Improving oncoplastic breast tumor bed localization for radiotherapy planning using image registration algorithms

Abstract

Knowledge about tumor bed localization and its shape analysis is a crucial factor for preventing irradiation of healthy tissues during supportive radiotherapy and as a result, cancer recurrence. The localization process is especially hard for tumors placed nearby soft tissues, which undergo complex, nonrigid deformations. Among them, breast cancer can be considered as the most representative example. A natural approach to improving tumor bed localization is the use of image registration algorithms. However, this involves two unusual aspects which are not common in typical medical image registration: the real deformation field is discontinuous, and there is no direct correspondence between the cancer and its bed in the source and the target 3D images respectively. The tumor no longer exists during radiotherapy planning. Therefore, a traditional evaluation approach based on known, smooth deformations and target registration error are not directly applicable. In this work, we propose alternative artificial deformations which model the tumor bed creation process. We perform a comprehensive evaluation of the most commonly used deformable registration algorithms: B-Splines free form deformations (B-Splines FFD), different variants of the Demons and TV-L optical flow. The evaluation procedure includes quantitative assessment of the dedicated artificial deformations, target registration error calculation, 3D contour propagation and medical experts visual judgment. The results demonstrate that the currently, practically applied image registration (rigid registration and B-Splines FFD) are not able to correctly reconstruct discontinuous deformation fields. We show that the symmetric Demons provide the most accurate soft tissues alignment in terms of the ability to reconstruct the deformation field, target registration error and relative tumor volume change, while B-Splines FFD and TV-L optical flow are not an appropriate choice for the breast tumor bed localization problem, even though the visual alignment seems to be better than for the Demons algorithm. However, no algorithm could recover the deformation field with sufficient accuracy in terms of vector length and rotation angle differences.

Przetwarzanie, analiza i wizualizacja obrazów oraz sygnałów medycznych

“…2021a), wspomaganie diagnostyki bazującej na obrazach mikroskopowych (Wodzinski i Müller 2021, Wodzinski i Skalski 2021a) czy też analiza obrazów mózgu pozyskanych z wykorzystaniem rezonansu magnetycznego (Wodzinski 2021). Dopasowanie obrazów umożliwia ilościowe i jakościowe badania zmian objętości oraz kształtu poszczególnych narządów lub zmian patologicznych (Wodzinski i in. 2018) czy też wspomaganie zabiegów chirurgicznych (Wodzinski i Skalski 2021b).…”

Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization

“…However, the approach using CT is still widely used in clinical practice. Nevertheless, the CT scans are often acquired before the surgery, which creates a chance to apply image registration (IR) algorithms to find a deformation field mapping the scan acquired before the surgery to the scan acquired after the surgery [6]. The deformation field can then transform the segmented tumor in the preoperative image to localize the BTB in the postoperative scan.…”

“…Unfortunately, the BTB localization using IR algorithms is not straightforward, and the traditional algorithms usually fail [6]. The reason is the absence of a breast tumor in the scan acquired after the surgery.…”

“…The Demons algorithm is used as the reference because previous studies have shown the most promising result among the classical registration algorithms [6]. The nonrigid registration based on the instance optimization is implemented using PyTorch, similarly to the affine registration.…”

The application of multiple metrics in deformable image registration for target volume delineation of breast tumor bed

“…For more comprehensive evaluation, Wodzinski et al assessed different state-of -the-art i-DIR methods. 23,24 Furthermore, artificial deformations, which model the tumor bed creation process, were proposed to validate the deformation correctness. 23 However, the studies focused mainly on tumor bed but not boost-CTV.…”