Machine Learning and Radiomics Applications in Esophageal Cancers Using Non-Invasive Imaging Methods—A Critical Review of Literature

Abstract

Esophageal cancer (EC) is of public health significance as one of the leading causes of cancer death worldwide. Accurate staging, treatment planning and prognostication in EC patients are of vital importance. Recent advances in machine learning (ML) techniques demonstrate their potential to provide novel quantitative imaging markers in medical imaging. Radiomics approaches that could quantify medical images into high-dimensional data have been shown to improve the imaging-based classification system in characterizing the heterogeneity of primary tumors and lymph nodes in EC patients. In this review, we aim to provide a comprehensive summary of the evidence of the most recent developments in ML application in imaging pertinent to EC patient care. According to the published results, ML models evaluating treatment response and lymph node metastasis achieve reliable predictions, ranging from acceptable to outstanding in their validation groups. Patients stratified by ML models in different risk groups have a significant or borderline significant difference in survival outcomes. Prospective large multi-center studies are suggested to improve the generalizability of ML techniques with standardized imaging protocols and harmonization between different centers.

Atom Search Optimization with the Deep Transfer Learning-Driven Esophageal Cancer Classification Model

“…e occurrence rate was slowly increasing in males in Japan. Even though progressions were made in surgery and perioperative management policies, long-term diagnosis of esophageal cancer, speci cally in advanced levels, will be lower; the ve-year endurance rate of patients with phase IV EC is nearly 20% in Japan [1]. In several cases, the indication of common digestive signs associated with EC, like di culty in swallowing and heartburn, happens in developed phases.…”

Artificial intelligence and its impact on the domains of universal health coverage, health emergencies and health promotion: An overview of systematic reviews

“…The application of AI in UHC domain is mainly focused on neoplasms (N = 28) [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] , [41] , [42] , [43] , [44] , [45] , [46] , [47] and secondary on mental health (N = 17) [48] , [49] , [50] , [51] , [52] , [53] , [54] , [55] , [56] , [57] , [58] , [59] , [60] , [61] , [62] , [63] , [64] as depicted in Fig. 2 .…”

A review of radiomics and genomics applications in cancers: the way towards precision medicine

“…In addition, human delineation presents an inevitable bias resulting from different observers that cannot be ignored, leading to a lack of robustness due to intraand interobserver variations [25]. Semiautomatic delineation uses computer algorithms to segment the ROIs/ VOIs but usually needs to be corrected and calibrated manually [26]. Some primary open-source or commercial software could be applied to conduct semiautomatic segmentation, such as 3D Slicer [27], ITK-SNAP [28], LIFE [29], MITK [30], and ImageJ [31].…”