Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Mar 14;25(1):88.
doi: 10.1186/s12880-025-01614-3.

AI-ready rectal cancer MR imaging: a workflow for tumor detection and segmentation

Heather M Selby #  1 , Yewon A Son #  2 , Vipul R Sheth  3 , Todd H Wagner  2 , Erqi L Pollom  4 , Arden M Morris  2
Affiliations

AI-ready rectal cancer MR imaging: a workflow for tumor detection and segmentation

Heather M Selby et al. BMC Med Imaging. .

Abstract

Background: Magnetic Resonance (MR) imaging is the preferred modality for staging in rectal cancer; however, despite its exceptional soft tissue contrast, segmenting rectal tumors on MR images remains challenging due to the overlapping appearance of tumor and normal tissues, variability in imaging parameters, and the inherent subjectivity of reader interpretation. For studies requiring accurate segmentation, reviews by multiple independent radiologists remain the gold standard, albeit at a substantial cost. The emergence of Artificial Intelligence (AI) offers promising solutions to semi- or fully-automatic segmentation, but the lack of publicly available, high-quality MR imaging datasets for rectal cancer remains a significant barrier to developing robust AI models.

Objective: This study aimed to foster collaboration between a radiologist and two data scientists in the detection and segmentation of rectal tumors on T2- and diffusion-weighted MR images. By combining the radiologist's clinical expertise with the data scientists' imaging analysis skills, we sought to establish a foundation for future AI-driven approaches that streamline rectal tumor detection and segmentation, and optimize workflow efficiency.

Methods: A total of 37 patients with rectal cancer were included in this study. Through radiologist-led training, attendance at Stanford's weekly Colorectal Cancer Multidisciplinary Tumor Board (CRC MDTB), and the use of radiologist annotations and clinical notes in Epic Electronic Health Records (EHR), data scientists learned how to detect and manually segment tumors on T2- and diffusion-weighted pre-treatment MR images. These segmentations were then reviewed and edited by a radiologist. The accuracy of the segmentations was evaluated using the Dice Similarity Coefficient (DSC) and Jaccard Index (JI), quantifying the overlap between the segmentations delineated by the data scientists and those edited by the radiologist.

Results: With the help of radiologist annotations and radiology notes in Epic EHR, the data scientists successfully identified rectal tumors in Slicer v5.7.0 across all evaluated T2- and diffusion-weighted MR images. Through radiologist-led training and participation at Stanford's weekly CRC MDTB, the data scientists' rectal tumor segmentations exhibited strong agreement with the radiologist's edits, achieving a mean DSC [95% CI] of 0.965 [0.939-0.992] and a mean JI [95% CI] of 0.943 [0.900, 0.985]. Discrepancies in segmentations were attributed to over- or under-segmentation, often incorporating surrounding structures such as the rectal wall and lumen.

Conclusion: This study demonstrates the feasibility of generating high-quality labeled MR datasets through collaboration between a radiologist and two data scientists, which is essential for training AI models to automate tumor detection and segmentation in rectal cancer. By integrating expertise from radiology and data science, this approach has the potential to enhance AI model performance and transform clinical workflows in the future.

Keywords: Artificial intelligence; Magnetic resonance imaging; Rectal cancer; Segmentation.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: This study was approved by the Institutional Review Board (IRB) of Stanford University (IRB-62555) in accordance with the ethical guidelines established by the Stanford School of Medicine and U.S. federal regulations. All procedures were carried out in compliance with institutional and national ethical standards for research involving human participants. Informed consent was obtained from all participants included in the study. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Workflow for the detection and manual segmentation of rectal tumors on MR image. Created in BioRender, Selby, H. (2025) https://BioRender.com/i64b822
Fig. 2
Fig. 2
Overview of image pre-processing steps for data harmonization, including (1) ROI segmentation (yellow) delineated by data scientist and reviewed by radiologist, (2) isotropic resampling of MR images to a uniform spacing of 1 mm in the x, y, and z planes using B-Spline interpolation to preserve anatomical details, while segmentations were resampled using Nearest Neighbor (NN) interpolation to maintain boundary precision, balancing anatomical accuracy with segmentation fidelity, (3) extraction of the maximum connected component to remove stray pixels (denoted by red arrow), and (4) hole filling. Created in BioRender. Selby, H. (2025) https://BioRender.com/e41a602
Fig. 3
Fig. 3
Selected axial T2-weighted MR slices from a patient (SFX-024 in Table 3) with T3N0 rectal cancer, highlighting challenging segmentation cases. The MR images are overlaid with tumor segmentations that were (a) over-segmented, (b) under-segmented, and (c) over-segmented by the data scientist (yellow) with subsequent edits by the radiologist (purple). Image (d) presents a 3D visualization of the segmented tumor, illustrating the spatial relationship and overlap between the data scientist’s segmentation (yellow) and the radiologist’s edits (purple). The DSC and JI between the segmentations performed by the data scientist and the radiologist were 0.98 and 0.96, respectively. Created in BioRender. Selby, H. (2025) https://BioRender.com/t87e418
See this image and copyright information in PMC

References

    1. Horvat N, Papanikolaou N, Koh DM. Radiomics Beyond the Hype: A Critical Evaluation Toward Oncologic Clinical Use. Radiol Arti Intell. 2024;6(4):e230437. 10.1148/ryai.230437. - PMC - PubMed
    1. Horvat N, Carlos Tavares Rocha C, Clemente Oliveira B, Petkovska I, Gollub MJ. MRI of Rectal Cancer: Tumor Staging, Imaging Techniques, and Management. Radiographics Rev Publ Radiol Soc N Am Inc. 2019;39(2):367–87. 10.1148/rg.2019180114. - PMC - PubMed
    1. Rao SX, Zeng MS, Chen CZ, Li RC, Zhang SJ, Xu JM, et al. The value of diffusion-weighted imaging in combination with T2-weighted imaging for rectal cancer detection. Eur J Radiol. 2008;65(2):299–303. 10.1016/j.ejrad.2007.04.001. - PubMed
    1. Song I, Kim SH, Lee SJ, Choi JY, Kim MJ, Rhim H. Value of diffusion-weighted imaging in the detection of viable tumour after neoadjuvant chemoradiation therapy in patients with locally advanced rectal cancer: comparison with Tformula image weighted and PET/CT imaging. Br J Radiol. 2012;85(1013):577–86. 10.1259/bjr/68424021. - PMC - PubMed
    1. Sharma N, Aggarwal L. Automated medical image segmentation techniques. J Med Phys. 2010;35(1):3. 10.4103/0971-6203.58777. - PMC - PubMed

LinkOut - more resources