• Annals of Surgical Treatment and Research
• /
• 제95권6호
• /
• pp.297-302
• /
• 2018

Purpose: Increased robotic surgery is attended by increased reports of complications, largely due to limited operative view and lack of tactile sense. These kinds of obstacles, which seldom occur in open surgery, are challenging for beginner surgeons. To enhance robotic surgery safety, we created an augmented reality (AR) model of the organs around the thyroid glands, and tested the AR model applicability in robotic thyroidectomy. Methods: We created AR images of the thyroid gland, common carotid arteries, trachea, and esophagus using preoperative CT images of a thyroid carcinoma patient. For a preliminary test, we overlaid the AR images on a 3-dimensional printed model at five different angles and evaluated its accuracy using Dice similarity coefficient. We then overlaid the AR images on the real-time operative images during robotic thyroidectomy. Results: The Dice similarity coefficients ranged from 0.984 to 0.9908, and the mean of the five different angles was 0.987. During the entire process of robotic thyroidectomy, the AR images were successfully overlaid on the real-time operative images using manual registration. Conclusion: We successfully demonstrated the use of AR on the operative field during robotic thyroidectomy. Although there are currently limitations, the use of AR in robotic surgery will become more practical as the technology advances and may contribute to the enhancement of surgical safety.

• 한국의학물리학회지:의학물리
• /
• 제32권4호
• /
• pp.172-178
• /
• 2021

Purpose: This study aimed to develop a deep learning architecture combining two task models to generate synthetic computed tomography (sCT) images from low-tesla magnetic resonance (MR) images to improve metallic marker visibility. Methods: Twenty-three patients with cervical cancer treated with intracavitary radiotherapy (ICR) were retrospectively enrolled, and images were acquired using both a computed tomography (CT) scanner and a low-tesla MR machine. The CT images were aligned to the corresponding MR images using a deformable registration, and the metallic dummy source markers were delineated using threshold-based segmentation followed by manual modification. The deformed CT (dCT), MR, and segmentation mask pairs were used for training and testing. The sCT generation model has a cascaded three-dimensional (3D) U-Net-based architecture that converts MR images to CT images and segments the metallic marker. The performance of the model was evaluated with intensity-based comparison metrics. Results: The proposed model with segmentation loss outperformed the 3D U-Net in terms of errors between the sCT and dCT. The structural similarity score difference was not significant. Conclusions: Our study shows the two-task-based deep learning models for generating the sCT images using low-tesla MR images for 3D ICR. This approach will be useful to the MR-only workflow in high-dose-rate brachytherapy.

• Korean Journal of Radiology
• /
• 제23권10호
• /
• pp.1009-1018
• /
• 2022

Objective: This study aimed to investigate the feasibility of using artificial intelligence (AI) to identify normal chest radiography (CXR) from the worklist of radiologists in a health-screening environment. Materials and Methods: This retrospective simulation study was conducted using the CXRs of 5887 adults (mean age ± standard deviation, 55.4 ± 11.8 years; male, 4329) from three health screening centers in South Korea using a commercial AI (Lunit INSIGHT CXR3, version 3.5.8.8). Three board-certified thoracic radiologists reviewed CXR images for referable thoracic abnormalities and grouped the images into those with visible referable abnormalities (identified as abnormal by at least one reader) and those with clearly visible referable abnormalities (identified as abnormal by at least two readers). With AI-based simulated exclusion of normal CXR images, the percentages of normal images sorted and abnormal images erroneously removed were analyzed. Additionally, in a random subsample of 480 patients, the ability to identify visible referable abnormalities was compared among AI-unassisted reading (i.e., all images read by human readers without AI), AI-assisted reading (i.e., all images read by human readers with AI assistance as concurrent readers), and reading with AI triage (i.e., human reading of only those rendered abnormal by AI). Results: Of 5887 CXR images, 405 (6.9%) and 227 (3.9%) contained visible and clearly visible abnormalities, respectively. With AI-based triage, 42.9% (2354/5482) of normal CXR images were removed at the cost of erroneous removal of 3.5% (14/405) and 1.8% (4/227) of CXR images with visible and clearly visible abnormalities, respectively. In the diagnostic performance study, AI triage removed 41.6% (188/452) of normal images from the worklist without missing visible abnormalities and increased the specificity for some readers without decreasing sensitivity. Conclusion: This study suggests the feasibility of sorting and removing normal CXRs using AI with a tailored cut-off to increase efficiency and reduce the workload of radiologists.

• Korean Journal of Radiology
• /
• 제22권11호
• /
• pp.1850-1857
• /
• 2021

Objective: The purpose of this study was to assess whether a deep learning (DL) algorithm could enable simultaneous noise reduction and edge sharpening in low-dose lumbar spine CT. Materials and Methods: This retrospective study included 52 patients (26 male and 26 female; median age, 60.5 years) who had undergone CT-guided lumbar bone biopsy between October 2015 and April 2020. Initial 100-mAs survey images and 50-mAs intraprocedural images were reconstructed by filtered back projection. Denoising was performed using a vendor-agnostic DL model (ClariCT.AI^TM, ClariPI) for the 50-mAS images, and the 50-mAs, denoised 50-mAs, and 100-mAs CT images were compared. Noise, signal-to-noise ratio (SNR), and edge rise distance (ERD) for image sharpness were measured. The data were summarized as the mean ± standard deviation for these parameters. Two musculoskeletal radiologists assessed the visibility of the normal anatomical structures. Results: Noise was lower in the denoised 50-mAs images (36.38 ± 7.03 Hounsfield unit [HU]) than the 50-mAs (93.33 ± 25.36 HU) and 100-mAs (63.33 ± 16.09 HU) images (p < 0.001). The SNRs for the images in descending order were as follows: denoised 50-mAs (1.46 ± 0.54), 100-mAs (0.99 ± 0.34), and 50-mAs (0.58 ± 0.18) images (p < 0.001). The denoised 50-mAs images had better edge sharpness than the 100-mAs images at the vertebral body (ERD; 0.94 ± 0.2 mm vs. 1.05 ± 0.24 mm, p = 0.036) and the psoas (ERD; 0.42 ± 0.09 mm vs. 0.50 ± 0.12 mm, p = 0.002). The denoised 50-mAs images significantly improved the visualization of the normal anatomical structures (p < 0.001). Conclusion: DL-based reconstruction may enable simultaneous noise reduction and improvement in image quality with the preservation of edge sharpness on low-dose lumbar spine CT. Investigations on further radiation dose reduction and the clinical applicability of this technique are warranted.

• 핵의학기술
• /
• 제14권2호
• /
• pp.166-171
• /
• 2010

SPECT/CT는 SPECT의 기능학적 영상과 CT의 해부학적인 영상의 융합(Fusion)을 통하여 기존의 SPECT에서 병소의 위치 및 범위를 감별하는데 어려웠던 문제들을 해결하여 진단적 정보에 도움을 줄수 있다. Infinia Hawkeye 4 (GE Healthcare)의 SPECT/CT 감마카메라 영상과 진단용 CT영상을 융합(Fusion)하여 3D (three Dimension)로 Rendering 할 수 있는 Volumetrix Suite의 유용성을 소개하고자 한다. 본원에 SPECT/CT 검사를 하기 위해 내원한 환자 중, 동일한 검사부위의 진단용 CT영상이 있는 환자(Bone, Venography, Parathyroid, WBC)를 대상으로 Volumetrix Suite (Volumetrix IR, Volumetrix 3D)를 적용하였다. Infinia Hawkeye 4의 SPECT영상과 CT영상을 획득한 후 두 영상을 2D (two Dimension)로 융합(Fusion)하였다. Infinia Hawkeye4 SPECT/CT에서의 CT는 해부학적 정보에 한계가 있어, 3D (three Dimension) Rendering을 하기 위해서는 정보량이 많은 진단용 CT영상을 PACS상에서 DICOM (Digital Imaging and Communications in Medicine) File로 전송하여 Infinia Hawkeye4의 Xeleris Workstation에서 IR (Image Registration)한 후 Intergrating 3D (three Dimension)로 융합(Fusion)하여 2D(two Dimension)영상을 3D (three Dimension)영상으로 Rendering하였다. Volumetrix Suite program을 이용함으로써 Infinia Hawkeye 4의 SPECT/CT 영상과 별도로 촬영한 고해상도 진단용 CT영상을 3D Rendering하여 병소의 위치 및 범위를 감별하는데 좀 더 명확한 해부학적 정보를 얻을 수 있었다. 따라서, 해부학적인 정보가 부족한 핵의학영상을 보완함으로서 더 많은 정보를 제공해 핵의학영상 검사의 진단능력을 향상시킬 수 있을 것이라고 기대된다.

• Korean Journal of Radiology
• /
• 제22권8호
• /
• pp.1289-1299
• /
• 2021

Objective: We aimed to evaluate the usefulness of arterial subtraction images for predicting the viability of hepatocellular carcinoma (HCC) after locoregional therapy (LRT) using gadoxetic acid-enhanced MRI and the Liver Imaging Reporting and Data System treatment response (LR-TR) algorithm. Materials and Methods: This study included 90 patients (mean age ± standard deviation, 57 ± 9 years) who underwent liver transplantation or resection after LRT and had 73 viable and 32 nonviable HCCs. All patients underwent gadoxetic acid-enhanced MRI before surgery. Two radiologists assessed the presence of LR-TR features, including arterial phase hyperenhancement (APHE) and LR-TR categories (viable, nonviable, or equivocal), using ordinary arterial-phase and arterial subtraction images. The reference standard for tumor viability was surgical pathology. The sensitivity of APHE for diagnosing viable HCC was compared between ordinary arterial-phase and arterial subtraction images. The sensitivity and specificity of the LR-TR algorithm for diagnosing viable HCC was compared between the use of ordinary arterial-phase and the use of arterial subtraction images. Subgroup analysis was performed on lesions treated with transarterial chemoembolization (TACE) only. Results: The sensitivity of APHE for viable HCCs was higher for arterial subtraction images than ordinary arterial-phase images (71.2% vs. 47.9%; p < 0.001). LR-TR viable category with the use of arterial subtraction images compared with ordinary arterial-phase images showed a significant increase in sensitivity (76.7% [56/73] vs. 63.0% [46/73]; p = 0.002) without significant decrease in specificity (90.6% [29/32] vs. 93.8% [30/32]; p > 0.999). In a subgroup of 63 lesions treated with TACE only, the use of arterial subtraction images showed a significant increase in sensitivity (81.4% [35/43] vs. 67.4% [29/43]; p = 0.031) without significant decrease in specificity (85.0% [17/20] vs. 90.0% [18/20]; p > 0.999). Conclusion: Use of arterial subtraction images compared with ordinary arterial-phase images improved the sensitivity while maintaining specificity for diagnosing viable HCC after LRT using gadoxetic acid-enhanced MRI and the LR-TR algorithm.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1998년도 추계학술대회
• /
• pp.279-280
• /
• 1998

A new algorithm is developed that can reduce the metal artifact on CT caused by pedicle screws. Metal artifact has been recognized as a major problem in precise reading of CT images. In particular, spine surgeons have been bothered with the artifact appearing on CT taken after pedicle screw insertion. To reduce the artifact, our new algorithm first finds the center line from CT images, and then overlays an exact size screw image on the CT. The exact screw is obtained from an actual design specifications of screw, and the CT images are processed to maximize bone margins while minimizing screw images through adjusting the window width and level. 실험 결과 단순한 Window W/L 조절로는 해결되지 않는군요. This algorithm provides spine surgeons with more accurate CT images and thus better interpretation of CT to ascertain the success or failure of pedicle screw insertion.

• Korean Journal of Radiology
• /
• 제2권4호
• /
• pp.192-196
• /
• 2001

Objective: To compare conventional and diffusion-weighted MR imaging in terms of their depiction of the abnormalities occurring in Creutzfeldt-Jakob disease. Materials and Methods: We retrospectively analyzed the findings of conventional (T2-weighted and fluid-attenuated inversion recovery) and diffusion-weighted MR imaging in four patients with biopsy-proven Creutzfeldt-Jakob disease. The signal intensity of the lesion was classified by visual assessment as markedly high, slightly high, or isointense, relative to normal brain parenchyma. Results: Both conventional and diffusion-weighted MR images demonstrated bilateral high signal intensity in the basal ganglia in all four patients. Cortical lesions were observed on diffusion-weighted MR images in all four, and on fluid-attenuated inversion recovery MR images in one, but in no patient on T2-weighted images. Conventional MR images showed slightly high signal intensity in all lesions, while diffusion-weighted images showed markedly high signal intensity in most. Conclusion: Diffusion-weighted MR imaging is more sensitive than its conventional counterpart in the depiction of Creutzfeldt-Jakob disease, and permits better detection of the lesion in both the cerebral cortices and basal ganglia.

• Journal of the Korean Wood Science and Technology
• /
• 제47권3호
• /
• pp.265-276
• /
• 2019

In our previous study, the LeNet3 model successfully classified images from the transverse surfaces of five Korean softwood species (cedar, cypress, Korean pine, Korean red pine, and larch). However, a practical limitation exists in our system stemming from the nature of the training images obtained from the transverse plane of the wood species. In real-world applications, it is necessary to utilize images from the longitudinal surfaces of lumber. Thus, we improved our model by training it with images from the longitudinal and transverse surfaces of lumber. Because the longitudinal surface has complex but less distinguishable features than the transverse surface, the classification performance of the LeNet3 model decreases when we include images from the longitudinal surfaces of the five Korean softwood species. To remedy this situation, we adopt ensemble methods that can enhance the classification performance. Herein, we investigated the use of ensemble models from the LeNet and MiniVGGNet models to automatically classify the transverse and longitudinal surfaces of the five Korean softwoods. Experimentally, the best classification performance was achieved via an ensemble model comprising the LeNet2, LeNet3, and MiniVGGNet4 models trained using input images of $128{\times}128{\times}3pixels$ via the averaging method. The ensemble model showed an F1 score greater than 0.98. The classification performance for the longitudinal surfaces of Korean pine and Korean red pine was significantly improved by the ensemble model compared to individual convolutional neural network models such as LeNet3.

• 한국측량학회지
• /
• 제40권5호
• /
• pp.439-448
• /
• 2022

MV (Moving Vehicle) detection using satellite imagery is important for traffic monitoring and provides a wide range of observations. Specifically, MV detection methods utilizing the time lag in single-pass optical satellite images have been studied for detecting MVs from a single set of images. Because of limitations in detecting MVs outside of roads, most previous studies required road information to limit the moving object to cars on the road. However, it is difficult to obtain road information from inaccessible areas. Therefore, this study proposed a new method for detecting MVs regardless of their locations from single-pass optical satellite images without using additional data. WV-3 (Worldview-3) satellite images were used, and a spatial correlation coefficient map was proposed to detect spatial displacement which denotes MVs across two WV-3 MS images. Finally, evaluation was performed through quantitative metrics and visual inspection. The evaluation results revealed that the proposed method can detect MV movements from the single-pass satellite images. On the contrary, misdetected or undetected MVs due to radiometric differences between the images could be identified by visual inspection. The performance of the proposed method can be improved by minimizing radiometric variations and adding conditions that are robust to radiometric differences between the images.