Feasibility of artificial intelligence-driven interfractional monitoring of organ changes by mega-voltage computed tomography in intensity-modulated radiotherapy of prostate cancer

Purpose: High-dose radiotherapy (RT) for localized prostate cancer requires careful consideration of target position changes and adjacent organs-at-risk (OARs), such as the rectum bladder. Therefore, daily monitoring OAR is crucial in minimizing interfractional dosimetric uncertainties. For efficient internal condition patients, we assessed feasibility an auto-segmentation OARs on acquired images, megavoltage computed tomography (MVCT), via a commercial artificial intelligence (AI)-based solution this study. Materials Methods: We collected MVCT images weekly during entire course RT 100 patients treated with helical TomoTherapy system. Based manually contoured body outline, bladder including area, rectal balloon regions trained commercially available fully convolutional (FC)-DenseNet model tested its auto-contouring performance. Results: optimally determined hyperparameters, FC-DenseNet successfully auto-contoured all interest showing high dice similarity coefficient (DSC) over 0.8 small mean surface distance (MSD) within 1.43 mm reference to data. With well-trained AI model, have efficiently monitored patient's through six scans, analyzing DSC, MSD, centroid, volume differences.Conclusion: verified utilizing AI-based low-quality images. In future, will establish fast accurate organ system determining time adaptive replanning.