Gokhan Gunay1, Manh Ha Luu1, Adriaan Moelker2, Theo van Walsum1, Stefan Klein1. 1. Departments of Medical Informatics & Radiology, Biomedical Imaging Group Rotterdam, 3015CN Rotterdam, The Netherlands. 2. Departments of Radiology and Nuclear Medicine, Erasmus University Medical Center Rotterdam, 3015CN Rotterdam, The Netherlands.
Abstract
PURPOSE: In CT-guided liver tumor ablation interventions, registration of a preoperative contrast-enhanced CT image to the intraoperative CT image is hypothesized to improve guidance. This is a highly challenging registration task due to differences in patient poses and large deformations, and therefore high registration errors are expected. In this study, our objective is to develop a method that enables users to locally improve the registration where the registration fails, with minimal user interaction. METHODS: The method is based on a conventional nonrigid intensity-based registration framework, extended with a novel point-to-surface penalty. The point-to-surface penalty serves to improve the alignment of the liver boundary, while requiring minimal user interaction during the intervention: annotating some points on the liver surface at those regions where the conventional registration seems inaccurate. RESULTS: The method is evaluated on 18 clinical datasets. It improves registration accuracy compared with the conventional nonrigid registration in terms of average surface distance (from 2.75 to 2.05 mm) and target registration error (from 6.92 to 5.8 mm). CONCLUSIONS: In this study, we introduce a semiautomated registration algorithm that improves the accuracy of image registration.
PURPOSE: In CT-guided liver tumor ablation interventions, registration of a preoperative contrast-enhanced CT image to the intraoperative CT image is hypothesized to improve guidance. This is a highly challenging registration task due to differences in patient poses and large deformations, and therefore high registration errors are expected. In this study, our objective is to develop a method that enables users to locally improve the registration where the registration fails, with minimal user interaction. METHODS: The method is based on a conventional nonrigid intensity-based registration framework, extended with a novel point-to-surface penalty. The point-to-surface penalty serves to improve the alignment of the liver boundary, while requiring minimal user interaction during the intervention: annotating some points on the liver surface at those regions where the conventional registration seems inaccurate. RESULTS: The method is evaluated on 18 clinical datasets. It improves registration accuracy compared with the conventional nonrigid registration in terms of average surface distance (from 2.75 to 2.05 mm) and target registration error (from 6.92 to 5.8 mm). CONCLUSIONS: In this study, we introduce a semiautomated registration algorithm that improves the accuracy of image registration.
Authors: Anando Sen; Brian M Anderson; Guillaume Cazoulat; Molly M McCulloch; Dalia Elganainy; Brigid A McDonald; Yulun He; Abdallah S R Mohamed; Baher A Elgohari; Mohamed Zaid; Eugene J Koay; Kristy K Brock Journal: Med Phys Date: 2020-02-12 Impact factor: 4.071
Authors: Kyle A Hasenstab; Guilherme Moura Cunha; Atsushi Higaki; Shintaro Ichikawa; Kang Wang; Timo Delgado; Ryan L Brunsing; Alexandra Schlein; Leornado Kayat Bittencourt; Armin Schwartzman; Katie J Fowler; Albert Hsiao; Claude B Sirlin Journal: Eur Radiol Exp Date: 2019-10-26