ESTRO 2022 - Abstract Book

S1121

Abstract book

ESTRO 2022

There was significant variability between the three observers’ contours. The majority of all contours that were drawn had a central region where all observers agreed on the presence of tumor, but significant variability in the contours on the edges of the tumor. Manual contours took on average 10 seconds per image, while the deep learning model is able to segment an image in 0.009 ± 0.004 seconds per image on the test set. Preliminary tests confirmed that real-time tumor segmentation from a video feed is possible at this rate. Conclusion Manual tumor segmentation in endoscopy images of rectal cancer patients is prone to significant inter-observer variability. There is uncertainty associated with discerning the edges of a tumor. Deep learning can be used to not only detect regions that are likely to contain tumors but also accurately estimate the regions that are the most likely to cause inter-observer variability. The fact that the model can run in real-time and accurately shows tumor regions with their associated uncertainty makes this method appealing for clinical implementation. In future studies, the model’s generalizability will be investigated by using data from different types of cameras, observers from more institutions, and using more classes.

PO-1326 PET/CT parameters to predict survival and recurrence in patients with locally advanced anal cancer

M. PERAZZI 1 , J. Castelli 1 , R. De Crevoisier 1 , A. Lievre 2 , X. Palard-Novello 3 , A. Devillers 3 , V. Guimas 4 , R. Le Scodan 5 , K. Gnep 1

1 Eugène Marquis Center, Radiotherapy, Rennes, France; 2 Rennes University Hospital, Gastroenterology, Rennes, France; 3 Eugène Marquis Center, Nuclear Medicine, Rennes, France; 4 Institut de Cancérologie de l’Ouest, Radiotherapy, Nantes, France; 5 Saint-Grégoire Private Hospital, Radiotherapy, Saint-Gregoire, France

Purpose or Objective