Abstract Background Endoscopic improvement (EI), Mayo endoscopic subscore of 0 or 1, is considered a therapeutic target in ulcerative colitis (UC) treatment. The potential to estimate EI non-invasively is an advantage of intestinal ultrasound (IUS). In a former study, we developed a new ultrasound parameter, the submucosal index (SMI), calculated as the ratio of the thickness of the submucosa to the thickness of the bowel wall (BWT), and reported that combining BWT and SMI could be a practical criterion for estimating EI without the assessment of bowel wall vascularity using color Doppler evaluation. In the present study, we validated the EI estimation ability of our B-mode-based UC criterion, Kyorin Ultrasound Criterion for UC (KUC-UC; BWT<3.8mm and SMI<50%), using an external cohort. Methods Patients who underwent IUS and CS at Kyorin University Hospital were included. The inclusion criteria were (1) both IUS and endoscopy (sigmoidoscopy or CS) for UC were performed between July 2021 and June 2022, (2) the interval between IUS and endoscopy was within 15 days, and (3) no change or addition of therapeutic agents between the two examinations. IUS findings including BWT, SMI, and modified Limberg score were evaluated. In the modified Limberg score, the threshold of normal BWT was less than 3 mm, while the original Limberg score employed less than 4 mm. Results The KUC-UC criterion defined as BWT < 3.8 mm and SMI < 50.0% is practical because it uses only B-mode findings and does not require complex calculations. We tested the ability of this criterion to estimate EI in UC. A total of 122 colon segments were evaluated by transabdominal IUS and CS. The criteria showed a sensitivity of 67.3%, specificity of 97.1%, PPV of 94.6%, and NPV of 80.0%. In the previous development cohort, the PPV and NPV of KUC-UC were 95.5% and 82.7%, respectively. Based on the above, this validation cohort is considered equivalent to the development cohort. BWT < 3 mm and MUC have been reported as useful criteria for EI estimation. In the present cohort, BWT <3mm showed a sensitivity of 67.3%, specificity of 91.4%, PPV of 85.4%, and NPV of 79.0%, and MUC demonstrated a sensitivity of 75.0%, specificity of 88.6%, PPV of 83.0%, and NPV of 82.0%. These results indicate that KUC-UC has a considerable potential to estimate EI as BWT < 3 mm or MUC. Conclusion External validation showed that KUC-UC, using only B-mode findings without complicated calculations, is feasible, accurate sonographic criteria to estimate the EI of UC. The KUC-UC criterion could be one of the options for estimating EI in UC.