An approach for cancer classification using optimization driven deep learning

Abstract

The normal and cancer cell tissues exhibit different gene expressions. Therefore, gene expression data are the effective source for cancer classification, in which the usage of the original gene expression data is challenging due to their high dimension and small size of the data samples. This article proposes a fractional biogeography-based optimization-based deep convolutional neural network (FBBO-based deep CNN) for cancer classification. The developed FBBO is the integration of the fractional calculus (FC) in the biogeography-based optimization (BBO), which aims at determining the optimal weights for tuning the deep CNN. Initially, the gene expression data is pre-processed and subjected to dimensional reduction using the probabilistic principal component analysis (PPCA). The selected features are used for cancer classification enabling a high degree of robustness and accuracy. The experimental analysis using the Colon dataset and Leukemia dataset reveals that the proposed classifier acquired maximal accuracy, sensitivity, specificity, precision, and F-Measure of 0.98.

Application of artificial intelligence in cancer diagnosis and tumor nanomedicine

“…They integrated a Softmax classifier into this feature extraction method for gastric cancer classification, achieving a remarkable overall accuracy of 98.7% for late-stage gastric cancer and 97.3% for early gastric cancer detection using breath analysis. Devendran et al 87 contributed to the field by introducing a novel approach in cancer classification, leveraging a fractional biogeography-based Fig. 6 Deep CNN layout and classification of skin lesions: the internal representation of four significant disease classes by the convolutional neural network (CNN) was visualized using t-SNE, a method designed for visualizing high-dimensional data.…”

RN-Autoencoder: Reduced Noise Autoencoder for classifying imbalanced cancer genomic data

“…Devendran et al. [ 38 ] reduced the high dimensionality of the genomic data by utilizing Probabilistic Principal Component Analysis (PPCA) to enhance the performance of their cancer classification proposal. Following that, the resulting features were classified using a CNN, with the weights optimized using Fractional Biogeography-Based Optimization (FBBO).…”

“…[ 30 ] 2021 2 Wrapper Fractional-ASO Deep RNN Al Mamun et al [ 31 ] 2021 12 mrCAE - Multiple Stages Feature Selection Majumder et al [ 33 ] 2022 4 ANOVA, IG MLP, 1DCNN, 2DCNN Saberi-Movahed et al [ 34 ] 2022 9 DR-FS-MFMR = Matrix Factorization + Minimum Redundancy Unsupervised clustering Bustamam et al [ 35 ] 2021 2 SVM-RFE + ABC SVM Samieinasab et al [ 47 ] 2022 1 Ensemble (Variance Inflation Factor, Pearson’s Correlation, Information Gain) Ensemble (Boosting, Bagging, Voting) Single Stage Feature Extraction Devendran et al. [ 38 ] 2021 2 PPCA FBBO + CNN Majji et al [ 39 ] 2021 4 Non-negative matrix factorization JayaALO-based Deep RNN Singh et al [ 48 ] 2022 2 PCA C5.0, AdaBoost, CART, GBM, NB, RF, SVM, AdaBoost Bacha et al [ 50 ] 2022 …”