . . . . . . . . . . . . . . . . . . . . . . . . 77 3.3 Data augmentation . . . . . . . . . . . . . . . . . . . . . . . . . 81 3.4 Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 3.5 Model’s Hyper parameters optimization . . . . . . . . . . . . . . 85 3.5.1 Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 3.5.2 Implementation . . . . . . . . . . . . . . . . . . . . . . 86 4 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.1 Results with /without metadata . . . . . . . . . . . . . . . . . 90 4.2 Results with Automatic hyperparameters Selection (CNN-AHPS) 92 5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . |
| En ligne : | http://thesis.univ-biskra.dz/id/eprint/5463 |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| TINF/160 | Théses de doctorat | bibliothèque sciences exactes | Consultable |
