Cone-Beam CT Segmentation for Intraoperative Electron Radiotherapy Based on U-Net Variants with Transformer and Extended LSTM Approaches
Authors:
- Sara VocknerSVSara Vockner
- Matthias MattkeMMMatthias Mattke
- Ivan M. MessnerIMIvan M. Messner
- Christoph GaisbergerCGChristoph Gaisberger
- Franz ZehentmayrFranz Zehentmayr
- Klarissa EllmauerKEKlarissa Ellmauer
- Elvis RuznicERElvis Ruznic
- Josef KarnerJKJosef Karner
- Gerd FastnerGFGerd Fastner
- Roland ReitsamerRRRoland Reitsamer
- Falk RoederFalk Roeder
- Markus StanaMSMarkus Stana
Open Access
Abstract
Artificial Intelligence (AI) applications are increasingly prevalent in radiotherapy, including commercial software solutions for automatic segmentation of anatomical structures for 3D Computed Tomography (CT). However, their use in intraoperative electron radiotherapy (IOERT) remains limited. In particular, no AI solution is available for contouring cone beam CT (CBCT) images acquired with a mobile CBCT device. The U-Net convolutional neural network architecture has gained huge success for medical image segmentation but still has difficulties capturing the global context. To increase the accuracy in CBCT segmentation for IOERT, three different AI architectures were trained and evaluated. The features of the natural language processing models Transformer and xLSTM were added to the popular U-Net architecture and compared with the standard U-Net and manual segmentation performance. These networks were trained and tested using 55 CBCT scans obtained from breast cancer patients undergoing IOERT in the department of radiotherapy and radiation oncology in Salzburg, and each architecture’s segmentation performance was assessed using the dice coefficient (DSC) as a similarity measure. The average DSC values were 0.83 for the standard U-Net, 0.88 for the U-Net with transformer features, and 0.66 for the U-Net with xLSTM. The hybrid U-Net architecture, including Transformer features, achieved the best segmentation accuracy, demonstrating an improvement of 5% on average over the standard U-Net, while the U-Net with xLSTM showed inferior performance compared to the standard U-Net. With the help of automatic contouring, synthetic CT images can be generated, and IOERT challenges related to the time-consuming nature of 3D image-based treatment planning can be addressed.Keywords
This publication has 28 references indexed in Scilit:
- IOeRT conventional and FLASH treatment planning system implementation exploiting fast GPU Monte Carlo: The case of breast cancerPhysica Medica, 2024
- Advantages of transformer and its application for medical image segmentation: a surveyBioMedical Engineering OnLine, 2024
- Investigating focal spot position drift in a mobile imaging system equipped with a monobloc‐based x‐ray generatorMedical Physics, 2023
- A fully automatic AI system for tooth and alveolar bone segmentation from cone-beam CT imagesNature Communications, 2022
- Technical evaluation of the cone‐beam computed tomography imaging performance of a novel, mobile, gantry‐based X‐ray system for brachytherapyJournal of Applied Clinical Medical Physics, 2021
- ESTRO IORT Task Force/ACROP recommendations for intraoperative radiation therapy with electrons (IOERT) in breast cancerRadiotherapy and Oncology, 2020
- Intraoperative Tumor Bed Boost With Electrons in Breast Cancer of Clinical Stages I Through III: Updated 10-Year ResultsInternational Journal of Radiation Oncology - Biology - Physics (IJROBP), 2018
- Metrics for evaluating 3D medical image segmentation: analysis, selection, and toolBMC Medical Imaging, 2015
- An Innovative Tool for Intraoperative Electron Beam Radiotherapy Simulation and Planning: Description and Initial Evaluation by Radiation OncologistsInternational Journal of Radiation Oncology - Biology - Physics (IJROBP), 2012
- Long Short-Term MemoryNeural Computation, 1997