S2934
RTT- RTT operational practice and workflow innovations
ESTRO 2026
compared to traditional clinician-led methods. Material/Methods: A retrospective analysis was conducted using timestamped workflow data from the ARIA® Oncology Information System over an eleven-month period. Voluming task completion before and after AI introduction was measured. Data were stratified by tumour group (prostate, gynaecological) and case complexity. Entries of 0 minutes and those exceeding 300 minutes were excluded to remove implausible outliers resulting from task interruption and incomplete use of Aria. Clinician planning tasks were reviewed to assess operational impact, with planning workloads of two prostate consultants analysed over a 42-week period to account for leave, clinical duties and plan sign off. Results: AI-assisted workflows demonstrated earlier completion of contouring tasks in prostate cases, with statistically significant reductions in volume completion times (p < 0.05). Effect sizes were modest (Cohen’s d ≈ –0.17 to –0.27), but consistent across high - volume prostate cohorts.Table One Prostate and Nodes Results
>5% CTV underdosing was found in 56% of fractions with the maximum underdose of -17.5% of planned value.
. Conclusion: Automated CBCT-based dose reconstruction enabled routine assessment of delivered dose without disrupting clinical workflow. Automated CBCT-based dose reconstruction reveals clinically significant dosimetric variations invisible to current verification methods. This approach can provide new insight into the clinical relevance of setup and anatomical variations that are currently considered acceptable within tolerance, helping to better understand the true delivered dose to critical structures in breast Darby, Sarah C., et al. "Risk of ischemic heart disease in women after radiotherapy for breast cancer." New England Journal of Medicine 368.11 (2013): 987-998. Keywords: DIBH, Offline ART, Automated CBCT Dose Calculation radiotherapy. References: Radiotherapy: A Retrospective Workflow Analysis Deirdre Dobson 1 , Dr Christopher Thomas 2 , Dr Teresa Guerro Urbano 1 1 Radiotherapy, Guys Cancer, London, United Kingdom. 2 Radiotherapy Physics, Guys Cancer, London, United Kingdom Purpose/Objective: Artificial Intelligence (AI) is increasingly used for auto- contouring in radiotherapy planning. While feasibility and accuracy are established, published evidence of its impact on workflow efficiency in NHS practice remains limited. This study evaluated whether AI- assisted contouring influences the timing of contour completion within the radiotherapy pathway, Digital Poster 3223 Evaluating AI-Assisted Contouring in NHS
Table Two Prostate results
Gynaecological cases showed smaller, non - significant differences, with greater variability. Table Three Gynae results
Table Four Gynae complex volumes results
Across prostate cohorts, cumulative weekly savings equated to ~3 hours of clinician time, reflecting ~9% of their planning workload.Table five Clinician efficiency savings
Conclusion: AI-assisted contouring is associated with earlier
Made with FlippingBook - Share PDF online