A deep learning-based model for postoperative resection assessment in glioblastoma: A comparative study.

Background Post-operative imaging of glioblastoma presents unique challenges for tumor segmentation due to surgical cavities, hemorrhage, and treatment-related changes. Although multiple open-source artificial intelligence (AI) tools have demonstrated strong performance in pre-operative settings, their utility in post-operative assessment has been insufficiently validated. Thus, we sought to develop an AI-based model for segmentation of postoperative images and to evaluate its performance. Method Our newly developed subtraction-based AI-based segmentation model (Dynapex BT) was tested on the LUMIERE and RHUH-GBM dataset, publicly available cohorts. Its performance was evaluated in three domains: (1) extent of resection (EOR) classification by experts, (2) segmentation performance for residual tumors with non-gross total resection, and (3) correlation between EOR classification and measured residual tumor volume. Performance of our model was compared with that of DeepBraTumIA, a widely used commercially available U-Net-based brain tumor segmentation tool. Results In the LUMIERE cohort, Dynapex BT demonstrated higher GTR classification accuracy compared to DeepBraTumIA (0.80 vs. 0.58). Dynapex BT also achieved significantly better segmentation performance compared to DeepBraTumIA (DSC: 0.815 vs. 0.406, p = 0.002; precision: 0.771 vs. 0.366, p < 0.001; recall: 0.888 vs. 0.583, p = 0.019). Dynapex BT maintained GTR classification accuracy in expert-annotated validation cohort (RHUH-GBM) of 0.8161. Conclusion Our automated postoperative segmentation model outperformed a widely used commercial U-Net-based model not only in segmentation accuracy but also in clinically relevant endpoints. Future studies in larger, multi-institutional cohorts are warranted to evaluate its clinical utility.