Abstract
Ultrasound (US) imaging is a complex imaging modality, where the tissues are typically characterised by an inhomogeneous image intensity and by a variable image definition at the boundaries that depends on the direction of the incident sound wave. For this reason, conventional image segmentation approaches where the regions of interest are represented by exact masks are inherently inefficient for US images. To solve this issue, we present the first application of a Bayesian convolutional neural network (CNN) based on Monte Carlo dropout on US imaging. This approach is particularly relevant for quantitative applications since differently from traditional CNNs, it enables to infer for each image pixel not only the probability of being part of the target but also the algorithm confidence (i.e. uncertainty) in assigning that probability. In this work, this technique has been applied on US images of the femoral cartilage in the framework of a new application, where high-refresh-rate volumetric US is used for guidance in minimally invasive robotic surgery for the knee. Two options were explored, where the Bayesian CNN was trained with the femoral cartilage contoured either on US, or on magnetic resonance imaging (MRI) and then projected onto the corresponding US volume. To evaluate the segmentation performance, we propose a novel approach where a probabilistic ground-truth annotation was generated combining the femoral cartilage contours from registered US and MRI volumes. Both cases produced a significantly better segmentation performance when compared against traditional CNNs, achieving a dice score coefficient increase of about 6% and 8%, respectively.
Highlights
Ultrasound (US) is broadly used to scan many body regions due to its capability to visualise both bony surfaces and soft tissues
An increase in the Max Dice Score Coefficient with Boundary Uncertainty (DSCBU) was observed for both the US and Magnetic resonance imaging (MRI)-based training, indicating the Bayesian convolutional neural network (CNN) superior ability to identify the femoral cartilage correctly
This implies that the Bayesian CNN allowed identifying pixels belonging to the cartilage, as confirmed through the MRIbased pseudo-ground-truth, that were not included in the US-based ground-truth because the cartilage was not sufficiently defined to be contoured in the respective image region
Summary
Ultrasound (US) is broadly used to scan many body regions (e.g. abdomen, musculoskeletal system) due to its capability to visualise both bony surfaces and soft tissues. We aim at enhancing this tissue representation by training a Bayesian CNN based on Monte Carlo dropout to predict for each image pixel the probability of being part of the target (the femoral cartilage) and the algorithm confidence (i.e. uncertainty) in assigning that probability. This approach was first introduced by [12] and it gained popularity especially in the computer vision field as it can be used for both detection [13], [14] and segmentation tasks [15] on any existing DL algorithm. This approach was able to recognise a larger range of pixels on the US images belonging to the cartilage, but requires that US and MRI datasets are registered
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
