【摘要】 肌肉减少症是指进行性、广泛性的骨骼肌质量及力量下降,以及由此导致的身体残疾、生活质量下降及死亡等不良后果的综合征。肌肉减少症常见于老年人以及肿瘤患者。肿瘤相关性肌肉减少症常见于多种实体瘤中,如消化道肿瘤、头颈部肿瘤、乳腺癌及肺癌等。其临床发病率高,不仅与放化疗不良反应以及围手术期并发症高度相关,更严重影响患者的生存预后。肌肉减少症的诊断方法可分为非影像学检查以及影像学检查。目前关于影像学评估诊断肿瘤相关性肌肉减少症的研究已经广泛开展起来,传统的诊断方法烦琐、费时费力。随着近年来计算机学科的发展,人工智能在海量数据、算力、算法的驱动下,可以快速精准做出影像学评估。本文对影像学评估肿瘤相关肌肉减少症的具体方法、人工智能助力评估方法的优劣势,以及荟萃相关研究的结果进行综述,旨在明确人工智能方法评价肿瘤相关肌肉减少症中的重要临床价值以及应用前景,为进一步开展大规模临床研究提供依据,同时提高临床医师对人工智能相关评估的关注度。更多还原

【Abstract】 Sarcopenia is a syndrome of progressive loss of skeletal muscle mass and strength, resulting in physical disability,reduced quality of life and death. Sarcopenia is common in the elderly and patients with cancer. Tumor combined with sarcopenia often occurs in the solid tumors such as gastrointestinal cancers, head and neck cancers, breast cancers and lung cancers. Due to frequent morbidity, tumor combined with sarcopenia is highly correlated with adverse reaction of radiation therapy and chemotherapy, and also with perioperative complications and diagnosis. Diagnostic methods for sarcopenia can be divided into non-imaging and imaging tests. At present, studies on imaging evaluation and diagnosis of tumor-associated sarcopenia have been carried out extensively.Traditional diagnostic methods are cumbersome, time-consuming and laborious. With the development of computer science in recent years, the method of artificial intelligence, driven by massive data, computing power and algorithms, can quickly and accurately accomplish imaging evaluation. This article reviews the specific imaging methods for the assessment of tumor-associated sarcopenia,the advantages and disadvantages of AI-assisted assessment methods, and the results of meta-studies. The aim is to clarify the important clinical value and application prospect of artificial intelligence method in evaluating tumor-associated sarcopenia, provide a basis for further large-scale clinical studies, and increase clinicians’ attention to AI-related evaluation.更多还原