2D/3D Deep Registration Along Trajectories With Spatiotemporal Context: Application to Prostate Biopsy Navigation.

OBJECTIVE: The accuracy of biopsy targeting is a major issue for prostate cancer diagnosis and therapy. However, navigation to biopsy targets remains challenging due to the limitations of transrectal ultrasound (TRUS) guidance added to prostate motion issues. This article describes a rigid 2D/3D deep registration method, which provides a continuous tracking of the biopsy location w.r.t the prostate for enhanced navigation. METHODS: A spatiotemporal registration network (SpT-Net) is proposed to localize the live 2D US image relatively to a previously aquired US reference volume. The temporal context relies on prior trajectory information based on previous registration results and probe tracking. Different forms of spatial context were compared through inputs (local, partial or global) or using an additional spatial penalty term. The proposed 3D CNN architecture with all combinations of spatial and temporal context was evaluated in an ablation study. For providing a realistic clinical validation, a cumulative error was computed through series of registrations along trajectories, simulating a complete clinical navigation procedure. We also proposed two dataset generation processes with increasing levels of registration complexity and clinical realism. RESULTS: The experiments show that a model using local spatial information combined with temporal information performs better than more complex spatiotemporal combination. CONCLUSION: The best proposed model demonstrates robust real-time 2D/3D US cumulated registration performance on trajectories. Those results respect clinical requirements, application feasibility, and they outperform similar state-of-the-art methods. SIGNIFICANCE: Our approach seems promising for clinical prostate biopsy navigation assistance or other US image-guided procedure.

External validation of a prostate biopsy simulator.

INTRODUCTION: Simulation-based training has proven to be a promising option allowing for initial and continuous training while limiting the impact of the learning curve on the patient. The Biopsym simulator was developed as a complete teaching environment for the prostate biopsy procedure. This paper presents the results of an external validation of this simulator, involving urology residents recruited during a regional teaching seminar. METHODS: Residents from 4 academic urology departments of the French Auvergne Rhône-Alpes region, who did not take part in the previous simulator validation studies, were enrolled. After a short presentation and standardized initiation session, residents carried out a simulated systematic 12-core biopsy procedure and were asked to fill in a questionnaire collecting their expectations and evaluation of the Biopsym simulator. The number of biopsies reaching each targeted sector, the total score provided by the simulator and the duration of the procedure were recorded. RESULTS: Twenty-three residents were recruited. The overall added value (/100) for learning was rated at a median of 100 (interquartile range 83-100), overall realism of the biopsy procedure at 80 (65-89). The median percentage of biopsies reaching the targeted sector was 66.7% (62-75). The median score provided by the simulator was 50% (37-60). For both, the difference between residents with or without prior biopsy experience was not statistically significant. The median duration of the simulated biopsy procedure was 4:58 (minutes: seconds) (3:49-6:00). Resident with prior experience required less time to complete the biopsy procedure 3:53 (3:39-4:56) vs. 5:10 (4:59-7:10), P=0.01. CONCLUSION: This external validation study confirms a high acceptance of the simulator by the target audience. To our knowledge, the Biopsym simulator is the only prostate biopsy simulator that demonstrated such validity as evaluated by clinicians, outside the center involved in its early development. LEVEL OF EVIDENCE: 3.