Enhancing Myocardial Disease Prediction with DOC-NET+ Architecture: A Custom Data Analysis Approach for the EMIDEC Challenge - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Enhancing Myocardial Disease Prediction with DOC-NET+ Architecture: A Custom Data Analysis Approach for the EMIDEC Challenge

Résumé

Cardiovascular diseases, including myocarditis and myocardial infarction (MI), pose substantial health risks due to their lifethreatening nature. This article delves into the complexities of distinguishing these conditions, as they share symptoms and diagnostic challenges. It underscores the vital role of advanced imaging techniques, such as cardiac magnetic resonance imaging (CMR), and emphasizes the emerging importance of artificial intelligence (AI), particularly the innovative DOC-Net+ architecture, in refining precision. Through a comprehensive exploration of the EMIDEC challenge, which combines clinical data with MRI insights, this study highlights AI's potential to enhance diagnosis accuracy. Notably, our implementation of both the DOC-NET and DOC-NET+ architectures yielded promising classification results, achieving 97% and 98% accuracy, respectively, on our newly established dataset.
Fichier principal
Vignette du fichier
Mariem_ICMLDE_23.pdf (804.17 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04451330 , version 1 (11-02-2024)

Identifiants

  • HAL Id : hal-04451330 , version 1

Citer

Mariem Dali, Rostom Kachouri, Narjes Benameur, Youness Laaroussi, Salam Laabidi. Enhancing Myocardial Disease Prediction with DOC-NET+ Architecture: A Custom Data Analysis Approach for the EMIDEC Challenge. The 2nd International conference on Machine Learning and Data Engineering (ICMLDE 2023), Nov 2023, Dehradun, India. ⟨hal-04451330⟩
17 Consultations
9 Téléchargements

Partager

Gmail Facebook X LinkedIn More