TY - JOUR
T1 - Resection cavity auto-contouring for patients with pediatric medulloblastoma using only CT information
AU - Hernandez, Soleil
AU - Nguyen, Callistus
AU - Gay, Skylar
AU - Duryea, Jack
AU - Howell, Rebecca
AU - Fuentes, David
AU - Parkes, Jeannette
AU - Burger, Hester
AU - Cardenas, Carlos
AU - Paulino, Arnold C.
AU - Pollard-Larkin, Julianne
AU - Court, Laurence
N1 - Publisher Copyright:
© 2023 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, LLC on behalf of The American Association of Physicists in Medicine.
PY - 2023/7
Y1 - 2023/7
N2 - Purpose: Target delineation for radiation therapy is a time-consuming and complex task. Autocontouring gross tumor volumes (GTVs) has been shown to increase efficiency. However, there is limited literature on post-operative target delineation, particularly for CT-based studies. To this end, we trained a CT-based autocontouring model to contour the post-operative GTV of pediatric patients with medulloblastoma. Methods: One hundred four retrospective pediatric CT scans were used to train a GTV auto-contouring model. Eighty patients were then preselected for contour visibility, continuity, and location to train an additional model. Each GTV was manually annotated with a visibility score based on the number of slices with a visible GTV (1 = < 25%, 2 = 25–50%, 3 = > 50–75%, and 4 = > 75–100%). Contrast and the contrast-to-noise ratio (CNR) were calculated for the GTV contour with respect to a cropped background image. Both models were tested on the original and pre-selected testing sets. The resulting surface and overlap metrics were calculated comparing the clinical and autocontoured GTVs and the corresponding clinical target volumes (CTVs). Results: Eighty patients were pre-selected to have a continuous GTV within the posterior fossa. Of these, 7, 41, 21, and 11 were visibly scored as 4, 3, 2, and 1, respectively. The contrast and CNR removed an additional 11 and 20 patients from the dataset, respectively. The Dice similarity coefficients (DSC) were 0.61 ± 0.29 and 0.67 ± 0.22 on the models without pre-selected training data and 0.55 ± 13.01 and 0.83 ± 0.17 on the models with pre-selected data, respectively. The DSC on the CTV expansions were 0.90 ± 0.13. Conclusion: We successfully automatically contoured continuous GTVs within the posterior fossa on scans that had contrast > ± 10 HU. CT-Based auto-contouring algorithms have potential to positively impact centers with limited MRI access.
AB - Purpose: Target delineation for radiation therapy is a time-consuming and complex task. Autocontouring gross tumor volumes (GTVs) has been shown to increase efficiency. However, there is limited literature on post-operative target delineation, particularly for CT-based studies. To this end, we trained a CT-based autocontouring model to contour the post-operative GTV of pediatric patients with medulloblastoma. Methods: One hundred four retrospective pediatric CT scans were used to train a GTV auto-contouring model. Eighty patients were then preselected for contour visibility, continuity, and location to train an additional model. Each GTV was manually annotated with a visibility score based on the number of slices with a visible GTV (1 = < 25%, 2 = 25–50%, 3 = > 50–75%, and 4 = > 75–100%). Contrast and the contrast-to-noise ratio (CNR) were calculated for the GTV contour with respect to a cropped background image. Both models were tested on the original and pre-selected testing sets. The resulting surface and overlap metrics were calculated comparing the clinical and autocontoured GTVs and the corresponding clinical target volumes (CTVs). Results: Eighty patients were pre-selected to have a continuous GTV within the posterior fossa. Of these, 7, 41, 21, and 11 were visibly scored as 4, 3, 2, and 1, respectively. The contrast and CNR removed an additional 11 and 20 patients from the dataset, respectively. The Dice similarity coefficients (DSC) were 0.61 ± 0.29 and 0.67 ± 0.22 on the models without pre-selected training data and 0.55 ± 13.01 and 0.83 ± 0.17 on the models with pre-selected data, respectively. The DSC on the CTV expansions were 0.90 ± 0.13. Conclusion: We successfully automatically contoured continuous GTVs within the posterior fossa on scans that had contrast > ± 10 HU. CT-Based auto-contouring algorithms have potential to positively impact centers with limited MRI access.
KW - autocontouring
KW - pediatric medulloblastoma
KW - radiation therapy
KW - target delineation
UR - http://www.scopus.com/inward/record.url?scp=85150647371&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85150647371&partnerID=8YFLogxK
U2 - 10.1002/acm2.13956
DO - 10.1002/acm2.13956
M3 - Article
C2 - 36917640
AN - SCOPUS:85150647371
SN - 1526-9914
VL - 24
JO - Journal of applied clinical medical physics
JF - Journal of applied clinical medical physics
IS - 7
M1 - e13956
ER -