• Korean Journal of Radiology
• /
• 제22권6호
• /
• pp.983-993
• /
• 2021

Objective: To investigate the image quality of ultralow-dose CT (ULDCT) of the chest reconstructed using a cycle-consistent generative adversarial network (CycleGAN)-based deep learning method in the evaluation of pulmonary tuberculosis. Materials and Methods: Between June 2019 and November 2019, 103 patients (mean age, 40.8 ± 13.6 years; 61 men and 42 women) with pulmonary tuberculosis were prospectively enrolled to undergo standard-dose CT (120 kVp with automated exposure control), followed immediately by ULDCT (80 kVp and 10 mAs). The images of the two successive scans were used to train the CycleGAN framework for image-to-image translation. The denoising efficacy of the CycleGAN algorithm was compared with that of hybrid and model-based iterative reconstruction. Repeated-measures analysis of variance and Wilcoxon signed-rank test were performed to compare the objective measurements and the subjective image quality scores, respectively. Results: With the optimized CycleGAN denoising model, using the ULDCT images as input, the peak signal-to-noise ratio and structural similarity index improved by 2.0 dB and 0.21, respectively. The CycleGAN-generated denoised ULDCT images typically provided satisfactory image quality for optimal visibility of anatomic structures and pathological findings, with a lower level of image noise (mean ± standard deviation [SD], 19.5 ± 3.0 Hounsfield unit [HU]) than that of the hybrid (66.3 ± 10.5 HU, p < 0.001) and a similar noise level to model-based iterative reconstruction (19.6 ± 2.6 HU, p > 0.908). The CycleGAN-generated images showed the highest contrast-to-noise ratios for the pulmonary lesions, followed by the model-based and hybrid iterative reconstruction. The mean effective radiation dose of ULDCT was 0.12 mSv with a mean 93.9% reduction compared to standard-dose CT. Conclusion: The optimized CycleGAN technique may allow the synthesis of diagnostically acceptable images from ULDCT of the chest for the evaluation of pulmonary tuberculosis.

• 한국방사선학회논문지
• /
• 제16권6호
• /
• pp.709-718
• /
• 2022

본 연구의 목적은 토모테라피 방사선치료에서 고밀도 알루미늄, 티타늄, 강철 금속 삽입물에 대한 단층촬영(CT)을 평가하고 자 하였다. 다양한 밀도의 원통형 막대를 포함한 금속 삽입물과 함께 원통형 토모팬텀을 이용하여 영상을 얻었다. 총 세 가지의 CT 영상에 대해 평균 CT 값(number)와 표준 편차를 구하고, 치료계획 선량평가도 수행하였다. 고밀도 금속 삽입물이 CT값과 변화가 가장 컸다. 타겟에 대한 선량평가(적합성 지수, CI)에서 반복적 금속 인공물 감소 알고리즘(iMAR)이 적용된 영상이 그렇지 않은 영상에 비해 약 20% 좋았으나 유의한 차이는 없었다. iMAR은 표적 및 장기의 묘사에 도움을 주고 토모테라피를 이용한 3차원 입체조형 방사선치료기술(3D-CRT)에서 불확실성을 줄이는 데 도움이 될 것으로 사료된다.

• 대한방사선기술학회지:방사선기술과학
• /
• 제44권6호
• /
• pp.655-661
• /
• 2021

Positron emission tomography (PET) images is affected by acquisition time, short acquisition times results in low gamma counts leading to degradation of image quality by statistical noise. Noise2Void(N2V) is self supervised denoising model that is convolutional neural network (CNN) based deep learning. The purpose of this study is to evaluate denoising performance of N2V for PET image with a short acquisition time. The phantom was scanned as a list mode for 10 min using Biograph mCT40 of PET/CT (Siemens Healthcare, Erlangen, Germany). We compared PET images using NEMA image-quality phantom for standard acquisition time (10 min), short acquisition time (2min) and simulated PET image (S2 min). To evaluate performance of N2V, the peak signal to noise ratio (PSNR), normalized root mean square error (NRMSE), structural similarity index (SSIM) and radio-activity recovery coefficient (RC) were used. The PSNR, NRMSE and SSIM for 2 min and S2 min PET images compared to 10min PET image were 30.983, 33.936, 9.954, 7.609 and 0.916, 0.934 respectively. The RC for spheres with S2 min PET image also met European Association of Nuclear Medicine Research Ltd. (EARL) FDG PET accreditation program. We confirmed generated S2 min PET image from N2V deep learning showed improvement results compared to 2 min PET image and The PET images on visual analysis were also comparable between 10 min and S2 min PET images. In conclusion, noisy PET image by means of short acquisition time using N2V denoising network model can be improved image quality without underestimation of radioactivity.

• 핵의학기술
• /
• 제18권1호
• /
• pp.69-75
• /
• 2014

최근 진단 분야에서 PET/CT는 종양학 분야는 물론 심장, 신경 등 여러 가지 분야에서 널리 활용되고 있다. 그 중 심장 분야에서 $^{13}N-NH_3$ 심근 관류 PET/CT는 MBF를 이용한 절대 심근 관류의 측정이 가능하며 심근 관류 SPECT에 비하여 공간 분해능과 대조도가 우수하고 CT를 이용한 보다 정확한 감쇄 보정이 가능하다는 장점을 가지고 있다. 본 연구에서는 $^{13}N-NH_3$ PET/CT 검사 시 각 재구성 방법에 따른 정량적 심근 관류, 반정량 심근 관류 지표의 변화와 image quality의 변화를 비교하였다. 2013년 1월부터 11월까지 서울대학교병원에서 $^{13}N-NH_3$ PET/CT 검사를 시행받은 14명(평균연령 $60.24{\pm}7.21$세, 평균 몸무게 $71.54{\pm}7.43kg$, 남자 8명, 여자 6명)의 데이터를 분석하였다. 안정기, 부하기 각 10분 listmode scan의 data를 이용 동적, 정적 영상을 FBP, iterative2D, Iterative3D, TrueX의 재구성 방법을 이용하여 재구성하였다. 안정기, 부하기의 각 동적 영상을 이용하여 MBF에서 재구성 방법에 따른 RCA, LAD, LCX의 각 territory 별 심근관류와 global, reserve (stress/rest)값을 비교하였다. 또한 정적 영상을 이용 QPS에서 각 재구성 방법에 따른 extent와 TPD 를 비교하였으며 재구성 방법 별 정적 영상의 snapshoot을 제작하여 영상의 해상력과 노이즈, 판독의 용이성을 기반으로 핵의학과 판독의 5명의 blind test를 시행하였다. 각 재구성 방법에 따라 vendor에서 권고 되는 iterative2D 대비 정량적 심근 관류의 CFR은 최소 -18.68% (P=0.0002)에서 최대 7.91% (P<0.0001)의 변화를 보였으며, 반 정량적 지표들은 안정기에서 extent는 최소 -0.86%p (P=0.1953)에서 최대 5.36%p (P<0.0001), TPD는 최소 -0.57%p (P=0.2053)에서 최대 4.36%p (P<0.0001), 부하기에서 extent는 최소 1.93%p (P=0.4275)에서 최대 5.43%p (P=0.0003), TPD는 최소 1.57%p (P=0.4595)에서 3.93%p (P<0.0001) 증가된 값을 보였다. 영상의 해상력과 노이즈, 판독의 용이성을 기반으로 핵의학과 판독의 5명의 blind test에서는 FBP로 재구성된 영상이 최하위로 평가 되었으며 TrueX, iterative2D, iterative3D 순으로 평가되어 iterative3D로 재구성된 영상이 판독 시 가장 우수한 영상을 평가되었다. 각 병원은 장비에 따른 각 재구성 방법에 의한 정량적 심근 관류의 변화, 반 정략적 지표들의 과대 또는 과소평가되는 정도를 확인하고 병원 장비의 실정에 적합한 동적, 정적 재구성 방법을 확립하여 진단에 보다 유용하고 정확한 심근 관류 값을 제공하여야 할 것이다.

• Journal of Computational Design and Engineering
• /
• 제2권2호
• /
• pp.113-118
• /
• 2015

An important problem in low-dose CT is the image quality degradation caused by photon starvation. There are a lot of algorithms in sinogram domain or image domain to solve this problem. In view of strong self-similarity contained in the special sinusoid-like strip data in the sinogram space, we propose a novel non-local filtering, whose average weights are related to both the image FBP (filtered backprojection) reconstructed from restored sinogram data and the image directly FBP reconstructed from noisy sinogram data. In the process of sinogram restoration, we apply a non-local method with smoothness parameters adjusted adaptively to the variance of noisy sinogram data, which makes the method much effective for noise reduction in sinogram domain. Simulation experiments show that our proposed method by filtering in both image and projection domains has a better performance in noise reduction and details preservation in reconstructed images.

• 대한의용생체공학회:의공학회지
• /
• 제30권6호
• /
• pp.476-481
• /
• 2009

The most troublesome artifacts in micro computed tomography (micro-CT) are ring artifacts. The ring artifacts are caused by non-uniform sensitivity and defective pixels of the x-ray detector. These ring artifacts seriously degrade the quality of CT images. In flat-panel detector based micro-CT systems, the ring artifacts are hardly removed by conventional correction methods of digital radiography, because very small difference of detector pixel signals may make severe ring artifacts. This paper presents a novel method to remove ring artifacts in flat-panel detector based micro-CT systems. First, the bad lines of a sinogram which are caused by defective pixels of the detector are identified, and then, they are corrected using a cubic spline interpolation technique. Finally, a ring artifacts free image is reconstructed from the corrected projections. We applied the method to various kinds of objects and found that the image qualities were much improved.

• 대한영상의학회지
• /
• 제82권6호
• /
• pp.1477-1492
• /
• 2021

목적 Dedicated breast CT (이하 DBCT)는 유방 압박의 고통이 없는 영상 진단 기법으로 최근 주목받고 있다. 본 연구에서는 DBCT 영상에서 약하게 조영증강된 작은 병변의 검출률을 높이기 위해 피사체의 기하학적 정보를 이용하여 최적의 영상 문턱값을 제공하는 adaptive image rescaling (이하 AIR) 기법을 제안하였다. 대상과 방법 5개의 동일 크기의 구멍과 서로 다른 크기의 구멍을 가지는 두 개의 디스크를 각각 제작하고, 이를 60 kVp와 100 kVp로 스캔하여 single-energy CT (이하 SECT), dual-energy CT (이하 DECT), 그리고 AIR 영상을 생성하였다. 전임상 평가를 위해 돼지 조직 영상도 획득하였다. Image contrast (이하 IC)와 contrast-to-noise ratio (이하 CNR)로 화질을 평가하였으며, student's t test를 이용하여 영상 간 화질의 차이를 검증하였다. 결과 AIR의 평균 IC (0.70)는 DECT (0.94)의 74.5%로 나타났으며, SECT (0.22) 보다 318.2% 높았다. 또한 AIR의 평균 CNR (5.08)은 SECT (14.30)의 35.5%로 나타났고 DECT (2.28) 보다 222.8% 높게 측정되었다. 돼지 조직의 전임상 평가 결과도 비슷한 양상을 보였다. 결론 AIR은 SECT보다 높은 영상 대조도를 가지며, 50% 선량만으로도 DECT에 비견할 만한 화질 성능을 제공할 수 있음을 확인하였다. 따라서 AIR은 DBCT 영상에서 약하게 조영증강된 병변의 검출률을 개선할 수 있을 것으로 생각된다.

• 대한방사선기술학회지:방사선기술과학
• /
• 제45권2호
• /
• pp.127-134
• /
• 2022

In order to overcome the image quality limitations of the conventional C-arm, a flat panel detector (FPD) is used to enhance spatial resolution, detective quantum efficiency, frame rate, and dynamic range. Three-dimensional (3D) visualized information can be obtained from C-arm computed tomography (CT) equipped with an FPD, which can reduce patient discomfort and provide various medical information to health care providers by conducting procedures in the interventional procedure room without moving the patient to the CT scan room. Unlike a conventional C-arm device, a C-arm CT requires different basic safety and essential performance evaluation criteria; therefore, in this study, basic safety and essential performance evaluation criteria to protect patients, medical staff, and radiologists were derived based on International Electrotechnical Commission (IEC) standards, the Ministry of Food and Drug Safety (MFDS) standards in Korea, and the rules on the installation and operation of special medical equipment in Korea. As a result of the study, six basic safety evaluation criteria related to electrical and mechanical radiation safety (leakage current, collision protection, emergency stopping device, overheating, recovery management, and ingress of water or particulate matter into medical electrical (ME) equipment and ME systems: footswitches) and 14 essential performance evaluation criteria (accuracy of tube voltage, accuracy of tube current, accuracy of loading time, accuracy of current time product, reproducibility of radiation output, linearity and consistency in radiography, half layer value in X-ray equipment, focal size and collimator, relationship between X-ray field and image reception area, consistency of light irradiation versus X-ray irradiation, performance of the mechanical device, focal spot to skin distance accuracy, image quality evaluation, and technical characteristic of cone-beam computed tomography) were selected for a total of 20 criteria.

• Journal of Veterinary Science
• /
• 제21권4호
• /
• pp.55.1-55.11
• /
• 2020

Background: Computed tomography urography (CTU), based on the excretion of contrast medium after its injection, allows visualization of the renal parenchyma and the renal collecting system. Objectives: To determine the optimal contrast medium dose allocation ratio to apply in split-bolus CTU in dogs. Methods: This prospective, experimental, exploratory study used 8 beagles. In 3-phase CTU, unenhanced-, nephrographic-, and excretory-phase images were obtained with a single injection of 600 mg iodine/kg iohexol. In split-bolus CTU, two different contrast medium allocation ratios (30% and 70% for split CTU 1; 50% and 50% for split CTU 2) were used. Unenhanced phase image and a synchronous nephrographic-excretory phase image were acquired. Results: Although the attenuation of the renal parenchyma was significantly lower when using both split CTUs than the 3-phase CTU, based on qualitative evaluation, the visualization score of the renal parenchyma of split CTU 1 was as high as that of the 3-phase CTU, whereas the split CTU 2 score was significantly lower than those of the two others. Artifacts were not apparent, regardless of CTU protocol. The diameter and opacification of the ureter in both split CTUs were not significantly different from those using 3-phase CTU. Conclusions: Split-bolus CTU with a contrast medium allocation ratio of 30% and 70% is feasible for evaluating the urinary system and allows sufficient enhancement of the renal parenchyma and appropriate distention and opacification of the ureter, with similar image quality to 3-phase CTU in healthy dogs. Split-bolus CTU has the advantages of reducing radiation exposure and the number of CT images needed for interpretation.

• Korean Journal of Radiology
• /
• 제23권7호
• /
• pp.752-762
• /
• 2022

Objective: To compare a deep learning-based reconstruction (DLR) algorithm for pediatric abdominopelvic computed tomography (CT) with filtered back projection (FBP) and iterative reconstruction (IR) algorithms. Materials and Methods: Post-contrast abdominopelvic CT scans obtained from 120 pediatric patients (mean age ± standard deviation, 8.7 ± 5.2 years; 60 males) between May 2020 and October 2020 were evaluated in this retrospective study. Images were reconstructed using FBP, a hybrid IR algorithm (ASiR-V) with blending factors of 50% and 100% (AV50 and AV100, respectively), and a DLR algorithm (TrueFidelity) with three strength levels (low, medium, and high). Noise power spectrum (NPS) and edge rise distance (ERD) were used to evaluate noise characteristics and spatial resolution, respectively. Image noise, edge definition, overall image quality, lesion detectability and conspicuity, and artifacts were qualitatively scored by two pediatric radiologists, and the scores of the two reviewers were averaged. A repeated-measures analysis of variance followed by the Bonferroni post-hoc test was used to compare NPS and ERD among the six reconstruction methods. The Friedman rank sum test followed by the Nemenyi-Wilcoxon-Wilcox all-pairs test was used to compare the results of the qualitative visual analysis among the six reconstruction methods. Results: The NPS noise magnitude of AV100 was significantly lower than that of the DLR, whereas the NPS peak of AV100 was significantly higher than that of the high- and medium-strength DLR (p < 0.001). The NPS average spatial frequencies were higher for DLR than for ASiR-V (p < 0.001). ERD was shorter with DLR than with ASiR-V and FBP (p < 0.001). Qualitative visual analysis revealed better overall image quality with high-strength DLR than with ASiR-V (p < 0.001). Conclusion: For pediatric abdominopelvic CT, the DLR algorithm may provide improved noise characteristics and better spatial resolution than the hybrid IR algorithm.