COVID-19 chest X-ray detection through blending ensemble of CNN snapshots

Abstract

The novel COVID-19 pandemic, has effectively turned out to be one of the deadliest events in modern history, with unprecedented loss of human life, major economic and financial setbacks and has set the entire world back quite a few decades. However, detection of the COVID-19 virus has become increasingly difficult due to the mutating nature of the virus, and the rise in asymptomatic cases. To counteract this and contribute to the research efforts for a more accurate screening of COVID-19, we have planned this work. Here, we have proposed an ensemble methodology for deep learning models to solve the task of COVID-19 detection from chest X-rays (CXRs) to assist Computer-Aided Detection (CADe) for medical practitioners. We leverage the strategy of transfer learning for Convolutional Neural Networks (CNNs), widely adopted in recent literature, and further propose an efficient ensemble network for their combination. The DenseNet-201 architecture has been trained only once to generate multiple snapshots, offering diverse information about the extracted features from CXRs. We follow the strategy of decision-level fusion to combine the decision scores using the blending algorithm through a Random Forest (RF) meta-learner. Experimental results confirm the efficacy of the proposed ensemble method, as shown through impressive results upon two open access COVID-19 CXR datasets - the largest COVID-X dataset, as well as a smaller scale dataset. On the large COVID-X dataset, the proposed model has achieved an accuracy score of 94.55% and on the smaller dataset by Chowdhury et al., the proposed model has achieved a 98.13% accuracy score.

Keywords: Blending; COVID-19; Chest X-ray; Classifier fusion; Deep learning; Ensemble.

Fig. 1

Weekly distribution of covid positive cases and death counts (worldwide) as of 25th March,2022 .

Fig. 2

Sample images of CXRs for all three classes taken from the COVID-X dataset.

Fig. 3

Schematic diagram of our proposed methodology which consists of: (I) Acquisition and Preprocessing of input CXRs, (II) Transfer Learning upon the DenseNet-201 CNN architecture, (III) Generation of multiple Snapshots with Cosine Annealing (Section 3.4.2) with only one training phase, and (IV) Ensemble of classifiers using blending algorithm with RF meta-learner to yield prediction, available for medical practitioners.

Fig. 4

Sample chest X-ray scan images taken from COVID-X dataset  showing: (a) COVID-19 positive and (b) Pneumonia cases.

Fig. 5

Shows a cyclic learning rate while following the cosine function providing a warm restart after every 10 epochs.

Fig. 6

Performance of base CNN classifiers.

Fig. 7

Confusion Matrices.