Real‐time electromagnetic seed drop detection for permanent implants brachytherapy: Technology overview and performance assessment. Issue 12 (2nd November 2016)

• Record Type: Journal Article
• Authors: Racine, E.; Hautvast, G.; Binnekamp, D.; Beaulieu, L.
• Is Part Of: Medical physics. Volume 43:Issue 12(2016)
• Page Start: 6217
• Page End: 6225
• Publication Date: 2016-11-02
• Subjects: brachytherapy -- dosimetry -- magnetic permeability -- phantoms -- prosthetics; Therapeutic applications, including brachytherapy -- Dose‐volume analysis -- Biomedical engineering; Prostheses implantable into the body -- Radiation therapy -- Scintigraphy; brachytherapy -- permanent implants -- electromagnetic tracking -- interventional guidance -- seed positioning accuracy; Dosimetry -- Brachytherapy -- Ultrasonography -- Coils -- Computer software -- Tracking devices -- Image processing -- Three dimensional sensing -- Computed tomography; Medical physics -- Periodicals; Medical physics; Geneeskunde; Natuurkunde; Toepassingen; Biophysics; Periodicals; Periodicals; Electronic journals; 610.153
• DOI: 10.1118/1.4966135
• Languages: English
• ISSNs: 0094-2405

Abstract

Abstract : Purpose: To describe the principles and report on the performance of a novel real‐time electromagnetic (EM) seed drop detection technology for permanent implants brachytherapy procedures. Methods: A novel EM hollow needle prototype was recently developed by Philips. It possesses standard 3D tracking capability as well as a seed drop detection mechanism, both performed from a single custom built EM sensor. The detection mechanism is based on the magnetic permeability changes in the sensor as the seeds pass through. Drop position estimates are generated by the tracking information at the dropping instants. Three validation experiments were carried out in this study. First, the robustness of the detection mechanism was tested in free air with four different seed types. Detection waveforms were measured and commented. The accuracy of the seed drop position estimates was then evaluated using both 2D and 3D experiments. The procedures consisted of dropping seeds in phantoms, recording the drop position estimates, and finally registering the resulting spatial distributions on reference ones obtained by accurate modalities. Seeds were dropped on a specially designed plastic support adapted to brachytherapy template dimensions for 2D experiments, and true seed positions (reference distribution) were obtained by optical detection. In 3D experiments, seeds were dropped in edible gelatin and reference distributions were obtained by localizing the implants from CT scans of the phantoms. Results: All four seed types were correctly detected by the needle prototype. In total, 250 seeds were dropped on the plastic support, and 96 were dropped in gelatin phantoms. The detection rate was 100% in both cases. The minimum, maximum, and average drop position errors were, respectively, 0.1(+1.6/ − 0.1), 2.9(+1.4/ − 1.5), and 0.9(+1.4/ − 0.7) mm for 2D, and 0.1(+1.0/ − 0.1), 2.1(+1.1/ − 0.8), and 0.6(+1.2/ − 0.5) mm for 3D experiments. Conclusions: The hollow needle prototype combines both EM tracking and automatic seed drop detection in a compact and convenient form. The EM detection mechanism is robust, and the seed drop position estimates appear sufficiently accurate for potential integration of the technology to current brachytherapy treatment planning systems. In that context, it would serve as a valuable tool for rapid dosimetry validation in real‐time treatment delivery.