TABLE 2.
Research analysis.
| Study | Cancer dataset | Objective | Technique | Acc |
|---|---|---|---|---|
| Dwivedi, (2016) | Leukemia | Cancer classification employing microarray gene-expression data using deep learning | ANN | 98% |
| Yuan et al. (2016) | 12 selected types of cancer | Cancer type classification using deep learning and somatic point mutations | DeepGene | 94% |
| Motieghader et al., 2017 | 6 different cancer | Cancer classification using microarray data using genetic algorithm | Genetic Algorithm | 94% |
| Aziz et al., 2017 | 5 gene microarray datasets | Microarray data classification using novel hybrid method | Artificial Bee Colony (ABC) | 95% |
| Tumuluru and Ravi, (2017) | Colon and Leukemia data | Implementing deep neural networks for cancer classification | GOA-based DBN | 95% |
| Extraction, (2017) | Breast cancer | Integrated deep neural networks to predict breast cancer | Deep-SVM | 70% |
| Danaee et al. (2017) | Breast cancer | Relevant gene identification for better cancer classification | Stacked Denoising Autoencoder (SDAE) | 98% |
| Urda and Moreno, (2017) | 3 cancer databases | Investigating RNA-sequence gene expression data utilizing deep learning | Regularized linear model (standard LASSO) and two deep learning models | 75% |
| Salman, (2018) | TCGA | Analyzing the Effect of meta heuristic iteration on the neural networks in cancer data | GA and FWA | 98% |
| Ching et al. (2018) | TCGA RNA-Sequence data | Evaluating deep learning technique for tumor detection | Cox-nnet. | -- |
| Cho et al. (2018) | TCGA LUAD | Examining the relationship between specific gene mutations and lung cancer survival | Information gain, chi-squared test | -- |
| Xiao et al. (2018b) | RNA-sequence data sets of three cancers | Analyzing deep learning technique to predict cancer employing RNA sequence data | Sparse Auto-Encoder (SSAE) | 98% |
| Lin et al. (2018) | TCGA Leukemia | Introduced deep learning to Predicting Prognosis of Leukemia | Stacked Autoencoders | 83% |
| Alomari et al. (2018) | 10 microarray datasets | Implementing a novel strategy for gene selection based on a hybrid technique | Hybrid Bat-inspired Algorithm | 100% |
| Parvathavardhini and Manju, (2020) | Gene expression data of liver cancer | Cancer gene recognition using neuro-fuzzy approach | Neuro-Fuzzy method | 96% |
| Ahn and Lee, (2018) | TCGA | Recognition of cancer tissues using RNA-Sequence data | Deep neural network (DNN) | 99.7% |
| Kong and Yu, (2018) | Two RNA-seq expression datasets | Extracting features for RNA-Sequence data classification | Forest Deep Neural Network (fDNN) | 90.4% |
| Guo et al., (2018) | Multiple Cancer datasets | Cancer subtype classification using RNA-Sequence gene expression data | BCD Forest | 92.8% |
| Chen et al. , (2018) | mRNA datasets from the GDC repository | Cancer type recognition using neural network | Deep Learning models | 98% |
| Sevakula et al. (2018) | 36 datasets from the GEMLeR repository | Implemented transfer learning for molecular cancer classification | Sparse Autoencoders on gene expression data | 98% |
| Joshi and Park, (2019) | LUAD | Lung Cancer Subtype Classification using deep learning model | Sparse Cross-modal Superlayered Neural Network | 99% |
| Gao et al., (2019) | gene expression data | Cancer subtype prediction using gene expression data | Deep cancer subtype classification (DeepCC) | 90% |
| Basavegowda and Dagnew, (2020) | 8 microarray cancer datasets | Deep neural networks for classifying microarray cancer data | 7-layer deep neural network architecture | 90% |
| Huynh et al., (2019) | TCGA | Developed hybrid approach for classifying RNA-sequence data | Deep convolutional neural network (DCNN) | 95% |
| Xu et al., (2019a) | RNA-seq gene expression data | Cancer subtype classification | Deep flexible neural forest (DFNForest) | 76% |
| Xiao et al., (2018a) | LUAD, BRCA, and STAD | Cancer type prediction using deep learning model | ensemble-based approach | 97% |
| Guia, (2019) | RNA-sequence data from Pan-Cancer Atlas | Cancer type Classification using RNA-sequence data | DeepGx Convolutional neural network (CNN) | 95.65% |
| Huang et al. (2020) | TCGA cancers | Cancer survival prediction from RNA-sequence data | AECOX (AutoEncoder with Cox regression network) | -- |
| Shon et al. (2021) | TCGA stomach cancer dataset | Stomach cancer prediction using gene expression data | CNN | 96% |
| García-díaz et al., (2019) | 5 types of cancer | Multiclass cancer classification of gene expression RNA-Sequence data | Extreme Learning Machine algorithm | 98.81% |
| Kim et al., (2020) | TCGA | Cancer prediction using gene expression data | NN, SVM, KNN, RF | 94% |
| Jerez et al., (2020) | 31 Tumor types | Prediction of cancer survival using gene-expression data | Transfer learning with CNN | 73% |
| Panda, (2017) | 10 most common UCI Cancer datasets | Analyzed microarray cancer data using deep neural networks | Elephant search optimization based deep learning approach | 92% |
| He and Luo, (2020) | 15 different cancer types | Prediction of the tissue-of-origin of cancer types on basis of RNA-sequence data | a novel NN model | 80% |
| Abdollahi et al., (2021) | Diabetes, heart, cancer dataset | Disease prediction model for the healthcare system | neural network-based ensemble learning | 100% |
| Torkey et al., (2021) | RNA-seq data of three datasets | Cancer survival analysis for microarray dataset. | AutoCox and AutoRandom | 98% |
| Wessels et al., (2021) | prostate cancer patients | Prediction of lymph node metastasis straight from tumor histology in prostate malignancy | convolutional neural network | 62% |
| Chaunzwa et al., (2021) | 311 NSCLC patients at Massachusetts General Hospital | Tumor detection using CT images | convolutional neural network | 71% |
| Gupta, (2021) | cervical cancer dataset | Prediction of Cervical Cancer risk factors | Ensemble model | 99.7% |
| Gupta and Gupta, (2021a) | five benchmark datasets | Cancer diagnosis analysis along with imbalanced classes | Stacked Ensemble Model | 98% |