Development of Machine Learning Model for Selecting the 1st Coil in the Treatment of Cerebral Aneurysms by Coil Embolization.

To achieve good treatment outcomes in coil embolization for cerebral aneurysms, it is important to select an appropriate 1st coil for each aneurysm since it serves as a frame to support the subsequent coils to be deployed. However, its selection as appropriate size and length from a wide variety of lineups is not easy, especially for inexperienced neurosurgeons. We developed a machine learning model (MLM) to predict the optimal size and length of the 1st coil by learning information on patients and aneurysms that were previously treated with coil embolization successfully. The accuracy rates of the MLM for the test data were 86.3% and 83.4% in the prediction of size and length, respectively. In addition, the accuracy rates for the 30 cases showed good prediction by the MLM when compared with two different skilled neurosurgeons. Although the accuracy rate of the well-experienced neurosurgeon is similar to MLM, the inexperienced neurosurgeon showed a worse rate and can benefit from the method.Clinical Relevance- The developed MLM has the potential to assist in the selection of the 1st coil for aneurysms. A technically and cost efficient supply chain in the treatment of aneurysms may also be achieved by MLM application.

Ultra-High-Resolution Computed Tomography of the Lung: Image Quality of a Prototype Scanner.

PURPOSE: The image noise and image quality of a prototype ultra-high-resolution computed tomography (U-HRCT) scanner was evaluated and compared with those of conventional high-resolution CT (C-HRCT) scanners. MATERIALS AND METHODS: This study was approved by the institutional review board. A U-HRCT scanner prototype with 0.25 mm x 4 rows and operating at 120 mAs was used. The C-HRCT images were obtained using a 0.5 mm x 16 or 0.5 mm x 64 detector-row CT scanner operating at 150 mAs. Images from both scanners were reconstructed at 0.1-mm intervals; the slice thickness was 0.25 mm for the U-HRCT scanner and 0.5 mm for the C-HRCT scanners. For both scanners, the display field of view was 80 mm. The image noise of each scanner was evaluated using a phantom. U-HRCT and C-HRCT images of 53 images selected from 37 lung nodules were then observed and graded using a 5-point score by 10 board-certified thoracic radiologists. The images were presented to the observers randomly and in a blinded manner. RESULTS: The image noise for U-HRCT (100.87 ± 0.51 Hounsfield units [HU]) was greater than that for C-HRCT (40.41 ± 0.52 HU; P < .0001). The image quality of U-HRCT was graded as superior to that of C-HRCT (P < .0001) for all of the following parameters that were examined: margins of subsolid and solid nodules, edges of solid components and pulmonary vessels in subsolid nodules, air bronchograms, pleural indentations, margins of pulmonary vessels, edges of bronchi, and interlobar fissures. CONCLUSION: Despite a larger image noise, the prototype U-HRCT scanner had a significantly better image quality than the C-HRCT scanners.

Improvement of spatial resolution in the longitudinal direction for isotropic imaging in helical CT.

Experiments were conducted to confirm the isotropic spatial resolution of multislice CT with a 0.5 mm slice thickness. Isotropic spatial resolution means that the spatial resolution in the transaxial plane (X-Y plane) and that in the longitudinal direction (Z direction) are equivalent. To obtain point spread function (PSF) values in the X-Y-Z directions, three-dimensional voxel data were obtained by helical scanning of a bead phantom. The modulation transfer function (MTF) values were then obtained by three-dimensional Fourier transform of the PSF. Evaluation of the spatial resolution in the X-Y-Z directions by the MTF values showed that the spatial resolution in the Z direction does not depend on the reconstruction kernel used. It was also found that the spatial resolution in the Z direction, as compared with that in the X-Y plane, is superior with the standard kernel for the abdomen and is inferior with the high-definition kernel for the ears/bones. By performing sharpening filter processing in the Z direction with a high-definition kernel, comparable spatial resolution could be obtained in the X-Y-Z directions. It was confirmed that adjusting the spatial resolution in the Z direction with the reconstruction kernel used is an effective method for isotropic imaging.

Feldkamp-based cone-beam reconstruction for gantry-tilted helical multislice CT.

Depending on the clinical application, it is frequently necessary to tilt the gantry of an x-ray CT system with respect to the patient and couch. For single-slice fan-beam systems, tilting the gantry introduces no errors or artifacts. Most current systems, however, are helical multislice systems with up to 16 slices. The multislice helical reconstruction algorithms used to create CT images must be modified to account for tilting of the gantry. If they are not, the quality of reconstructed images will be poor with the presence of significant artifacts, such as smearing and double-imaging of anatomical structures. Current CT systems employ three primary types of reconstruction algorithms: helical fan-beam approximation, advanced single-slice rebinning, and Feldkamp-based algorithms. This paper presents a generalized helical cone-beam Feldkamp-based algorithm that is valid for both tilted and nontilted orientations of the gantry. Unlike some of the other algorithms, generalization of the Feldkamp algorithm to include gantry tilt is simple and straightforward with no significant increase in computational complexity. The effect of gantry tilt for helical Feldkamp reconstruction is to introduce a lateral shift in the isocenter of the reconstructed slice of interest, which is a function of the tilt, couch speed, and view angle. The lateral shift is easily calculated and incorporated into the helical Feldkamp backprojection algorithm. A tilt-generalized helical Feldkamp algorithm has been developed and incorporated into Aquilion 16-slice CT (Toshiba, Japan) scanners. This paper describes modifications necessary for the tilt generalization and its verification.