Do Preprocessing and Class Imbalance Matter to the Deep Image Classifiers for COVID-19 Detection? An Explainable Analysis
IEEE Transactions on Artificial Intelligence
; 4(2):229-241, 2023.
Artículo
en Inglés
| Scopus | ID: covidwho-2292006
ABSTRACT
In a world withstanding the waves of a raging pandemic, respiratory disease detection from chest radiological images using machine-learning approaches has never been more important for a widely accessible and prompt initial diagnosis. A standard machine-learning disease detection workflow that takes an image as input and provides a diagnosis in return usually consists of four key components, namely input preprocessor, data irregularities (like class imbalance, missing and absent features, etc.) handler, classifier, and a decision explainer for better clarity. In this study, we investigate the impact of the three primary components of the disease-detection workflow leaving only the deep image classifier. We specifically aim to validate if the deep classifiers may significantly benefit from additional preprocessing and efficient handling of data irregularities in a disease-diagnosis workflow. To elaborate, we explore the applicability of seven traditional and deep preprocessing techniques along with four class imbalance handling approaches for a deep classifier, such as ResNet-50, in the task of respiratory disease detection from chest radiological images. While deep classifiers are more capable than their traditional counterparts, explaining their decision process is a significant challenge. Therefore, we also employ three gradient visualization algorithms to explain the decision of a deep classifier to understand how well each of them can highlight the key visual features of the different respiratory diseases. © 2020 IEEE.
Class imbalanced classification; deep image classifier; gradient visualization; image preprocessing; medical image analysis; respiratory disease detection; Computer aided diagnosis; Computerized tomography; Data visualization; Image analysis; Image classification; Machine components; Machine learning; Medical imaging; Pulmonary diseases; Visualization; Computed tomography; Data preprocessing; Disease detection; Image Classifiers; Imbalanced classification; Lung; Medical diagnostic imaging; COVID-19
Texto completo:
Disponible
Colección:
Bases de datos de organismos internacionales
Base de datos:
Scopus
Idioma:
Inglés
Revista:
IEEE Transactions on Artificial Intelligence
Año:
2023
Tipo del documento:
Artículo
Similares
MEDLINE
...
LILACS
LIS