Identification and characterization of focal ground-glass opacity in the lungs by high-resolution CT using thin-section multidetector helical CT: experimental study using a chest CT phantom

Evaluating medical images using deep convolutional neural networks: A simulated CT phantom image study

BACKGROUND

Applied research on artificial intelligence, mainly in deep learning, is widely performed. If medical images can be evaluated using artificial intelligence, this could substantially improve examination efficiency.

OBJECTIVE

We investigated an evaluation system for medical images with different noise characteristics using a deep convolutional neural network.

METHODS

Simulated computed tomography images are the targets of the system. We used an AlexNet trained with natural images for the deep convolutional neural network and a support vector machine for classification. Synthetic computed tomography images with circular and rectangular signal bodies at different levels of contrast and added Gaussian noise were used for training and testing.

RESULTS

Two transfer learning methods were tested: classification by a re-trained support vector machine using the AlexNet features, and a method that fine-tuned the deep convolutional neural network. Using the first method, all the test image noise levels could be classified correctly. The fine-tuning method achieved an accuracy rate of 92.6%.

CONCLUSIONS

An image quality evaluation method using artificial intelligence will be useful for clinical images and different image quality indices in the future.

Use of an Anthropomorphic Chest Model to Evaluate Multiple Scanning Protocols for High-Definition and Standard-Definition Computed Tomography to Detect Small Pulmonary Nodules

BACKGROUND This study aimed to use the LUNGMAN N1 anthropomorphic chest model to evaluate protocols for high-definition computed tomography (HDCT) and standard-definition CT (SDCT) to detect and compare small pulmonary nodules and determine the most appropriate low-dose scanning protocols. MATERIAL AND METHODS HDCT imaging used the Discovery HD750 scanner (80, 100, 120 and 140 kVp; 360, 320, 280, 240, 200, 160, 120, 80, 40, and 20 mA), and SDCT imaging used the Lightspeed VCT scanner (80, 120, and 140 kVp; 360, 320, 280, 240, 200, 160, 120, 80, 40, and 20 mA). The LUNGMAN N1 anthropomorphic chest model contained artificial pulmonary nodules (diameter: 5, 8, 10, and 12 mm). Low-dose scanning protocols were used in image acquisition. Two experienced radiologists evaluated the image quality. The combinations of voltage, tube current, image noise, and radiation dose were recorded. Consistency of the image quality between raters was assessed by kappa statistical analysis. RESULTS Seventy CT scans of pulmonary nodules (diameter, 5-12 mm) were performed. There was a high degree of consistency for image quality between the two observers (K=0.929 for 5 mm nodules; K=0.819 for overall image quality). For 8 mm nodules, 100% were detected on both SDCT and HDCT. HDCT outperformed SDCT by 5%, in terms of effective dose. There was no significant difference in image quality between the SDCT and HDCT scanners. CONCLUSIONS Using an anthropomorphic chest model, the identification and image quality using SDCT was similar to that of HDCT for small pulmonary nodules between 5-12 mm.

Computer-assisted solid lung nodule 3D volumetry on CT: influence of scan mode and iterative reconstruction: a CT phantom study

OBJECTIVE

To evaluate the effect of high-resolution scan mode and iterative reconstruction on lung nodule 3D volumetry.

METHODS

Solid nodules with various sizes (5, 8, 10 and 12 mm) were placed inside a chest phantom. CT images were obtained with various tube currents, scan modes (conventional mode, high-resolution mode) and iterative reconstructions [0, 50 and 100 % blending of adaptive statistical iterative reconstruction (ASiR) and filtered back projection]. The nodule volumes were calculated using semiautomatic software and compared with the assumed volume from the nodules.

RESULTS

The mean absolute and relative percentage error improved when using iterative reconstruction especially when using the conventional scan mode; however, this effect was not significant. Significant reduction in volume overestimation was observed when using high-resolution scan mode (P = 0.011).

CONCLUSION

The high-resolution mode significantly reduces the volume overestimation of 3D volumetry. Iterative reconstruction shows a reduction in volume overestimation and error margin especially with the conventional scan mode; however, this effect was not significant.

CT volumetry of artificial pulmonary nodules using an ex vivo lung phantom: influence of exposure parameters and iterative reconstruction on reproducibility

OBJECTIVES

To evaluate the influence of exposure parameters and raw-data based iterative reconstruction (IR) on the measurement variability of computer-aided nodule volumetry on chest multidetector computed tomography (MDCT).

MATERIALS AND METHODS

N=7 porcine lung explants were inflated in a dedicated ex vivo phantom and prepared with n=162 artificial nodules. MDCT was performed eight consecutive times (combinations of 120 and 80 kV with 120, 60, 30 and 12 mAs), and reconstructed with filtered back projection (FBP) and IR. Nodule volume and diameter were measured semi-automatically with dedicated software. The absolute percentage measurement error (APE) was computed in relation to the 120 kV 120 mAs acquisition. Noise was recorded for each nodule in every dataset.

RESULTS

Mean nodule volume and diameter were 0.32 ± 0.15 ml and 12.0 ± 2.6mm, respectively. Although IR reduced noise by 24.9% on average compared to FBP (p<0.007), APE with IR was equal to or slightly higher than with FBP. Mean APE for volume increased significantly below a volume computed tomography dose index (CTDI) of 1.0 mGy: for 120 kV 12 mAs APE was 3.8 ± 6.2% (FBP) vs. 4.0 ± 5.2% (IR) (p<0.007); for 80 kV 12 mAs APE was 8.0 ± 13.0% vs. 9.3 ± 15.8% (n.s.), respectively. Correlating APE with image noise revealed that at identical noise APE was higher with IR than with FBP (p<0.05).

CONCLUSIONS

Computer-aided volumetry is robust in a wide range of exposure settings, and reproducibility is reduced at a CTDI below 1.0 mGy only, but the error rate remains clinically irrelevant. Noise reduction by IR is not detrimental for measurement error in the setting of semi-automatic nodule volumetry on chest MDCT.

Determination of a standard deviation that could minimize radiation exposure in an automatic exposure control for pulmonary thin-section computed tomography

PURPOSE

The aim of this study was to determine a standard deviation (SD) that most reduces the radiation dose without sacrificing the diagnostic accuracy of thin-section computed tomography (CT) for clinical use.

MATERIALS AND METHODS

A total of 120 patients were examined by multidetector CT. They were assigned to one of four SD groups: 8, 9, 11, and 12. Each SD group consisted of 30 patients. The CT images of the same patients with SD10 that had formerly been examined were used for comparison. Two radiologists independently evaluated the degrees of image noise and diagnostic acceptability of the pulmonary diseases using a point score grading system. We compared the scores between each SD and the SD10 group.

RESULTS

Generally, image noise was significantly more prominent in the higher-SD groups. The mean score of diagnostic acceptability was significantly lower in the SD12 group (4.2 ± 1.6) than in the SD10 group (4.6 ± 1.1) group (P < 0.001), whereas no difference was present between the SD8 (4.9 ± 0.7), SD9 (4.8 ± 1.0), and SD11 (4.4 ± 1.5) groups and the SD10 group (4.7 ± 1.1, 4.6 ± 1.4, 4.6 ± 1.1, respectively).

CONCLUSION

Thin-section CT with SD12 is not acceptable. SD11 seems to be the setting with the lowest radiation dose while providing acceptable imaging quality for pulmonary thin-section CT.

Volumetric measurement pulmonary ground-glass opacity nodules with multi-detector CT: effect of various tube current on measurement accuracy--a chest CT phantom study

RATIONALE AND OBJECTIVES

The purpose of this study was to evaluate the effect of various tube currents on the accuracy of volumetric measurements of ground-glass opacity (GGO) nodules using a chest phantom.

MATERIALS AND METHODS

A chest phantom containing 13 artificial GGO nodules with known volumes was scanned using a 64-slice computed tomographic scanner at different tube currents (30, 60, 90, 120, 150, 180, and 210 mA). Volumetric measurements were performed using software. The relative percentage error and the absolute percentage error between the volume measures on computed tomography and the reference-standard volumes were calculated. Correlations between the mean absolute percentage error and the mean attenuation of nodules and between the ratio of solid component and the mean attenuation of nodules were analyzed.

RESULTS

The relative percentage errors showed that there was substantial underestimation of nodule volumes at 30, 60, and 90 mA and substantial overestimation of volumes at 120, 150, 180, and 210 mA, but there was no statistically significant difference in absolute percentage errors (P = .876). Pearson's correlation coefficient of the mean absolute percentage errors of nodules on volumetric measurement versus the mean attenuation value of nodules showed a negative correlation, and the ratio of solid component to whole nodule versus the mean attenuation of nodules showed a positive correlation.

CONCLUSION

Volume measurement is a promising method for the quantification of GGO nodule volume. It is important to know that different tube currents can affect the accuracy of volumetric measurements.

Cone-beam technique for 64-MDCT of lung: image quality comparison with stepwise (step-and-shoot) technique

OBJECTIVE

The purpose of this study was to use phantom and patient data acquired with 64-MDCT to compare the image quality and characteristics of helical high-resolution CT images obtained with cone-beam reconstruction with those of stepwise high-resolution CT images obtained with fan-beam reconstruction.

SUBJECTS AND METHODS

We reconstructed helical high-resolution CT images with cone-beam technique and stepwise high-resolution CT images with fan-beam technique. In the phantom study, we measured high-contrast spatial resolution and image noise using a phantom. Streak artifact was evaluated by five radiologists using the phantom. In the clinical phase of the study, two radiologists independently evaluated high-resolution helical and stepwise CT images of the lung fields of 30 patients with diffuse lung disease. Using a 3-point ordinal scale, the radiologists assessed the sharpness of peripheral vessels and interlobular fissures, artifacts, and graininess in the lung fields; overall image quality; and the sharpness of the contour of the left ventricle.

RESULTS

In high-contrast spatial resolution, the contrast curves in each spatial frequency were similar on the helical and stepwise images. In the clinical study, there was no statistically significant difference between helical and stepwise images with respect to sharpness of the contour of the left ventricle, peripheral vessels, or interlobular fissures (p>0.05). With respect to streak artifacts and graininess in the lung fields, helical images received a significantly higher quality grade than did stepwise images (p<0.05).

CONCLUSION

Our phantom and clinical evaluation showed that the quality of high-resolution CT images of the lung obtained with helical scanning was comparable with the quality of stepwise scans.