Zhu HB, Xu D, Ye M, et al. - This research was sought to develop and validate a deep learning algorithm applying pre‐and post‐chemotherapy magnetic resonance imaging (MRI) to prognosticate pathological response in colorectal liver metastases (CRLM). Between January 2013 and November 2016, CRLM patients from center one who had ≤ 5 lesions and were scheduled to receive preoperative chemotherapy followed by liver resection, were included prospectively and chronologically classified into a training cohort (80% of patients) and a testing cohort (20% of patients). They included a total of 155 patients (328 lesions); chronologically, 101 (264 lesions) in the training cohort and 54 (64 lesions) in the testing cohort. In patients with CRLM, the MRI‐based deep learning model provided an accurate prediction of pathological tumor response to preoperative chemotherapy and may inform individualized treatment.