Prospective clinical research of radiomics and deep learning in oncology: A translational review. (November 2022)

• Record Type: Journal Article
• Authors: Zhang, Xingping; Zhang, Yanchun; Zhang, Guijuan; Qiu, Xingting; Tan, Wenjun; Yin, Xiaoxia; Liao, Liefa
• Is Part Of: Critical reviews in oncology/hematology. Volume 179(2022)
• Publication Date: 2022-11
• Subjects: DL Deep learning -- ML Machine learning -- ROI Region of interest -- VOI Volume of interest -- RQS Radiomics quality score -- TRIPOD Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis -- RT Radiotherapy -- CT Computed tomography -- SAT Soft tissue sarcoma -- MR Magnetic resonance -- PET- Positron emission tomography- -- US Ultrasound -- SWE Shear-wave elastography -- 3CB Three-compartment breast -- CEM Contrast-enhanced mammography -- PET Positron emission tomography -- MVI Microvascular invasion -- PFS Progression-free survival -- OS Overall survival -- DFS- Disease-free survival- -- AUC Area under the curve -- DTS Digital tomosynthesis -- HCC Hepatocellular carcinoma -- LNM Lymph node metastasis -- PTC Papillary thyroid carcinoma -- H&N Head and neck -- OPR Overall patient risk -- IDH Isocitrate dehydrogenase -- hFSRT Hypofractionated stereo- tactic radiotherapy -- NAC Neoadjuvant chemotherapy -- CCRT Concurrent chemoradiation therapy -- PDP Photodynamic priming -- RFS Recurrence-free survival -- pCR Pathologic complete response -- CNN Convolutional neural network -- DFS Disease-free survival -- HR Hazard ratio -- DSC Dice similarity coefficient -- IBSI Image Biomarkers Standardization Initiative -- BD Basic adherence; Prospective studies -- Clinical deployment -- Radiomics and deep learning -- Methodological robustness -- Oncology; Oncology -- Periodicals; Hematology -- Periodicals; 616.994
• DOI: 10.1016/j.critrevonc.2022.103823
• Languages: English
• ISSNs: 1040-8428

Abstract

Abstract: Radiomics and deep learning (DL) hold transformative promise and substantial and significant advances in oncology; however, most methods have been tested in retrospective or simulated settings. There is considerable interest in the biomarker validation, clinical utility, and methodological robustness of these studies and their deployment in real-world settings. This review summarizes the characteristics of studies, the level of prospective validation, and the overview of research on different clinical endpoints. The discussion of methodological robustness shows the potential for independent external replication of prospectively reported results. These in-depth analyses further describe the barriers limiting the translation of radiomics and DL into primary care options and provide specific recommendations regarding clinical deployment. Finally, we propose solutions for integrating novel approaches into the treatment environment to unravel the critical process of translating AI models into the clinical routine and explore strategies to improve personalized medicine. Highlights: Revisiting prospective clinical research of radiomics and deep learning in oncology. Visual analysis of prospective validation levels and methodological robustness. Propose clinical deployment endpoints based on application overview and challenges. Build an implementation framework to integrate AI models into the clinical process. Unveiling the translation of academic results in related fields to clinical routines.