Deep learning for patient-specific quality assurance of volumetric modulated arc therapy: Prediction accuracy and cost-sensitive classification performance

• Published in: Physica medica 2024-09, Vol.125, p.104500, Article 104500
• Main Authors: Li, Guangjun, Duan, Lian, Xie, Lizhang, Hu, Ting, Wei, Weige, Bai, Long, Xiao, Qing, Liu, Wenjie, Zhang, Lei, Bai, Sen, Yi, Zhang
• Format: Article
• Language: English
• Subjects: Gamma passing rate; MLC; Patient-specific quality assurance; ResNet; Volumetric modulated arc therapy
• ISSN: 1120-1797; 1724-191X; 1724-191X
• DOI: 10.1016/j.ejmp.2024.104500

•Deep learning enhances VMAT PSQA efficiency.•3D-MResNet model shows promising accuracy.•Optimal Th-p improves classification performance.•Low error rates at stringent gamma criteria.•Deep learning reduces VMAT PSQA workloads. To evaluate a deep learning model’s performance in predicting and classifying patient-specific quality assurance (PSQA) results for volumetric modulated arc therapy (VMAT), aiming to streamline PSQA workflows and reduce the onsite measurement workload. A total of 761 VMAT plans were analyzed using 3D-MResNet to process multileaf collimator images and monitor unit data, with the gamma passing rate (GPR) as the output. Thresholds for the predicted GPR (Th-p) and measured GPR (Th-m) were established to aid in PSQA decision-making, using cost curves and error rates to assess classification performance. The mean absolute errors of the model for the test set were 1.63 % and 2.38 % at 3 %/2 mm and 2 %/2 mm, respectively. For the classification of the PSQA results, Th-m was 88.3 % at 2 %/2 mm and 93.3 % at 3 %/2 mm. The lowest cost-sensitive error rates of 0.0127 and 0.0925 were obtained when Th-p was set as 91.2 % at 2 %/2 mm and 96.4 % at 3 %/2 mm, respectively. Additionally, the 2 %/2 mm classifier also achieved a lower total expected cost of 0.069 compared with 0.110 for the 3 %/2 mm classifier. The deep learning classifier under the 2 %/2 mm gamma criterion had a sensitivity and specificity of 100 % (10/10) and 83.5 % (167/200), respectively, for the test set. The developed 3D-MResNet model can accurately predict and classify PSQA results based on VMAT plans. The introduction of a deep learning model into the PSQA workflow has considerable potential for improving the VMAT PSQA process and reducing workloads.