Deep learning in cancer diagnosis, prognosis and treatment selection
Глубокое обучение в диагностике рака, прогнозировании исходов и выборе лечения
2021-09-27
SCID: 54.1/m4xsktch
Discuss with AI
cancer diagnosisdeep learningexplainable deep learningmulti-omics data integrationprecision oncology
Figures from the paper
Abstract (AI)
Deep learning is a subdiscipline of artificial intelligence that uses a machine learning technique called artificial neural networks to extract patterns and make predictions from large data sets. The increasing adoption of deep learning across healthcare domains together with the availability of highly characterised cancer datasets has accelerated research into the utility of deep learning in the analysis of the complex biology of cancer. While early results are promising, this is a rapidly evolving field with new knowledge emerging in both cancer biology and deep learning. In this review, we provide an overview of emerging deep learning techniques and how they are being applied to oncology. We focus on the deep learning applications for omics data types, including genomic, methylation and transcriptomic data, as well as histopathology-based genomic inference, and provide perspectives on how the different data types can be integrated to develop decision support tools. We provide specific examples of how deep learning may be applied in cancer diagnosis, prognosis and treatment management. We also assess the current limitations and challenges for the application of deep learning in precision oncology, including the lack of phenotypically rich data and the need for more explainable deep learning models. Finally, we conclude with a discussion of how current obstacles can be overcome to enable future clinical utilisation of deep learning.
Key Findings
1
Current precision-oncology applications are limited by insufficiently phenotypically rich data and inadequate model explainability.
2
Deep learning is increasingly applied to complex cancer biology using large, highly characterized datasets for predictive oncology analyses.
3
Integrating multiple data types could support decision tools for cancer diagnosis, prognosis, and treatment management.
4
Overcoming data and interpretability barriers is necessary for future clinical adoption of deep learning in oncology.
5
The review covers deep learning applications to genomic, methylation, transcriptomic, and histopathology-based genomic inference data.
Research Object
Deep learning applications in oncology (for analysis of cancer-related omics and histopathology data)
Research Subject
deep-learning applications for cancer diagnosis, prognosis, treatment selection, and precision-oncology decision support, including data integration and model limitations
Publication Details
Publication Date
2021-09-27
Journal
Publisher
ISSN
Cited by
1090
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai11
A Unified Approach to Interpreting Model Predictions2017
edgeR : a Bioconductor package for differential expression analysis of digital gene expression data2009
WGCNA: an R package for weighted correlation network analysis2008
"Why Should I Trust You?": Explaining the Predictions of Any Classifier2016
Generative adversarial networks2020
The Cancer Genome Atlas Pan-Cancer analysis project2013
Mutational landscape and significance across 12 major cancer types2013
Key challenges for delivering clinical impact with artificial intelligence2019
International network of cancer genome projects2010
Explainability for artificial intelligence in healthcare: a multidisciplinary perspective2020
Pan-cancer image-based detection of clinically actionable genetic alterations2020
Cited by3
Application of Machine Learning for Differentiating Bone Malignancy on Imaging: A Systematic Review2023
Annotation-Free Deep Learning-Based Prediction of Thyroid Molecular Cancer Biomarker BRAF (V600E) from Cytological Slides2023
Artificial Intelligence in Medical Diagnostics: Foundations, Clinical Applications, and Future Directions2026