A pre-operative imaging landmark to define the rectum would optimise clinical care of rectal cancer patients and research efforts to improve outcomes. The sigmoid take-off has been suggested as an imaging landmark for the rectosigmoid junction (RSJ). This study aimed to investigate whether this imaging definition of the rectum was validated by surgical specimen analysis. This prospective study recruited 20 patients undergoing surgery and undertook radiological and pathological analysis of their rectal specimens. The radiological landmark of the sigmoid take-off was identified on pre-operative magnetic resonance imaging (MRI), and the distance to the anterior peritoneal reflection was measured by two readers. After surgery, the distance from the beginning of the sigmoid mesocolon to the anterior peritoneal reflection to the beginning of the sigmoid mesocolon on the specimen was measured, and compared to the distance on MRI using Pearson's Correlation Coefficient and Bland-Altman plots. In 17 patients, the mean distance from the anterior peritoneal reflection to the RSJ on MRI was 20.3 mm and 23.1 mm for two readers, and on pathology was 20.6 mm. The mean differences between MRI and specimen measurements were -0.31 mm (-2.83 to 2.20 mm), and 2.51 mm (95% confidence interval -0.31 to 5.33 mm) for each reader, with correlation coefficients of 0.77 and 0.81. The sigmoid take-off has been validated on specimen analysis to be an imaging landmark that defines the termination of the rectum. This anatomical landmark can be used to classify tumours and guide treatment and research of sigmoid colon and rectal cancer. Crown Copyright © 2020. Published by Elsevier Ltd. All rights reserved.