Mid to low-lying rectal cancer patient with enlarged lateral lymph node befor treatment and receiving lateral lymph node dissection would be enrolled. Radiomics parameters of lateral lymph nodes would be extracted by expert software. Then a part of the nodes would be used as the training set to build the prediction model, and the another part nodes would be used as the validation set.

After performing lateral lymph node dissection for the enrolled patients, we would find the targeted lateral node which was matched with the enlarged one in imaging from the specimen. Then the nodes would be send for pathological examination. Then we would extract the radiomics parameters of the nodes from the pretreatment CT and MRI with ITK-SNAP software. According to the pathological status, lymph nodes and corresponding radiomics parameters would be divided into positive and negative groups. Then significant parameters would be selected by R software. Then the prediction model with nomogram would be developed. 70% nodes would be used for for model development and the other 30% would be used for validation.

Inclusion Criteria:

• middle to low-lying rectal adenocarcinoma patients with enlarged lateral lymph node ≥5mm for short diameter in CT and/or MRI

Exclusion Criteria:

• other kind of rectal tumor including neuroendocrine and malignant.
• high location rectal cancer.
• recurrent rectal cancer

patients with rectal cancer.

individual participant data (IPD) would not be available to other researchers.