Clinical evaluation of a deep learning model for segmentation of target volumes in breast cancer radiotherapy

Buelens, P.; Ir, S Willems; Vandewinckele, Liesbeth; Crijns, Wouter; Maes, Frederik; Weltens, Caroline

doi:10.1016/j.radonc.2022.04.015

Cited by 9 publications

(6 citation statements)

References 28 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 43 Additionally, there is a need for standardised validation metrics to evaluate clinical acceptability and commissioning of AS models. 44 The CDV measures used in this study indicate AS variations in the x, y, and z axes, highlighting the importance of modifications based on clinical relevance. The software solution used for data processing was validated against a digital phantom through IBSI international collaboration.…”

Section: Discussionmentioning

confidence: 85%

Clinical evaluation of atlas-based auto-segmentation in breast and nodal radiotherapy

Welgemoed

Spezi

Riddle

et al. 2023

The British Journal of Radiology

View full text Add to dashboard Cite

Objectives: Accurate contouring of anatomical structures allows for high-precision radiotherapy planning, targeting the dose at treatment volumes and avoiding organs at risk. Manual contouring is time-consuming with significant user variability, whereas auto-segmentation (AS) has proven efficiency benefits but requires editing before treatment planning. This study investigated whether atlas-based AS (ABAS) accuracy improves with template atlas group size and character-specific atlas and test case selection. Methods and materials: One clinician retrospectively contoured the breast, nodes, lung, heart, and brachial plexus on 100 CT scans, adhering to peer-reviewed guidelines. Atlases were clustered in group sizes, treatment positions, chest wall separations, and ASs created with Mirada software. The similarity of ASs compared to reference contours was described by the Jaccard similarity coefficient (JSC) and centroid distance variance (CDV). Results: Across group sizes, for all structures combined, the mean JSC was 0.6 (SD 0.3, p = .999). Across atlas-specific groups, 0.6 (SD 0.3, p = 1.000). The correlation between JSC and structure volume was weak in both scenarios (adjusted R2−0.007 and 0.185). Mean CDV was similar across groups but varied up to 1.2 cm for specific structures. Conclusions: Character-specific atlas groups and test case selection did not improve accuracy outcomes. High-quality ASs were obtained from groups containing as few as ten atlases, subsequently simplifying the application of ABAS. CDV measures indicating auto-segmentation variations on the x, y, and z axes can be utilised to decide on the clinical relevance of variations and reduce AS editing. Advances in knowledge: High-quality ABASs can be obtained from as few as ten template atlases. Atlas and test case selection do not improve AS accuracy. Unlike well-known quantitative similarity indices, volume displacement metrics provide information on the location of segmentation variations, helping assessment of the clinical relevance of variations and reducing clinician editing. Volume displacement metrics combined with the qualitative measure of clinician assessment could reduce user variability.

show abstract

Section: Discussionmentioning

confidence: 85%

Clinical evaluation of atlas-based auto-segmentation in breast and nodal radiotherapy

Welgemoed

Spezi

Riddle

et al. 2023

The British Journal of Radiology

View full text Add to dashboard Cite

show abstract

“…Performance was comparable between CTV right and left breast in studies that reported CTV performance in both breasts. 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 Dai et al performed a multi-institutional study that reported better performance for DLAS models on planning CT scans compared with scanning CT scans. 21 Lymph node CTVs are the most difficult fields for DLAS models.…”

Section: Resultsmentioning

confidence: 99%

“…Studies reported worse DLAS performance for CTV internal mammary lymph nodes (DSC, 0.51-0.60) and Rotter's space (DSC, 0.63). 23 , 24 Choi et al reported on several different lymph node levels with the lowest DLAS model performance for ESTRO guideline left CTV supraclavicular nodes (DSC, 0.7). 18 Almberg et al reported worst DLAS model performance for CTV interpectoral lymph nodes (DSC, 0.7).…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Review of Deep Learning Based Autosegmentation for Clinical Target Volume: Current Status and Future Directions

Matoska,

Patel,

Liu

et al. 2024

Advances in Radiation Oncology

View full text Add to dashboard Cite

“…Our results are consistent with other previous studies addressing this same topic. 6 , 18 , 19 We must mention that no lymphatic pathway contours were considered in this segment, due to technical reasons. The software version available at the moment of the study counted with the American Society for Radiation Oncology templates, while our contours are routinely performed according to the European Society for Radiotherapy and Oncology recommendations.…”

Section: Discussionmentioning

confidence: 99%

Artificial Intelligence–Based Autosegmentation: Advantages in Delineation, Absorbed Dose-Distribution, and Logistics

Sarria,

Kugel,

Roehner

et al. 2024

Advances in Radiation Oncology

View full text Add to dashboard Cite

Clinical evaluation of a deep learning model for segmentation of target volumes in breast cancer radiotherapy

Cited by 9 publications

References 28 publications

Clinical evaluation of atlas-based auto-segmentation in breast and nodal radiotherapy

Clinical evaluation of atlas-based auto-segmentation in breast and nodal radiotherapy

Review of Deep Learning Based Autosegmentation for Clinical Target Volume: Current Status and Future Directions

Artificial Intelligence–Based Autosegmentation: Advantages in Delineation, Absorbed Dose-Distribution, and Logistics

Contact Info

Product

Resources

About