Hercules: Deep Hierarchical Attentive Multilevel Fusion Model With Uncertainty Quantification for Medical Image Classification
Article
Abdar, Moloud, Fahami, Mohammad Amin, Rundo, Leonardo, Radeva, Petia, Frangi, Alejandro F., Acharya, U. Rajendra, Khosravi, Abbas, Lam, Hak-Keung, Jung, Alexander and Nahavandi, Saeid. 2023. "Hercules: Deep Hierarchical Attentive Multilevel Fusion Model With Uncertainty Quantification for Medical Image Classification." IEEE Transactions on Industrial Informatics. 19 (1), pp. 274-285. https://doi.org/10.1109/TII.2022.3168887
| Article Title | Hercules: Deep Hierarchical Attentive Multilevel Fusion Model With Uncertainty Quantification for Medical Image Classification |
|---|---|
| ERA Journal ID | 41168 |
| Article Category | Article |
| Authors | Abdar, Moloud, Fahami, Mohammad Amin, Rundo, Leonardo, Radeva, Petia, Frangi, Alejandro F., Acharya, U. Rajendra, Khosravi, Abbas, Lam, Hak-Keung, Jung, Alexander and Nahavandi, Saeid |
| Journal Title | IEEE Transactions on Industrial Informatics |
| Journal Citation | 19 (1), pp. 274-285 |
| Number of Pages | 12 |
| Year | 2023 |
| Publisher | IEEE (Institute of Electrical and Electronics Engineers) |
| Place of Publication | United States |
| ISSN | 1551-3203 |
| 1941-0050 | |
| Digital Object Identifier (DOI) | https://doi.org/10.1109/TII.2022.3168887 |
| Web Address (URL) | https://ieeexplore.ieee.org/document/9763036 |
| Abstract | The automatic and accurate analysis of medical images (e.g., segmentation,detection, classification) are prerequisites for modern disease diagnosis and prognosis. Computer-aided diagnosis (CAD) systems empower accurate and effective detection of various diseases and timely treatment decisions. The past decade witnessed a spur in deep learning (DL)-based CADs showing outstanding performance across many health care applications. Medical imaging is hindered by multiple sources of uncertainty ranging fromnteasurement (aleatoric) errors, physiological variability, and limited medical knowledge (epistemic errors). However, uncertainty quantification (UQ) in most existing DL methods is insufficiently investigated, particularly in medical image analysis. Therefore, to address this gap, in this article, we propose a simple yet novel hierarchical attentive multilevel feature fusion model with an uncertainty-aware module for medical image classification coined Hercules . This approach is tested on several real medical image classification challenges. The proposed Hercules model consists of two main feature fusion blocks, where the former concentrates on attention-based fusion with uncertainty quantification module and the latter uses the raw features. Hercules was evaluated across three medical imaging datasets, i.e., retinal OCT, lung CT, and chest X-ray. Hercules produced the best classification accuracy in retinal OCT (94.21%), lung CT (99.59%), and chest X-ray (96.50%) datasets, respectively, against other state-of-the-art medical image classification methods. |
| Keywords | Attention mechanisms; deep learning (DL); uncertainty quantification; medical image classification; feature fusion; early fusion |
| ANZSRC Field of Research 2020 | 400306. Computational physiology |
| Public Notes | File reproduced in accordance with the copyright policy of the publisher/author. |
| Byline Affiliations | Deakin University |
| Isfahan University of Technology, Iran | |
| University of Cambridge, United Kingdom | |
| University of Barcelona, Spain | |
| University of Leeds, United Kingdom | |
| Catholic University of Leuven, Belgium | |
| Ngee Ann Polytechnic, Singapore | |
| Singapore University of Social Sciences (SUSS), Singapore | |
| Asia University, Taiwan | |
| King's College London, United Kingdom | |
| Aalto University, Finland |
Permalink -
https://research.usq.edu.au/item/z1v47/hercules-deep-hierarchical-attentive-multilevel-fusion-model-with-uncertainty-quantification-for-medical-image-classification
Download files
Accepted Version
| Hercules_UncertaintyQuantification_MedicalImageClassification_IEEE_TII2022.pdf | ||
| File access level: Anyone | ||
68
total views155
total downloads5
views this month5
downloads this month