Search

In all of medical profession a broad field of research is dedicated to seek less invasive and low-risk forms of therapy with the ultimate goal of non-invasive therapy, particularly in neoplasmic diseases. Theranostic platforms, combining diagnostic and therapeutic approaches within one system, have thus garnered interest to augment invasive surgical, chemical, and ionizing interventions. Magnetic particle imaging (MPI) offers, with its versatile tracer material (superparamagnetic iron oxide nanoparticles, SPIOs), a quite recent alternative to established radiation based diagnostic modalities. In addition, MPI lends a bimodal theranostic frame allowing to combine tomographic imaging with therapy techniques using the very same SPIOs. In this work, we show for the first time the interleaved combination of MPI-based imaging, therapy (highly localized magnetic fluid hyperthermia) and therapy safety control (MPI-based thermometry) within one theranostic platform in all three spatial dimensions.

Purpose 3D acquisitions are often acquired to assess the result in orthopedic trauma surgery. With a mobile C-Arm system, these acquisitions can be performed intra-operatively. That reduces the number of required revision surgeries. However, due to the operation room setup, the acquisitions typically cannot be performed such that the acquired volumes are aligned to the anatomical regions. Thus, the multiplanar reconstructed (MPR) planes need to be adjusted manually during the review of the volume. In this paper, we present a detailed study of multi-task learning (MTL) regression networks to estimate the parameters of the MPR planes. Approach First, various mathematical descriptions for rotation, including Euler angle, quaternion, and matrix representation, are revised. Then, three different MTL network architectures based on the PoseNet are compared with a single task learning network. Results Using a matrix description rather than the Euler angle description, the accuracy of the regressed normals improves from $7.7^{\circ}$ to $7.3^{\circ}$ in the mean value for single anatomies. The multi-head approach improves the regression of the plane position from $7.4mm$ to $6.1mm$, while the orientation does not benefit from this approach. Conclusions The results show that a multi-head approach can lead to slightly better results than the individual tasks networks. The most important benefit of the MTL approach is that it is a single network for standard plane regression for all body regions with a reduced number of stored parameters.

Flat panel computed tomography is used intraoperatively to assess the result of surgery. Due to workflow issues, the acquisition typically cannot be carried out in such a way that the axis aligned multiplanar reconstructions (MPR) of the volume match the anatomically aligned MPRs. This needs to be performed manually, adding additional effort during viewing the datasets. A PoseNet convolutional neural network (CNN) is trained such that parameters of anatomically aligned MPR planes are regressed. Different mathematical approaches to describe plane rotation are compared, as well as a cost function is optimized to incorporate orientation constraints. The CNN is evaluated on two anatomical regions. For one of these regions, one plane is not orthogonal to the other two planes. The plane's normal can be estimated with a median accuracy of 5{\deg}, the in-plane rotation with an accuracy of 6{\deg}, and the position with an accuracy of 6 mm. Compared to state-of-the-art algorithms the labeling effort for this method is much lower as no segmentation is required. The computation time during inference is less than 0.05 s., Comment: Accepted for MICCAI 2020

X-ray based measurement and guidance are commonly used tools in orthopaedic surgery to facilitate a minimally invasive workflow. Typically, a surgical planning is first performed using knowledge of bone morphology and anatomical landmarks. Information about bone location then serves as a prior for registration during overlay of the planning on intra-operative X-ray images. Performing these steps manually however is prone to intra-rater/inter-rater variability and increases task complexity for the surgeon. To remedy these issues, we propose an automatic framework for planning and subsequent overlay. We evaluate it on the example of femoral drill site planning for medial patellofemoral ligament reconstruction surgery. A deep multi-task stacked hourglass network is trained on 149 conventional lateral X-ray images to jointly localize two femoral landmarks, to predict a region of interest for the posterior femoral cortex tangent line, and to perform semantic segmentation of the femur, patella, tibia, and fibula with adaptive task complexity weighting. On 38 clinical test images the framework achieves a median localization error of 1.50 mm for the femoral drill site and mean IOU scores of 0.99, 0.97, 0.98, and 0.96 for the femur, patella, tibia, and fibula respectively. The demonstrated approach consistently performs surgical planning at expert-level precision without the need for manual correction., Comment: Accepted for MICCAI 2019