Artificial intelligence (AI) and machine learning (ML) in precision oncology: a review on enhancing discoverability through multiomics integration

Abstract

Multiomics data including imaging radiomics and various types of molecular biomarkers have been increasingly investigated for better diagnosis and therapy in the era of precision oncology. Artificial intelligence (AI) including machine learning (ML) and deep learning (DL) techniques combined with the exponential growth of multiomics data may have great potential to revolutionize cancer subtyping, risk stratification, prognostication, prediction and clinical decision-making. In this article, we first present different categories of multiomics data and their roles in diagnosis and therapy. Second, AI-based data fusion methods and modeling methods as well as different validation schemes are illustrated. Third, the applications and examples of multiomics research in oncology are demonstrated. Finally, the challenges regarding the heterogeneity data set, availability of omics data, and validation of the research are discussed. The transition of multiomics research to real clinics still requires consistent efforts in standardizing omics data collection and analysis, building computational infrastructure for data sharing and storing, developing advanced methods to improve data fusion and interpretability, and ultimately, conducting large-scale prospective clinical trials to fill the gap between study findings and clinical benefits.

Machine Learning and Bias in Medical Imaging: Opportunities and Challenges

“…Indeed, multiple studies have shown that unsupervised learning approaches applied to a combination of clinical, genetic, and imaging data can identify novel distinct—and likely more inclusive—phenotypes of disease that can serve to improve predictions of both disease outcomes and their responses to interventions. 58…”

“…While extending the benefits of radiomics, which generates imaging biomarkers of disease, AI methods also offer benefits to multisource data including genetic, epigenetic, transcriptomic, proteomic, metabolomic, and other types of molecular biomarker data. As recommended by experts in cardiovascular imaging, as well as other fields, 57,58 the contributions of AI methods applied to multiomics data are at least 2-fold. First, AI methods add new computational capabilities for harnessing vast amounts of high-dimensional data and analyzing their complex inter relations.…”

“…Indeed, multiple studies have shown that unsupervised learning approaches applied to a combination of clinical, genetic, and imaging data can identify novel distinct-and likely more inclusive-phenotypes of disease that can serve to improve predictions of both disease outcomes and their responses to interventions. 58 As with any robust model development process, a comprehensive workflow will involve iterative training and retraining to optimize performance as well as mitigate bias. 91 Upon completion of a final model, initial and then routine reporting of model performance across data sets can be facilitated using tools such as model cards for algorithms and data sheets for data sets-intended to provide optimal transparency and updated guidance on how to use, validate, and interpret results from a given model applied to a given data set.…”

Intricacies of Human-AI Interaction in Dynamic Decision-Making for Precision Oncology: A Case Study in Response-Adaptive Radiotherapy

“…Development of novel cancer therapies 1,2 and increasing availability of longitudinal multi-omics data 3,4 have improved our ability to prescribe personalized, adaptive treatment strategies [5][6][7] capable of dynamically reacting to individual cancer progression while improving efficacy and minimizing side effects. However, the dynamic nature of adaptive strategies compounded by a wide range of clinical options, high data dimensionality, uncertainty in assessing treatment response, and the uncertainty in the future course of disease, present challenges in tailoring optimal strategies.…”

Editorial: Reshaping the diagnostic process in oncology: science versus technology

“…The study was conducted with the aim of elucidating the experiences of cancer survivors with healthcare systems in five European nations. In the current milieu of precision oncology, there has been a growing emphasis on multi-omics data, encompassing imaging radiomics and diverse molecular biomarkers, to refine diagnostic and therapeutic approaches ( 14 ). The integration of AI, encompassing machine learning (ML) and deep learning (DL), with the expanding accumulation of multi-omics data, holds considerable promise to bring transformative changes in areas such as cancer subtyping, risk stratification, prognostic evaluations, predictive analytics, and clinical decision-making processes ( 14 ).…”