Initial Clinical Evaluation of Image Fusion Based on Rigid Registration and Supporting Percutaneous Liver Tumor Ablation

Purpose: The purpose of this article is to present and evaluate a proposed rigid registration method in the application of real-time fusion of pre-operative 3DCT images and intraoperative 2DUS images. Methods: A universal hybrid automatic rigid registration method was proposed, which allows registration of preoperative contrast enhanced CT with intraoperative 2D ultrasound images (neither CEUS nor 3DUS is required), with the possibility of manual correction. The method is based on automatically detectable markers, clearly identified in preoperative CT images, and by optical position tracking system during the procedure. A two-step fusion accuracy assessment was used. The first stage, the initial fusion assessment, was carried out at the time of the procedure. The second, objective stage (the final fusion assessment) was carried out after the procedure, and was based on the image and location data collected during the procedure. The clinical evaluation of the method was performed on 20 patients. Results: For IFA and for Fusion Stability Assessment evaluation steps the following results were obtained, respectively: no fusion disorder: 10, good overlay: 8, permanently wrong fusion: 2 and no fusion disorders: 8, short-term fusion disorders: 9, frequent fusion disorders:3. The RMSE descriptive statistics (presenting order: median (first quartile third quartile) [min max]) was 8.87 (6.46 12.76) [5.04 18.84] mm. Conclusion: The results are qualitatively comparable with results obtained in other independent research, quantitatively comparable in accuracy, achieving mostly better results in terms of preparation time consumed and operating in real time, which justifies the possible clinical usage of the proposed method. (c) 2020 Published by Elsevier B.V. on behalf of Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences.