Evaluation of 2D and 3D ultrasound tracking algorithms and impact on ultrasound‐guided liver radiotherapy margins

• DOI: 10.1002/mp.13152
• Publication Date: 2018-08-31T07:23:14Z
• Authors: Valeria De Luca; Jyotirmoy Banerjee; Andre Hallack; Satoshi Kondo; Maxim Makhinya; Daniel Nouri; Lucas Royer; Amalia Cifor; Guillaume Dardenne; Orcun Goksel; Mark J. Gooding; Camiel Klink; Alexandre Krupa; Anthony Le Bras; Maud Marchal; Adriaan Moelker; Wiro J. Niessen; Bartlomiej W. Papiez; Alex Rothberg; Julia Schnabel; Theo van Walsum; Emma Harris; Muyinatu A. Ledij...; Christine Tanner
• Keywords: 3D ultrasound; Match moving; Respiratory compensation

Purpose Compensation for respiratory motion is important during abdominal cancer treatments. In this work we report the results of 2015 MICCAI Challenge on Liver Ultrasound Tracking and extend 2D to relate them clinical relevance in form reducing treatment margins hence sparing healthy tissues, while maintaining full duty cycle. Methods We describe methodologies estimating temporally predicting liver from continuous ultrasound imaging, used ultrasound‐guided radiation therapy. Furthermore, investigated trade‐off between tracking accuracy runtime combination with temporal prediction strategies their impact margins. Results Based sequences 39 volunteers, a mean 0.9 mm was achieved when combining 4 challenge submissions (1.2 3.3 mm). The two 3D 14 volunteers provided accuracies 1.7 1.8 mm. prediction, using faster (41 vs 228 ms) but less accurate (1.4 mm) method resulted substantially reduced (70% 39%) contrast mid‐ventilation margins, as it avoided non‐linear by keeping system latency low (150 400 ms). Acceleration best would improve margin reduction 75%. Conclusions estimation free‐breathing images can reduce in‐plane uncertainty Employing an avoiding be favorable. This approach has potential shorten time compared breath‐hold gated approaches, increase efficiency safety.