Real-Time Lung Tumor Tracking Using a CUDA Enabled Nonrigid Registration Algorithm for MRI.

OBJECTIVE: This study intends to develop an accurate, real-time tumor tracking algorithm for the automated radiation therapy for cancer treatment using Graphics Processing Unit (GPU) computing. Although a previous moving mesh based tumor tracking approach has been shown to be successful in delineating the tumor regions from a sequence of magnetic resonance image, the algorithm is computationally intensive and its computation time on standard Central Processing Unit (CPU) processors is too slow to be used clinically especially for automated radiation therapy system. METHOD: A re-implementation of the algorithm on a low-cost parallel GPU-based computing platform is utilized to accelerate this computation at a speed that is amicable to clinical usages. Several components in the registration algorithm such as the computation of similarity metric are inherently parallel which fits well with the GPU parallel processing capabilities. Solving a partial differential equation numerically to generate the mesh deformation is one of the computationally intensive components which has been accelerated by utilizing a much faster shared memory on the GPU. RESULTS: Implemented on an NVIDIA Tesla K40c GPU, the proposed approach yielded a computational acceleration improvement of over 5 times its implementation on a CPU. The proposed approach yielded an average Dice score of 0.87 evaluated over 600 images acquired from six patients. CONCLUSION: This study demonstrated that the GPU computing approach can be used to accelerate tumor tracking for automated radiation therapy for mobile lung tumors. Clinical Impact: Accurately tracking mobile tumor boundaries in real-time is important to automate radiation therapy and the proposed study offers an excellent option for fast tumor region tracking for cancer treatment.

ANHIR: Automatic Non-Rigid Histological Image Registration Challenge.

Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds of microscopy histology images in a fair and independent manner. We have assembled 8 datasets, containing 355 images with 18 different stains, resulting in 481 image pairs to be registered. Registration accuracy was evaluated using manually placed landmarks. In total, 256 teams registered for the challenge, 10 submitted the results, and 6 participated in the workshop. Here, we present the results of 7 well-performing methods from the challenge together with 6 well-known existing methods. The best methods used coarse but robust initial alignment, followed by non-rigid registration, used multiresolution, and were carefully tuned for the data at hand. They outperformed off-the-shelf methods, mostly by being more robust. The best methods could successfully register over 98% of all landmarks and their mean landmark registration accuracy (TRE) was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.

A Fully Convolutional Deep Neural Network for Lung Tumor Boundary Tracking in MRI.

Delineation of lung tumor from adjacent tissue from a series of magnetic resonance images (MRI) poses many difficulties due to the image similarities of the region of interest and surrounding area as well as the influence of respiration. However, accurate segmentation of the tumor region is essential in planning a radiation therapy to prevent healthy tissues from receiving excessive radiation. The manual delineation of the entire MRI sequence is tedious, time-consuming and costly. This study investigates how one can perform automatic tracking of tumor boundaries during radiation therapy using convolutional neural networks. We proposed to use a convolutional neural network architecture with modified Dice metric as the cost function. The proposed approach was evaluated over 600 images in comparison to expert manual contours. The proposed method yielded an average Dice score of $0.91 \pm 0.03$ and Hausdorff distance of $2.88 \pm 0.86$ mm. The proposed approach outperformed recent state-of-the-art methods in terms of accuracy in the delineation of the mobile tumors.

Tracking tumor boundary using point correspondence for adaptive radio therapy.

BACKGROUND AND OBJECTIVE: Tracking mobile tumor regions during the treatment is a crucial part of image-guided radiation therapy because of two main reasons which negatively affect the treatment process: (1) a tiny error will lead to some healthy tissues being irradiated; and (2) some cancerous cells may survive if the beam is not accurately positioned as it may not cover the entire cancerous region. However, tracking or delineation of such a tumor region from magnetic resonance imaging (MRI) is challenging due to photometric similarities of the region of interest and surrounding area as well as the influence of motion in the organs. The purpose of this work is to develop an approach to track the center and boundary of tumor region by auto-contouring the region of interest in moving organs for radiotherapy. METHODS: We utilize a nonrigid registration method as well as a publicly available RealTITracker algorithm for MRI to delineate and track tumor regions from a sequence of MRI images. The location and shape of the tumor region in the MRI image sequence varies over time due to breathing. We investigate two approaches: the first one uses manual segmentation of the first frame during the pretreatment stage; and the second one utilizes manual segmentation of all the frames during the pretreatment stage. RESULTS: We evaluated the proposed approaches over a sequence of 600 images acquired from 6 patients. The method that utilizes all the frames in the pretreatment stage with moving mesh based registration yielded the best performance with an average Dice Score of 0.89 ±â€¯0.04 and Hausdorff Distance of 3.38 ±â€¯0.10 mm. CONCLUSIONS: This study demonstrates a promising boundary tracking tool for delineating the tumor region that can deal with respiratory movement and the constraints of adaptive radiation therapy.

Parallel implementation of a nonrigid image registration algorithm for lung tumor boundary tracking in quasi real-time MRI.

This study presents an accelerated implementation of a two-dimensional moving mesh point correspondence algorithm using a GPU for tracking mobile tumor boundaries during radiation therapy. Normal CPU implementation of this algorithm is computationally intensive and time-consuming which limits its clinical utility, hence the need for a faster GPU implementation. One of the computationally intensive parts of the registration algorithm involves numerically solving a partial differential equation. In this paper we demonstrate that the computational performance of the algorithms can be improved by utilizing a shared memory implementation on the GPU. Evaluations in comparison to 600 manually drawn contours showed that the proposed GPU-based tracking of the tumor boundaries yielded similar level of accuracy as the CPU based approach with improved computational efficiency.

Lung tumor boundary tracking in MRI with moving mesh correspondences for adaptive radio therapy.

Delineation of lung tumor regions from magnetic resonance imaging (MRI) poses many difficulties due to MR signal similarities of the region of interest and surrounding area as well as the influence of respiration. However, accurate segmentation of the tumor region is of utmost importance in planning a radiation therapy since a small error can result in some healthy tissues to receive excessive radiation. This study presents a semi-automated method to delineate lung tumor regions from a sequence of MRIs. The proposed method uses a non-rigid image registration framework to propagate the boundaries of the tumor region in MRI acquired during a radiation treatment stage, given manual segmentation on frames acquired during pretreatment stage. We investigate two approaches: 1) the first one utilizes manual segmentation of the first frame during the pretreatment stage; and 2) the second one utilizes manual segmentation of all the frames during the pretreatment stage. We evaluated the proposed approaches over a sequence of 400 images acquired from 4 patients. The proposed method based on the utilization of all the frames yielded a Dice score of 0.90 ± 0.04 and a Hausdorff distance of 1.17 ± 0.35 pixels (2.83 ± 0.79 mm) in comparison to expert manual segmentation.