Assessment of deep learning and transfer learning for cancer prediction based on gene expression data

• Published in: BMC bioinformatics 2022-07, Vol.23 (1), p.1-262, Article 262
• Main Authors: Hanczar, Blaise, Bourgeais, Victoria, Zehraoui, Farida
• Format: Article
• Language: English
• Subjects: Analysis; Artificial Intelligence; Artificial neural networks; Bioinformatics; Cancer; Classification; Computer Science; Datasets; Deep learning; Deep neural network; Diagnosis; Experiments; Gene expression; Genetic aspects; Learning algorithms; Machine Learning; Medical prognosis; Methods; Neural and Evolutionary Computing; Neural networks; Phenotype prediction; Predictions; Statistics; Support vector machines; Training; Transfer learning
• ISSN: 1471-2105; 1471-2105
• DOI: 10.1186/s12859-022-04807-7

Machine learning is now a standard tool for cancer prediction based on gene expression data. However, deep learning is still new for this task, and there is no clear consensus about its performance and utility. Few experimental works have evaluated deep neural networks and compared them with state-of-the-art machine learning. Moreover, their conclusions are not consistent. We extensively evaluate the deep learning approach on 22 cancer prediction tasks based on gene expression data. We measure the impact of the main hyper-parameters and compare the performances of neural networks with the state-of-the-art. We also investigate the effectiveness of several transfer learning schemes in different experimental setups. Based on our experimentations, we provide several recommendations to optimize the construction and training of a neural network model. We show that neural networks outperform the state-of-the-art methods only for very large training set size. For a small training set, we show that transfer learning is possible and may strongly improve the model performance in some cases.