Piecewise adaptive weighted smoothing-based multivariate rosenthal correlative target projection for lung and uterus cancer prediction with big data

Abstract

Cancer is the uncontrolled growth and spread of abnormal cells in the body. Early detection and prediction of cancer are crucial aspects of modern healthcare aimed at greatly improving the chances of survival for patients by reducing mortality rates and the number of people affected by this disease. Due to the large volume of data generated in the medical industry, accurate cancer detection is a challenging task. Many cancer classification systems using machine learning and deep learning models have been developed but accurate cancer detection with minimal time consumption remains a major challenging issue in the big data applications. To enhance the accuracy of cancer detection with minimal time, the Piecewise Adaptive Weighted Smoothing-based Multivariate Rosenthal Correlative Target Projection (PAWS-MRCTP) technique is introduced. This technique aims to detect lung and uterus cancers while leveraging big data. The proposed PAWS-MRCTP technique comprises three main processes namely data acquisition, preprocessing, and feature selection. In the data acquisition phase, a large number of cancer patient data are collected from lung cancer and uterus cancer detection datasets. Subsequently, the collected patient data undergo preprocessing. The preprocessing stage comprises three key processes namely handling missing data, noisy data, and outlier data. Firstly, the proposed PAWS-MRCTP is employed to address missing values, utilizing the Piecewise Adaptive Constant Interpolation method based on multiple available data points. Noisy data are identified using Gower's weighted smoothing technique, which detects data containing random variations or errors. Subsequently, outlier data are identified and removed by applying Peirce's statistical test. As a result, the pre-processed dataset is obtained resulting to minimize the time complexity. With the pre-processed dataset, the feature selection process is carried out to minimize the dimensionality of the large dataset. The proposed PAWS-MRCTP technique utilizes the Multivariate Rosenthal correlative target feature projection technique to identify the most relevant features. By selecting significant features, this approach enhances the accuracy of lung cancer and uterus cancer detection with minimal time consumption. Experimental assessment is conducted with different evaluation metrics such as cancer detection accuracy, precision, and cancer detection time and space complexity. The observed result shows the effectiveness of the proposed PAWS-MRCTP technique with higher accuracy with minimum time than the existing methods.