A review of convolutional neural network-based computer-aided lung nodule detection system

Abstract

Worldwide, lung cancer is the leading cause of mortality and rapidly spreads. Lung tissue that is benign does not grow significantly, but lung tissue that is malignant grows rapidly and attacks the body, posing a grave threat to one's health. This paper provides a literature review of computer-aided detection (CAD) systems for lung cancer diagnosis. Preprocessing, segmentation, detection, and classification are the stages of the CAD system. This review divides the preprocessing into three stages: image smoothing, edge sharpening, and noise removal. Additionally, lung segmentation is divided into three stages: histogram-based thresholding, linked component analysis, and lung extraction. The detecting phase aids in decreasing the workload. Several techniques are briefly described, including random forest, support vector machine (SVM), naive bayes, K-nearest neighbor (KNN), and convolutional neural network (CNN). Classification is the final stage; the image is then identified as containing or not possessing nodules. The prospect of incorporating CNN-based deep learning techniques into the CAD system is discussed. This paper is superior to other review studies on this topic due to its comprehensive examination of pertinent literature and structured presentation. We anticipate that our study may aid professional researchers and radiologists in developing more effective CAD systems for detecting lung cancer.

Predicting lung cancer risk using explainable artificial intelligence

“…In order to do this, it may be necessary to evaluate the models' capacity to offer precise and understandable justifications for how they arrived at a given risk score, as well as the consistency of these justifications for various cases. − Identify areas for improvement: the next stage would be to pinpoint areas that require development or improvement based on the analysis of the current models [35], [36]. This could entail locating additional risk variables that could be incorporated into the models, enhancing the readability of current models, or creating brand-new machine learning techniques that are especially intended to be more understandable.…”

Seizure stage detection of epileptic seizure using convolutional neural networks

“…These classifiers were applied to two well-known datasets: Fitchburg Hospital (FH) and Children's Hospital Boston-Massachusetts institute of technology (CHB-MIT) [29]. In [30], [31] proposed an automated method using RF and iterative filtering to identify the EEG signals. This classification accuracy for D versus E was 96%, E versus ABCD was 98.4%, and 99.5% for the a against E subsets using the Bonn University dataset (A-E).…”

Automated 3D convolutional neural network architecture design using genetic algorithm for pulmonary nodule classification

“…It may also cause metastasis, the spread of cancer from the lung into other organs which increases the risk of death. That's why early detection and treatment significantly boost the success of treatment and decrease mortality rates in diagnosed patients [5], [6]. Computed tomography (CT) screening has proven effective in early detection of lung nodules, offering a potential solution to mitigate this situation and decreases lung cancer mortality rates [7].…”

“…Recently, computer-aided diagnosis (CAD) systems have emerged as a valuable tool for easing the burden on radiologists by providing objective prediction with non-invasive solution to aid radiologists to diagnose pulmonary nodules [3], [7], [9]. Typically, CAD systems for lung nodule detection involve five stages: i) image acquisition, ii) preprocessing, iii) lung segmentation, iv) nodule detection, and v) classification [6]. This study specifically focuses on the classification method for lung nodule.…”