Keisuke Yamakawa

Two-step iterative reconstruction of region-of-interest with truncated projection in computed tomography

Iteratively reconstructing data only inside the region of interest (ROI) is widely used to acquire CT images in less computation time while maintaining high spatial resolution. A method that subtracts projected data outside the ROI from full-coverage measured data has been proposed. A serious problem with this method is that the accuracy of the measured data confined inside the ROI decreases according to the truncation error outside the ROI. We propose a two-step iterative method that reconstructs image inside the full-coverage in addition to a conventional iterative method inside the ROI to reduce the truncation error inside full-coverage images. Statistical information (e.g., quantum-noise distributions) acquired by detected X-ray photons is generally used in iterative methods as a photon weight to efficiently reduce image noise. Our proposed method applies one of two kinds of weights (photon or constant weights) chosen adaptively by taking into consideration the influence of truncation error. The effectiveness of the proposed method compared with that of the conventional method was evaluated in terms of simulated CT values by using elliptical phantoms and an abdomen phantom. The standard deviation of error and the average absolute error of the proposed method on the profile curve were respectively reduced from 3.4 to 0.4 [HU] and from 2.8 to 0.8 [HU] compared with that of the conventional method. As a result, applying a suitable weight on the basis of a target object made it possible to effectively reduce the errors in CT images.

Background filtering for accuracy improvement in computed tomography with iterative region-of-interest reconstruction

Two methods for preventing the deterioration of the accuracy of iterative region-of-interest (ROI) reconstruction are proposed. Both methods apply filters; the first one applies them to the whole region the outside the region of interest (outside ROI) without distinguishing objects (“method 1”, hereafter); and the second one applies them to only the air and patient-table regions while masking other objects outside the ROI (“method 2”). The effectiveness of both methods was evaluated in terms of simulated CT values by using two different phantoms. Method 1 reduced the artifact intensity level by 86% (at most) compared with that obtained with the conventional method. In the case of an object with high attenuation coefficient, method 2 decreases the level more than method 1. In other words, method 2 improves reconstruction accuracy without causing deterioration by the filters. By selecting either method 1 or 2 in accordance with the attenuation coefficient in regions of objects to be imaged, it is possible to reduce the error level compared with the conventional method.

Development of optimized segmentation map in dual energy computed tomography

Dual energy computed tomography (DECT) has been widely used in clinical practice and has been particularly effective for tissue diagnosis. In DECT the difference of two attenuation coefficients acquired by two kinds of X-ray energy enables tissue segmentation. One problem in conventional DECT is that the segmentation deteriorates in some cases, such as bone removal. This is due to two reasons. Firstly, the segmentation map is optimized without considering the Xray condition (tube voltage and current). If we consider the tube voltage, it is possible to create an optimized map, but unfortunately we cannot consider the tube current. Secondly, the X-ray condition is not optimized. The condition can be set empirically, but this means that the optimized condition is not used correctly. To solve these problems, we have developed methods for optimizing the map (Method-1) and the condition (Method-2). In Method-1, the map is optimized to minimize segmentation errors. The distribution of the attenuation coefficient is modeled by considering the tube current. In Method-2, the optimized condition is decided to minimize segmentation errors depending on tube voltagecurrent combinations while keeping the total exposure constant. We evaluated the effectiveness of Method-1 by performing a phantom experiment under the fixed condition and of Method-2 by performing a phantom experiment under different combinations calculated from the total exposure constant. When Method-1 was followed with Method-2, the segmentation error was reduced from 37.8 to 13.5 %. These results demonstrate that our developed methods can achieve highly accurate segmentation while keeping the total exposure constant.