Virtual magnetic resonance lumbar spine images generated from computed tomography images using conditional generative adversarial networks.

INTRODUCTION: The aim of this study was to generate virtual Magnetic resonance (MR) from computed tomography (CT) using conditional generative adversarial networks (cGAN). METHODS: We selected examinations from 22 adults who obtained their CT and MR lumbar spine examinations. Overall, 4 examinations were used as test data, and 18 examinations were used as training data. A cGAN was trained to generate virtual MR images from the CT images using the corresponding MR images as targets. After training, the generated virtual MR images from test data in epochs 1, 10, 50, 100, 500, and 1000 were compared with the original ones using the mean square error (MSE) and structural similarity index (SSIM). Additionally, two radiologists also performed qualitative assessments. RESULTS: The MSE of the virtual MR images decreased as the epoch of the cGANs increased from the original CT images: 8876.7 ± 1192.9 (original CT), 1567.5 ± 433.9 (Epoch 1), 1242.4 ± 442.0 (Epoch 10), 1065.8 ± 478.1 (Epoch 50), 1276.1 ± 718.9 (Epoch 100), 1046.7 ± 488.2 (Epoch 500), and 1031.7 ± 400.0 (Epoch 1000). No considerable differences were observed in the qualitative evaluation between the virtual MR images and the original ones, except in the structure of the spinal canal. CONCLUSION: Virtual MR lumbar spine images using cGANs could be a feasible technique to generate near-MR images from CT without MR examinations for evaluation of the vertebral body and intervertebral disc. IMPLICATIONS FOR PRACTICE: Virtual MR lumbar spine images using cGANs can offer virtual CT images with sufficient quality for attenuation correction for PET or dose planning in radiotherapy.

Dual-layer detector CT of chest, abdomen, and pelvis with a one-third iodine dose: image quality, radiation dose, and optimal monoenergetic settings.

AIM: To compare the image quality and radiation dose of reduced iodine dose dual-layer detector (DL) computed tomography (CT) with those of a conventional 120 kVp protocol for chest-abdomen-pelvis CT (CAP-CT). MATERIALS AND METHODS: Forty patients with renal dysfunction (estimated glomerular filtrating ratio <45 ml/min/1.73 m2) underwent reduced iodine dose CAP-CT (120 kVp, 200 mg iodine/kg) on DLCT. Virtual monochromatic images (VMI) at 40-70 keV (5 keV interval) were reconstructed retrospectively. Forty matched patients who underwent conventional CAP-CT (120 kVp, 600 mg iodine/kg, iterative reconstruction) were included as controls. The size-specific dose estimate (SSDE), image noise, CT attenuation, and contrast-to-noise ratio (CNR) were compared between the protocols. Two radiologists rated image contrast, image noise, streak artefact, and diagnostic confidence on a five-point scale. RESULTS: The SSDE of the DLCT group was approximately 20% lower than that of the 120 kVp group (15.4±1.9 versus 19.4±2.3 mGy, p<0.01). DLCT-VMI provided almost constant image noise throughout the range of energies (differences of ≤13%), with the noise being equivalent or lower than 120 kVp in the abdomen. CT attenuation and CNR gradually increased as the energy decreased, with values comparable to 120 kVp being attained at around 45-50 keV. Although streak artefact was accentuated at 40-50 keV (p<0.01), the highest scores for diagnostic confidence were assigned at 40 and 45 keV, both of which were equivalent to 120 kVp (p=1.0). CONCLUSION: For CAP-CT with a one-third iodine dose, DLCT-VMI at 40-45 keV allows for a 20% reduction in radiation dose, while preserving image quality comparable to that of conventional 120 kVp protocol.

Dose calculation for asymmetric photon fields with independent jaws and multileaf collimators.

We have developed a simple method for dose calculation in dual asymmetric open and irregular fields with four independent jaws and multileaf collimators. Our calculation method extends the scatter correction method of Kwa et al. [Med. Phys. 21, 1599-1604 (1994)] based on the principle of Day's equivalent-field calculation. The scatter correction factor was determined by the ratio of the derived doses of a smaller asymmetric open field or irregular field to a larger symmetric field. The algorithm with the scatter correction method can be calculated from output factors, tissue maximum ratios, and off-axis ratios for conventional symmetric fields. The doses calculated by this method were compared with the measured doses for various asymmetric open and irregular fields. The agreement between the calculated and measured doses for 4 and 10 MV photon beams was within 0.5% at the geometric center of the asymmetric open fields. For the asymmetric irregular fields with the same geometrical center, agreement within 1% was found in most cases.

[Improvement of detectability of microcalcifications by magnification digital mammography].

Our aim in this study was to evaluate the potential utility of magnification mammography with a CR system by investigating the basic imaging parameters and detectability of microcalcifications in comparison with those of conventional screen-film systems. The basic imaging parameters were evaluated by measuring scatter fraction, modulation transfer function (MTF), Wiener spectrum, and incident dose for the various magnification factors. The detection of simulated microcalcifications in radiographs of a mammographic phantom and breast specimens were evaluated subjectively and quantitatively for screen-film and CR techniques with various magnification factors. The scatter fraction of digital magnification mammography decreased with increasing magnification factor. MTF of magnification digital mammography improved with increasing magnification factor. The detectability of microcalcifications with the CR system was significantly improved by magnification technique. From the above results, it is expected that the use of magnification mammography with a CR system will improve the detectability of microcalcification.

Physical and clinical evaluation of a 2,048 x 2,048-matrix image intensifier TV digital imaging system in bone radiography.

RATIONALE AND OBJECTIVES: To evaluate the potential utility of a 2,048 x 2,048-matrix image intensifier television digital radiography (DR) system versus a conventional screen-film (S-F) system for bone radiography. METHODS: Basic imaging properties were evaluated including resolution properties, Wiener spectra, and detectabilities of low-contrast signals. DR images were obtained with the same exposure (iso-dose) or one-third the exposure (low-dose) used with the S-F system. The visibility of pathologic details of metastatic disease on bone radiographs of 27 patients was evaluated subjectively by six radiologists. RESULTS: Resolution properties of the DR system were slightly superior to those of the S-F system at low-frequency range, but the S-F system showed considerably higher resolution properties at the high-frequency range. The noise levels for iso-dose DR were slightly greater than those for S-F imaging at low spatial frequency; however, low-dose digital radiographs showed higher noise levels. Visibility of details of diagnostic features on bone radiographs was similar with both systems, but the low-dose digital radiographs were slightly inferior. CONCLUSION: High-resolution image intensifier television DR systems may be clinically useful for bone radiography.

Detection of subtle microcalcifications: comparison of computed radiography and screen-film mammography.

Observer performance tests were conducted to compare the effects of image-processing technique (unsharp mask filtering) on the diagnostic accuracy of computed radiography (CR) with storage phosphors in the detection of microcalcifications. Comparison of detectability of microcalcifications with CR and with screen-film mammography was also performed. Clusters of microcalcifications (125-250 microns in diameter) were randomly superimposed on human breast specimens. Observer performance tests were carried out with receiver operating characteristic (ROC) analysis. The area under the ROC curve and the cumulative true-positive-localization fraction were both used as indexes of performance. Observer performance experiments with nine observers indicated that the two types of screen-film images used provided higher detectability than the CR images. The detectability of microcalcifications on the CR images improved slightly with use of the unsharp masking technique. However, no statistically significant difference was found between processed and unprocessed CR images, and detectability still did not reach the level achieved with the screen-film images.