Abstract
Arguably the human body has been one of the most sophisticated systems we encounter but until now we are still far from understanding its complexity. We have been trying to replicate human intelligence by way of artificial intelligence but with limited success. We have discovered the molecular structure in terms of genetics, performed gene editing to change an organism’s DNA and much more, but their translatability into the field of oncology has remained limited. Conventional machine learning methods achieved some degree of success in solving problems that we do not have an explicit algorithm. However, they are basically shallow learning methods, not rich enough to discover and extract intricate features that represent patterns in the real environment. Deep learning has exceeded human performance in pattern recognition as well as strategic games and are powerful for dealing with many complex problems. High-throughput sequencing and microarray techniques have generated vast amounts of data and allowed the comprehensive study of gene expression in tumor cells. The application of deep learning with molecular data enables applications in oncology with information not available from clinical diagnosis. This paper provides fundamental concepts of deep learning, an essential knowledge of cancer genetics, and a review of applications of deep learning to cancer oncology. Importantly, it provides an insightful knowledge of deep learning and an extensive discussion on its challenges. The ultimate purpose is to germinate ideas and facilitate collaborations between cancer biologists and deep learning researchers to address challenging oncological problems using advanced deep learning technologies.
Publisher
Publishing House for Science and Technology, Vietnam Academy of Science and Technology (Publications)
Reference75 articles.
1. Adjiri A. - Mutations May Not Be the Cause of Cancer, Oncology and Therapy 5 (1) (2017) 85-101.
2. Motofei I. G. - Biology of cancer; from cellular and molecular mechanisms to developmental processes and adaptation, Seminars in Cancer Biology, 2021, doi: https://doi.org/10.1016/j.semcancer.2021.10.003.
3. DeepMind - "AlphaGo." https://www.deepmind.com/research/highlightedresearch/ alphago (accessed June 13, 2022).
4. Krizhevsky A., Sutskever I., and Hinton G. - ImageNet Classification with Deep Convolutional Neural Networks, Advances in Neural Information Processing Systems 25 (2012) 1097-1105.
5. Russakovsky O., et al. - ImageNet Large Scale Visual Recognition Challenge, Int. J. Comput Vis 115 (2015) 211-252. doi: https://doi.org/10.1007/s11263-015-0816-y.