Phelma Formation 2022

5PMBMLD0 : Introduction à l'apprentissage automatique et à l'apprentissage profond - WPMBXAS2

  • Volumes horaires

    • CM 12.0
    • Projet 0
    • TD 0
    • Stage 0
    • TP 12.0

    Crédits ECTS

    Crédits ECTS 3.0

Objectif(s)

Provide a solid introduction to machine-learning fundamentals, then delve into state-of-the-art deep-learning techniques and their practical applications in medical imaging

Contact Dawood AL CHANTI

Contenu(s)

Introduction to basic and advanced deep-learning techniques for medical-image analysis

1. Machine-Learning Fundamentals: supervised classification & regression (linear and logistic)
2. Deep Feed-Forward Networks: fully connected (multilayer perceptron) architectures
3. Convolutional Neural Networks (CNNs): feature extraction for 2-D, 3-D and multi-modal images
4. Representation Learning with Autoencoders: latent-space modeling and anomaly detection
5. Explainable AI & Uncertainty Quantification: XAI methods tailored to clinical decision-making
6. Generative Models: GANs, VAEs and hybrid approaches for data augmentation and synthesis
7. Diffusion Models: state-of-the-art denoising and high-fidelity image generation
8. MedSAM: “Segment Anything” for Healthcare: prompt-based foundation models for universal medical-image segmentation



Prérequis

Python language MANDATORY
1st year mathematics course

Contrôle des connaissances

Semestre 9 - L'examen existe uniquement en anglais 

Continuous assessment
BE report + MCQ + Project



Rapport de BE : 50%
Examen Ecrit : 50%

Informations complémentaires

Semestre 9 - Le cours est donné uniquement en anglais EN

Cursus ingénieur->Double-Diplômes Ingénieur/Master->Semestre 9
Cursus ingénieur->Masters->Semestre 9

Bibliographie

Deep Learning, An MIT Press book, Ian Goodfellow and Yoshua Bengio and Aaron Courville
https://www.deeplearningbook.org/