Inter-observer and sequence variability in pelvic organs at risk delineation of Magnetic Resonance images

Variability analysis of organs at risk on MRI

Authors

  • Sijuan Huang State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-Sen University Cancer Center, Guangzhou 510060, Guangdong Province, China.
  • Wanjia Zheng Department of Radiation Oncology, Southern Theater Air Force Hospital of the People’s Liberation Army, Guangzhou 510050, Guangdong Province, China.
  • Xin Yang State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-Sen University Cancer Center, Guangzhou 510060, Guangdong Province, China.
  • Zesen Cheng School of Electronic and Computer Engineering, Peking University, Shenzhen, Guangdong province, 518055, China
  • Jinxing Liang Department of Radiation Oncology, The First Affiliated Hospital of Guangzhou University of Traditional Chinese Medicine, Guangzhou, Guangdong province, 510405, China
  • Enting Li Department of Radiology, The Eighth Affiliated Hospital of Sun Yat-sen University, Shenzhen, Guangdong province, 518033, China
  • Shaolin Mo Department of Radiology, The First People’s Hospital of Foshan, Foshan, Guangdong province, 528010, China
  • Yimei Liu State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-Sen University Cancer Center, Guangzhou 510060, Guangdong Province, China.

Abstract

Background This study evaluates the contouring variability among observers using MR images reconstructed by different sequences and quantifies the differences of automatic segmentation models for different sequences.

Materials and methods Eighty-nine patients with pelvic tumors underwent T1WI, T1dixonc, and T2WI MR imaging on a simulator. Two observers performed manual delineation of the bladder, anal canal, rectum, and femoral heads on all images. Contour differences were used to analyze the interobserver and intersequence variability. A single-sequence automatic segmentation network was established using the U-Net network, and the segmentation results were analyzed.

Results Variability analysis among observers showed that the bladder, rectum, and left femoral head on T1WI yielded the highest DSC and the lowest 95% HD (all three sequences). Regarding sequence variability analysis for the same observer, the difference between T1WI and T2WI was the smallest. DSC of the bladder, rectum, and femoral heads exceeded 0.88 for T1WI–T2WI. The differences between automatic segmentations and manual delineations were minimal on T2WI. The averaged DSC of automatic and manual segmentation of all organs on T2WI exceeded 0.81, and the averaged 95% HD value was lower than 7 mm. Similarly, the sequence variability analysis of automatic segmentation indicates that the automatic segmentation differences between T2WI and T1WI are minimal.

Conclusions T1WI and T2WI yielded better results in manual delineation and automatic segmentation, respectively. The analysis of variability among three sequences indicates that the yielded good similarity outcomes between the T1WI and T2WI cases in manual and automatic segmentation. We infer that the T1WI and T2WI (or their combination) can be used for MR-only radiation therapy.

Key words: MRI, Multiple sequences, variability, Automatic segmentation

Downloads

Additional Files

Published

2025-02-25

How to Cite

Huang, S., Zheng, W., Yang, X., Cheng, Z., Liang, J., Li, E., … Liu, Y. (2025). Inter-observer and sequence variability in pelvic organs at risk delineation of Magnetic Resonance images: Variability analysis of organs at risk on MRI. Radiology and Oncology, 59(1), 139–146. Retrieved from https://www.radioloncol.com/index.php/ro/article/view/4443

Issue

Vol. 59 No. 1 (2025)

Section

Clinical oncology

License

Copyright (c) 2025 Sijuan Huang, Wanjia Zheng, Xin Yang, Zesen Cheng, Jinxing Liang, Enting Li, Shaolin Mo, Yimei Liu

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

License to Publish

Please read the terms of this agreement, print, initial page 1, sign page 3, scan and send the document as one file attached to an e-mail to gsersa@onko-i.si

Copyright - PDF