Sarcopenia, also known as muscle decay, is associated with high morbidity among surgical patients. It is highly correlated with adverse clinical outcomes, such as increased postoperative complications, prolonged hospital stay, and increased mortality. Computed tomography (CT) is one of the main methods for diagnosing sarcopenia, which has the advantages of intuitiveness, rapidity, and accuracy. Clinical studies have shown that CT‐defined sarcopenia can help predict the clinical outcomes and prognosis of surgical patients and provide an important reference for the formulation of antitumor treatment protocols. In recent years, some scholars have tried to construct an intelligent CT‐based diagnostic model, which is expected to improve the diagnostic efficiency and establish standardized diagnostic criteria for CT‐defined sarcopenia. In this review, we summarize the recent progress in the understanding of the diagnosis of sarcopenia and its potential prognostic value in surgery.