Prog Med Phys.  2024 Dec;35(4):205-213. 10.14316/pmp.2024.35.4.205.

Institution-Specific Autosegmentation for Personalized Radiotherapy Protocols

• Cho W
• Yoo GS
• Kim WD
• Kim Y
• Kim JS
• Min BJ

• KMID: 2564158
• DOI: http://doi.org/10.14316/pmp.2024.35.4.205

Abstract

Purpose
This study explores the potential of artificial intelligence (AI) in optimizing radiotherapy protocols for personalized cancer treatment. Specifically, it investigates the role of AI-based segmentation tools in improving accuracy and efficiency across various anatomical regions.
Methods
A dataset of 500 anonymized patient computed tomography scans from Chungbuk National University Hospital was used to develop and validate AI models for segmenting organs-atrisk. The models were tailored for five anatomical regions: head and neck, chest, abdomen, breast, and pelvis. Performance was evaluated using Dice Similarity Coefficient (DSC), Mean Surface Distance, and the 95th Percentile Hausdorff Distance (HD95).
Results
The AI models achieved high segmentation accuracy for large, well-defined structures such as the brain, lungs, and liver, with DSC values exceeding 0.95 in many cases. However, challenges were observed for smaller or complex structures, including the optic chiasm and rectum, with instances of segmentation failure and infinity values for HD95. These findings highlight the variability in performance depending on anatomical complexity and structure size.
Conclusions
AI-based segmentation tools demonstrate significant potential to streamline radiotherapy workflows, reduce inter-observer variability, and enhance treatment accuracy. Despite challenges with smaller structures, the integration of AI enables dynamic, patient-specific adaptations to anatomical changes, contributing to more precise and effective cancer treatments. Future work should focus on refining models for anatomically complex structures and validating these methods in diverse clinical settings.

Keyword

AI-driven segmentation; Personalized cancer treatment; Adaptive radiotherapy; Transfer learning

Figure

• 

  Fig. 1 Boxplots of segmentation performance metrics across regions (head and neck, chest, abdomen, breast, pelvis), highlighting accuracy (DSC), surface agreement (MSD), and boundary deviations (HD95). DSC, Dice Similarity Coefficient; MSD, Mean Surface Distance; HD95, 95th Percentile Hausdorff Distance.

• 

  Fig. 2 Comparison of Dice Similarity Coefficient (DSC) values across various anatomical regions (head and neck, chest, abdomen, breast, and pelvis) between the default vendor-provided segmentation model (green) and the fine-tuned model (orange). The boxplots represent the distribution of segmentation accuracy for each region, highlighting improvements achieved with fine-tuning.

Reference

References

1. Chen HHW, Kuo MT. 2017; Improving radiotherapy in cancer treatment: promises and challenges. Oncotarget. 8:62742–62758. DOI: 10.18632/oncotarget.18409. PMID: 28977985. PMCID: PMC5617545.

2. Baroudi H, Brock KK, Cao W, Chen X, Chung C, Court LE, et al. 2023; Automated contouring and planning in radiation therapy: what is 'clinically acceptable'? Diagnostics (Basel). 13:667. DOI: 10.3390/diagnostics13040667. PMID: 36832155. PMCID: PMC9955359.

3. Dona Lemus OM, Cao M, Cai B, Cummings M, Zheng D. 2024; Adaptive radiotherapy: next-generation radiotherapy. Cancers (Basel). 16:1206. DOI: 10.3390/cancers16061206. PMID: 38539540. PMCID: PMC10968833.

4. Fu Y, Zhang H, Morris ED, Glide-Hurst CK, Pai S, Traverso A, et al. 2022; Artificial intelligence in radiation therapy. IEEE Trans Radiat Plasma Med Sci. 6:158–181. DOI: 10.1109/TRPMS.2021.3107454. PMID: 35992632. PMCID: PMC9385128.

5. Jeong C, Goh Y, Kwak J. 2024; Challenges and opportunities to integrate artificial intelligence in radiation oncology: a narrative review. Ewha Med J. 47:e49. DOI: 10.12771/emj.2024.e49.

6. Krishnamurthy R, Mummudi N, Goda JS, Chopra S, Heijmen B, Swamidas J. 2022; Using artificial intelligence for optimization of the processes and resource utilization in radiotherapy. JCO Glob Oncol. 8:e2100393. DOI: 10.1200/GO.21.00393. PMID: 36395438. PMCID: PMC10166445.

7. Kawamura M, Kamomae T, Yanagawa M, Kamagata K, Fujita S, Ueda D, et al. 2024; Revolutionizing radiation therapy: the role of AI in clinical practice. J Radiat Res. 65:1–9. DOI: 10.1093/jrr/rrad090. PMID: 37996085. PMCID: PMC10803173.

8. Wong J, Huang V, Wells D, Giambattista J, Giambattista J, Kolbeck C, et al. 2021; Implementation of deep learning-based auto-segmentation for radiotherapy planning structures: a workflow study at two cancer centers. Radiat Oncol. 16:101. DOI: 10.1186/s13014-021-01831-4. PMID: 34103062. PMCID: PMC8186196.

9. Erdur AC, Rusche D, Scholz D, Kiechle J, Fischer S, Llorián-Salvador Ó, et al. 2014; Deep learning for autosegmentation for radiotherapy treatment planning: state-of-the-art and novel perspectives. Strahlenther Onkol. doi: 10.1007/s00066-024-02262-2. DOI: 10.1007/s00066-024-02262-2. PMID: 39105745.

10. Doolan PJ, Charalambous S, Roussakis Y, Leczynski A, Peratikou M, Benjamin M, et al. 2023; A clinical evaluation of the performance of five commercial artificial intelligence contouring systems for radiotherapy. Front Oncol. 13:1213068. DOI: 10.3389/fonc.2023.1213068. PMID: 37601695. PMCID: PMC10436522.

11. Hoque SMH, Pirrone G, Matrone F, Donofrio A, Fanetti G, Caroli A, et al. 2023; Clinical use of a commercial artificial intelligence-based software for autocontouring in radiation therapy: geometric performance and dosimetric impact. Cancers (Basel). 15:5735. DOI: 10.3390/cancers15245735. PMID: 38136281. PMCID: PMC10741804.

12. Shi F, Hu W, Wu J, Han M, Wang J, Zhang W, et al. 2022; Deep learning empowered volume delineation of whole-body organs-at-risk for accelerated radiotherapy. Nat Commun. 13:6566. DOI: 10.1038/s41467-022-34257-x. PMID: 36323677. PMCID: PMC9630370.

13. Smine Z, Poeta S, De Caluwé A, Desmet A, Garibaldi C, Brou Boni K, et al. 2025; Automated segmentation in planning-CT for breast cancer radiotherapy: a review of recent advances. Radiother Oncol. 202:110615. DOI: 10.1016/j.radonc.2024.110615. PMID: 39489430.

14. Warren S, Richmond N, Wowk A, Wilkinson M, Wright K. 2023; AI segmentation as a quality improvement tool in radiotherapy planning for breast cancer. IPEM Transl. 6-8:100020. DOI: 10.1016/j.ipemt.2023.100020.

15. Kouhen F, Gouach HE, Saidi K, Dahbi Z, Errafiy N, Elmarrachi H, et al. 2024; Synergizing expertise and technology: the artificial intelligence revolution in radiotherapy for personalized and precise cancer treatment. Gulf J Oncolog. 1:94–102. DOI: 10.20944/preprints202308.0992.v1.

16. Ren J, Eriksen JG, Nijkamp J, Korreman SS. 2021; Comparing different CT, PET and MRI multi-modality image combinations for deep learning-based head and neck tumor segmentation. Acta Oncol. 60:1399–1406. DOI: 10.1080/0284186X.2021.1949034. PMID: 34264157.

17. Song J, Zheng J, Li P, Lu X, Zhu G, Shen P. 2021; An effective multimodal image fusion method using MRI and PET for Alzheimer's disease diagnosis. Front Digit Health. 3:637386. DOI: 10.3389/fdgth.2021.637386. PMID: 34713109. PMCID: PMC8521941.

18. Basu S, Singhal S, Singh D. 2024; A systematic literature review on multimodal medical image fusion. Multimed Tools Appl. 83:15845–15913. DOI: 10.1007/s11042-023-15913-w.

19. Yang J, Soltan AAS, Clifton DA. 2022; Machine learning generalizability across healthcare settings: insights from multi-site COVID-19 screening. NPJ Digit Med. 5:69. DOI: 10.1038/s41746-022-00614-9. PMID: 35672368. PMCID: PMC9174159.

20. Tripathi S, Gabriel K, Dheer S, Parajuli A, Augustin AI, Elahi A, et al. 2023; Understanding biases and disparities in radiology AI datasets: a review. J Am Coll Radiol. 20:836–841. DOI: 10.1016/j.jacr.2023.06.015. PMID: 37454752.