Pneumonia detection from lung X‐ray images using local search aided sine cosine algorithm based deep feature selection method

Abstract

Pneumonia is a major cause of death among children below the age of 5 years, globally. It is especially prevalent in developing and underdeveloped nations where the risk factors for the disease such as unhygienic living conditions, high levels of pollution and overcrowding are higher. Radiological examination (usually X-ray scans) is conducted to detect pneumonia, yet it is prone to subjective variability and can lead to disagreements among different radiologists. To detect traces of pneumonia from X-ray images, a more robust method is therefore required, which can be achieved by using a computer-aided diagnosis (CAD) system. In this study, we develop a two-stage framework, using the combination of deep learning and optimization algorithms, which is both accurate and timeefficient. In its first stage, the proposed framework extracts feature using a customized deep learning model called DenseNet-201 following the concept of transfer learning to cope with the scanty available data. In the second stage, we then reduce the feature dimension using an improved sine cosine algorithm equipped with adaptive beta hill climbing-based local search algorithm. The optimized feature subset is utilized for the classification of "Pneumonia" and "Normal" X-ray images using a support vector machines classifier. Upon an evaluation on a publicly available data set, the proposed method demonstrates the

An adaptive and altruistic PSO-based deep feature selection method for Pneumonia detection from Chest X-rays

“…Thus, it is crucial to select the right set of hyperparameters to effectively use these algorithms. In this work, we follow some previous methods [8] , [22] in order to use the standard values of these parameters. The parameters used can be found in Table 3 .…”

“… Work Ref. Accuracy Precision Recall F1 Linag & Zheng [5] 90.50 89.10 96.70 92.70 Sharma et al [31] 90.68 – – – Stephen et al [32] 93.73 – – – Ibrahim et al [33] 94.43 – 98.19 – Saraiva et al [34] 95.30 98.86 94.77 96.77 Rajaraman et al [35] 96.20 97.70 96.20 97.00 Dey et al [11] 97.94 95.02 97.55 96.27 Mahmud et al [36] 98.10 98.00 98.50 98.30 Chattopadhyay et al [8] 98.36 98.98 98.79 98.88 Ours 98.41 98.80 99.02 98.91 …”

“…For this purpose, we use a transfer function, which normalizes the optimized values to the range of . We employ a standard transfer function in the domain of feature selection [8] , [22] . We generally refer to this function as the shaped transfer function.…”

“…The work by Chattopadhyay et al. [8] proposed a deep feature-selection technique with a Sine-Cosine Algorithm aided by a local search method. Unsupervised learning approaches have also been proposed in the past.…”

An Adaptive and Altruistic PSO-based Deep Feature Selection Method for Pneumonia Detection from Chest X-Rays

“…A fine-tuned network might have given a better feature representation, which would have been more useful for final classification. The work by Chattopadhyay et al [8] proposed a deep feature-selection technique with a Sine-Cosine Algorithm aided by a local search method. Unsupervised learning approaches have also been proposed in the past.…”