Machine Learning and Radiogenomics: Lessons Learned and Future Directions

  title={Machine Learning and Radiogenomics: Lessons Learned and Future Directions},
  author={John Kang and Tiziana Rancati and Sangkyun Lee and Jung Hun Oh and Sarah L. Kerns and Jacob G. Scott and Russell Schwartz and Seyoung Kim and Barry S. Rosenstein},
  journal={Frontiers in Oncology},
Due to the rapid increase in the availability of patient data, there is significant interest in precision medicine that could facilitate the development of a personalized treatment plan for each patient on an individual basis. Radiation oncology is particularly suited for predictive machine learning (ML) models due to the enormous amount of diagnostic data used as input and therapeutic data generated as output. An emerging field in precision radiation oncology that can take advantage of ML… 

The Era of Radiogenomics in Precision Medicine: An Emerging Approach to Support Diagnosis, Treatment Decisions, and Prognostication in Oncology

The integrated process of radiogenomics is summarized and the crucial strategies and statistical algorithms involved in current studies are introduced to facilitate computer-aided diagnosis, treatment, and prediction of the prognosis in patients with tumors in the routine clinical setting.

Artificial intelligence applications for pediatric oncology imaging

This article provides a focused review of emerging AI applications that are relevant for the pediatric oncology imaging community that can help to improve the accuracy and efficiency of cancer diagnosis, selection of personalized therapies and prediction of long-term outcomes.

Prostate Cancer Radiogenomics—From Imaging to Molecular Characterization

This work aimed to review the current literature of highly quantitative and qualitative results from well-designed studies for the diagnoses, treatment, and follow-up of prostate cancer, based on radiomics, genomics and radiogenomics research.

Radiomic and radiogenomic modeling for radiotherapy: strategies, pitfalls, and challenges

Technical approaches toward integrating clinical, biological, imaging, and treatment-specific data for more accurate prediction of tumor control probabilities or risk of radiation-induced side effects are discussed and classical analytical and data-driven approaches for outcomes prediction that serve as antecedents to both radiomic and radiogenomic strategies are reviewed.

Radiogenomics in Interventional Oncology

Radiogenomics, whether in terms of methodologies or applications, is still in the early stages of development and far from maturation, and there is room for development and increased value of these applications.

Radiogenomics of lung cancer

Novel concepts in ML and AI are highlighted and their potential applications in identifying radiobiogenomics of lung cancer, which involves image segmentation, feature extraction, and ML model to predict underlying tumor genotype and clinical outcomes.

Radiogenomics models in precision radiotherapy: from mechanistic to machine learning

A review of radiogenomics modeling frameworks and efforts towards genomically-guided radiotherapy and similar efforts to create clinical assays for normal tissue or tumor radiosensitivity are discussed.

Genomics models in radiotherapy: From mechanistic to machine learning.

A review of radiogenomics modeling frameworks and efforts toward genomically guided radiotherapy and medical oncology efforts to develop precision biomarkers is provided.

The application of radiomics in predicting gene mutations in cancer

The current role of radiogenomics in predicting gene mutations in brain, lung, colorectal, breast, and kidney tumours is summarized and may serve as a substitute tool for genetic testing.

Artificial intelligence in oncology, its scope and future prospects with specific reference to radiation oncology

Artificial intelligence with the use of machine learning and artificial neural networks has come up with faster and more accurate solutions for the problems faced by oncologist.



Radiogenomics and radiotherapy response modeling

An overview of radiogenomics starting with genotyping, data aggregation, and application of different modeling approaches based on modifying traditional radiobiological methods or application of advanced machine learning techniques is provided.

Big Data Analytics for Prostate Radiotherapy

Radiation therapy is a first-line treatment option for localized prostate cancer and radiation-induced normal tissue damage are often the main limiting factor for modern radiotherapy regimens.

Radiogenomics: Identification of Genomic Predictors for Radiation Toxicity.

Predicting outcomes in radiation oncology—multifactorial decision support systems

An overview of the factors that are correlated with outcome—including survival, recurrence patterns and toxicity—in radiation oncology is provided and the methodology behind the development of prediction models is discussed, which is a multistage process.

Machine learning for radiation oncology

This paper aimed to review the radiotherapy (RT) process by identifying the specific areas in which the quality and efficiency of ML can be increased and a workflow chart can be created.

Machine Learning Approaches for Predicting Radiation Therapy Outcomes: A Clinician's Perspective.

A Bayesian network approach for modeling local failure in lung cancer

The preliminary results show that the proposed integrated approach for predicting post-radiotherapy local failure in NSCLC patients can be used as an efficient method to develop predictive models of local failure and to interpret relationships among the different variables in the models.

Exploiting biological and physical determinants of radiotherapy toxicity to individualize treatment

Combining radiogenomics with individual estimates of DA might identify almost half of patients undergoing radical RT who might benefit from either dose escalation, suggesting improved tumour cure or reduced toxicity or both.

Establishment of a Radiogenomics Consortium.

Bayesian network ensemble as a multivariate strategy to predict radiation pneumonitis risk.

The authors' preliminary results demonstrate that the BN framework combined with an ensemble method can possibly improve prediction of RP under real-life clinical circumstances such as missing data or treatment plan adaptation.