1.
Motion-corrected coronary calcium scores by a convolutional neural network: a robotic simulating study.
Zhang, Y, van der Werf, NR, Jiang, B, van Hamersvelt, R, Greuter, MJW, Xie, X
European radiology. 2020;(2):1285-1294
Abstract
OBJECTIVE To classify motion-induced blurred images of calcified coronary plaques so as to correct coronary calcium scores on nontriggered chest CT, using a deep convolutional neural network (CNN) trained by images of motion artifacts. METHODS Three artificial coronary arteries containing nine calcified plaques of different densities (high, medium, and low) and sizes (large, medium, and small) were attached to a moving robotic arm. The artificial arteries moving at 0-90 mm/s were scanned to generate nine categories (each from one calcified plaque) of images with motion artifacts. An inception v3 CNN was fine-tuned and validated. Agatston scores of the predicted classification by CNN were considered as corrected scores. Variation of Agatston scores on moving plaque and by CNN correction was calculated using the scores at rest as reference. RESULTS The overall accuracy of CNN classification was 79.2 ± 6.1% for nine categories. The accuracy was 88.3 ± 4.9%, 75.9 ± 6.4%, and 73.5 ± 5.0% for the high-, medium-, and low-density plaques, respectively. Compared with the Agatston score at rest, the overall median score variation was 37.8% (1st and 3rd quartile, 10.5% and 68.8%) in moving plaques. CNN correction largely decreased the variation to 3.7% (1.9%, 9.1%) (p < 0.001, Mann-Whitney U test) and improved the sensitivity (percentage of non-zero scores among all the scores) from 65 to 85% for detection of coronary calcifications. CONCLUSIONS In this experimental study, CNN showed the ability to classify motion-induced blurred images and correct calcium scores derived from nontriggered chest CT. CNN correction largely reduces the overall Agatston score variation and increases the sensitivity to detect calcifications. KEY POINTS • A deep CNN architecture trained by CT images of motion artifacts showed the ability to correct coronary calcium scores from blurred images. • A correction algorithm based on deep CNN can be used for a tenfold reduction in Agatston score variations from 38 to 3.7% of moving coronary calcified plaques and to improve the sensitivity from 65 to 85% for the detection of calcifications. • This experimental study provides a method to improve its accuracy for coronary calcium scores that is a fundamental step towards a real clinical scenario.
2.
Radiation and contrast agent doses reductions by using 80-kV tube voltage in coronary computed tomographic angiography: a comparative study.
Cao, JX, Wang, YM, Lu, JG, Zhang, Y, Wang, P, Yang, C
European journal of radiology. 2014;(2):309-14
Abstract
OBJECTIVE To investigate the effects of 80-kilovoltage (kV) tube voltage coronary computed tomographic angiography (CCTA) with a reduced amount of contrast agent on qualitative and quantitative image quality parameters and on radiation dose in patients with a body mass index (BMI) <23.0 kg/m(2). METHODS One hundred and twenty consecutive patients with a BMI <23.0 kg/m(2) and a low calcium load undergoing retrospective electrocardiogram (ECG)-gated dual-source CCTA were randomized into two groups [standard-tube voltage (120-kV) vs. low-tube voltage (80-kV)]. The injection flow rate of contrast agent (350 mg I/mL) was adjusted to body weight of each patient (4.5-5.5 mL/s in the 120-kV group and 2.8-3.8 mL/s in the 80-kV group). Radiation and contrast agent doses were evaluated. Quantitative image quality parameters and figure of merit (FOM) of coronary artery were evaluated. Each coronary segment was evaluated for image quality on a 4-point scale. RESULTS Compared with the 120-kV group, effective dose and amount of contrast agent in the 80-kV group were decreased by 57.8% and 30.5% (effective dose:2.7 ± 0.5 vs. 6.4 ± 1.3 mSv; amount of contrast agent:57.1 ± 3.2 vs. 82.1 ± 6.1 mL; both p<0.0001), respectively. Image noise was 22.7 ± 2.1HU for 120-kV images and 33.2 ± 5.2 HU for 80-kV images (p<0.0001). Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in the proximal right coronary artery (RCA) and left main coronary artery (LMA) were all lower in 80-kV than 120-kV images (SNR in the proximal RCA: 16.5 ± 1.8 vs. 19.4 ± 2.8; SNR in the LMA: 16.3 ± 2.0 vs.19.6 ± 2.7; CNR in the proximal RCA: 19.4 ± 2.3 vs.22.9 ± 3.0; CNR in the LMA: 18.8 ± 2.4 vs. 22.7 ± 2.9; all p<0.0001). FOM were all significantly higher in 80-kV than 120-kV images (proximal RCA: 146.7 ± 45.1 vs. 93.4 ± 32.0; LMA: 139.1 ± 47.2 vs. 91.6 ± 31.1; all p<0.0001). There was no significant difference in image quality score between the two groups (3.3 ± 0.8 vs. 3.3 ± 0.8, p=0.068) despite decreased SNR and CNR of coronary artery in the 80-kV group. CONCLUSION The 80-kV protocol significantly reduces radiation and contrast agent doses in CCTA in patients with a low BMI <23.0 kg/m(2) and a low calcium load while maintaining image quality.
3.
Clinical value of [(18)F]FDG-PET/CT in the detection of metastatic medullary thyroid cancer.
Jiang, J, Yang, Z, Zhang, Y, Xu, X, Wang, M, Hu, S, Yao, Z, Pan, H, Zhang, Y, Li, D
Clinical imaging. 2014;(6):797-801
Abstract
PURPOSE To evaluate the value of fluorine-18 2-deoxy-2-d-glucose positron emission tomography/computed tomography ([(18)F]FDG-PET/CT) in the detection of metastatic medullary thyroid cancer. METHODS From November 2006 to November 2012, 50 medullary thyroid cancer patients (median age 48.7 years, range 18-76) who had a total thyroidectomy operation underwent whole-body [(18)F]FDG-PET/CT scans. The diagnostic accuracy of [(18)F]FDG-PET/CT was determined through both lesion-based and patient-based analyses. Further pathological tests were performed on all identified lesions or clinically followed for a minimum period of 6 months. RESULTS One hundred forty-four suspicious lesions were identified by organ-based analysis. Of these lesions, [(18)F]FDG-PET/CT detected 99 true-positive lesions, sensitivity was 73.3%, and specificity was 66.7%. On the patient-based analysis, the overall sensitivity and specificity were calculated as 65.7% and 92.3%, respectively. Using a cutoff calcitonin value of 1000 pg/ml, in patients with calcitonin lower than this value, sensitivity and specificity were 42.9% and 91.0%, respectively. In patients with calcitonin exceeding this value, they raised to 77.3% (χ(2)=4.392, P<.05) and 100% (χ(2)=0.197, P>.05), respectively. Compared with conventional imaging modality, PET/CT scans detected more lesions in 10 patients (20.4%) and correctly changed the treatment in 8 patients (16.3%). CONCLUSION [(18)F]FDG-PET/CT has excellent sensitivity and specificity, especially when the calcitonin value is higher than 1000 pg/ml for detecting metastatic medullary thyroid cancer. Compared to conventional morphologic imaging methods, it provides additional information for diagnosis.