Introduction: There is no clear evidence regarding the benefit of restaging for distant metastases after neoadjuvant radiotherapy (RTX) in patients with soft tissue sarcoma (STS) of the extremities... Show moreIntroduction: There is no clear evidence regarding the benefit of restaging for distant metastases after neoadjuvant radiotherapy (RTX) in patients with soft tissue sarcoma (STS) of the extremities and trunk wall. This study aimed to determine how often restaging of the chest identified metastatic disease that altered management in these patients.Methods: We performed a single-centre retrospective study from 2010 to 2020. All patients with non-metastatic STS of the extremities and trunk wall who were treated with neoadjuvant RTX and received a staging and restaging chest CT scan or X-ray for distant metastasis were included. The outcome of interest was change in treatment strategy due to restaging after neoadjuvant RTX.Results: Within the 144 patients who were staged and treated with neoadjuvant RTX, a restaging chest CT or X-ray was performed in 134 patients (93%). A change in treatment strategy due to new findings at restaging after RTX was observed in 26 out of 134 patients (19%). In 24 patients the scheduled resection of the primary STS was cancelled at restaging (24/134, 18%), given the findings at restaging. The other two patients did receive the intended local resection, but either with palliative intent, or as a part of a previously unplanned multimodality treatment.Conclusion: In approximately one in five patients restaging results in a change in treatment strategy. This underlines the added value of routine restaging for distant metastases with chest CT or X-ray after neoadjuvant RTX in patients with STS. (c) 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Show less
Vos, B.D. de; Berendsen, F.F.; Viergever, M.A.; Sokooti, H.; Staring, M.; Isgum, I. 2019
Image registration, the process of aligning two or more images, is the core technique of many (semi-)automatic medical image analysis tasks. Recent studies have shown that deep learning methods,... Show moreImage registration, the process of aligning two or more images, is the core technique of many (semi-)automatic medical image analysis tasks. Recent studies have shown that deep learning methods, notably convolutional neural networks (ConvNets), can be used for image registration. Thus far training of ConvNets for registration was supervised using predefined example registrations. However, obtaining example registrations is not trivial. To circumvent the need for predefined examples, and thereby to increase convenience of training ConvNets for image registration, we propose the Deep Learning Image Registration (DLIR) framework for unsupervised affine and deformable image registration. In the DLIR framework ConvNets are trained for image registration by exploiting image similarity analogous to conventional intensity-based image registration. After a ConvNet has been trained with the DLIR framework, it can be used to register pairs of unseen images in one shot. We propose flexible ConvNets designs for affine image registration and for deformable image registration. By stacking multiple of these ConvNets into a larger architecture, we are able to perform coarse-to-fine image registration. We show for registration of cardiac cine MRI and registration of chest CT that performance of the DLIR framework is comparable to conventional image registration while being several orders of magnitude faster. (C) 2018 Elsevier By. All rights reserved. Show less
