Developing lung cancer post-diagnosis system using pervasive data analytic framework. (January 2023)

• Record Type: Journal Article
• Authors: Pethuraj, Mohamed Shakeel; bin Mohd Aboobaider, Burhanuddin; Salahuddin, Lizawati Binti
• Is Part Of: Computers & electrical engineering. Volume 105(2023)
• Publication Date: 2023-01
• Subjects: Auto encoder learning -- Butterfly optimization -- Jaya optimization -- Lung cancer -- Pervasive data analysis -- Wearable sensors -- Feature correlation -- Data segregation; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108528
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Automatic lung cancer detection process by resolving the clinical correlation flaws Auto-encoder variations to improve cancer detection with minimal error Managing data & discreteness while analyzing cancer data by optimization techniques Abstract: The data from lung cancer patients using wearable sensors and clinical assessments after observation is available to predict the disease's recurrence. In recurrence prediction, pervasive data analysis is required to prevent flaws in clinical correlations and data observations. This article proposes a Pervasive Data Analytical Framework (PDAF) for recurrence prediction. The proposed framework incorporates three processes: data segregation using Butterfly Optimisation, feature correlation using Jaya Optimisation, and autoencoder prediction. First, the data from the wearable sensor is segregated using observation count for its availability and discreteness. It prevents missing errors under different observation sequences for which the correlation rate is determined using the next optimization. In the Jaya optimization process, the features correlate with the clinical assessments to improve precision. The autoencoder predicts the occurrence of previous missing and non-correlated inputs for maximizing the detection rate. Using the proposed framework, the maximum gains of 9.22% in accuracy, 9.29% in detection, and 7.96% in recommendations. Graphical abstract: Image, graphical abstract