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

PurposeManual contour work for radiation treatment planning takes significant time to ensure volumes are accurately delineated. The use of artificial intelligence with deep learning based autosegmentation (DLAS) models has made itself known in recent years to alleviate this workload. It is used for organs at risk (OAR) contouring with significant consistency in performance and time saving. The purpose of this study was to evaluate the performance of current published data for DLAS of clinical target volume (CTV) contours, identify areas of improvement, and discuss future directions.MethodologyA literature review was performed by utilizing the key words “Deep Learning” AND (“Segmentation” OR “Delineation”) AND “Clinical Target Volume” in an indexed search into PubMed. A total of 154 articles based on the search criteria were reviewed. The review considered the DLAS model used, disease site, targets contoured, guidelines utilized, and the overall performance.ResultsOf the 53 articles investigating DLAS of CTV, only 6 were published before 2020. Publications have increased in recent years, with 46 articles published between 2020-2023. The cervix (n=19) and the prostate (n=12) were studied most frequently. Most studies (n=43) involved a single institution. Median sample size was 130 patients (range: 5-1,052). The most common metrics utilized to measure DLAS performance were Dice similarity coefficient (DSC) followed by Hausdorff distance. Dosimetric performance was seldom reported (n=11). There was also variability in specific guidelines utilized (RTOG, ESTRO, and others). DLAS models had good overall performance for contouring CTV volumes for multiple disease sites, with most studies showing DSC values >0.7. DLAS models also delineated CTV volumes faster compared to manual contouring. However, some DLAS model contours still required at least minor edits, and future studies investigating DLAS of CTV volumes require improvement.ConclusionsDLAS demonstrates capability of completing CTV contour plans with increased efficiency and accuracy. However, most models are developed and validated by single institutions using guidelines followed by the developing institutions. Publications about DLAS of the CTV have increased in recent years. Future studies and DLAS models need to include larger datasets with different patient demographics, disease stages, validation in multi-institutional settings, and inclusion of dosimetric performance. Manual contour work for radiation treatment planning takes significant time to ensure volumes are accurately delineated. The use of artificial intelligence with deep learning based autosegmentation (DLAS) models has made itself known in recent years to alleviate this workload. It is used for organs at risk (OAR) contouring with significant consistency in performance and time saving. The purpose of this study was to evaluate the performance of current published data for DLAS of clinical target volume (CTV) contours, identify areas of improvement, and discuss future directions. A literature review was performed by utilizing the key words “Deep Learning” AND (“Segmentation” OR “Delineation”) AND “Clinical Target Volume” in an indexed search into PubMed. A total of 154 articles based on the search criteria were reviewed. The review considered the DLAS model used, disease site, targets contoured, guidelines utilized, and the overall performance. Of the 53 articles investigating DLAS of CTV, only 6 were published before 2020. Publications have increased in recent years, with 46 articles published between 2020-2023. The cervix (n=19) and the prostate (n=12) were studied most frequently. Most studies (n=43) involved a single institution. Median sample size was 130 patients (range: 5-1,052). The most common metrics utilized to measure DLAS performance were Dice similarity coefficient (DSC) followed by Hausdorff distance. Dosimetric performance was seldom reported (n=11). There was also variability in specific guidelines utilized (RTOG, ESTRO, and others). DLAS models had good overall performance for contouring CTV volumes for multiple disease sites, with most studies showing DSC values >0.7. DLAS models also delineated CTV volumes faster compared to manual contouring. However, some DLAS model contours still required at least minor edits, and future studies investigating DLAS of CTV volumes require improvement. DLAS demonstrates capability of completing CTV contour plans with increased efficiency and accuracy. However, most models are developed and validated by single institutions using guidelines followed by the developing institutions. Publications about DLAS of the CTV have increased in recent years. Future studies and DLAS models need to include larger datasets with different patient demographics, disease stages, validation in multi-institutional settings, and inclusion of dosimetric performance.