• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.10 no.1
• /
• pp.87-92
• /
• 2016

In this paper, we improved classification performance of benign and malignant lung nodules by including the parenchyma features. For small pulmonary nodules (4-10mm) nodules, there are a limited number of CT data voxels within the solid tumor, making them difficult to process through traditional CAD(computer aided diagnosis) tools. Increasing feature extraction to include the surrounding parenchyma will increase the CT voxel set for analysis in these very small pulmonary nodule cases and likely improve diagnostic performance while keeping the CAD tool flexible to scanner model and parameters. In AdaBoost learning using naive Bayes and SVM weak classifier, a number of significant features were selected from 304 features. The results from the COPDGene test yielded an accuracy, sensitivity and specificity of 100%. Therefore proposed method can be used for the computer aided diagnosis effectively.

• Journal of the Korean Society of Radiology
• /
• v.12 no.1
• /
• pp.85-91
• /
• 2018

Computed tomography, which obtains section images from reconstruction process using projection images, has been applied to various fields. The spatial resolution of the reconstructed image depends on the device used in CT system, the object, and the reconstruction process. In this paper, we investigates the effect of the number of projection images and the pixel size of the detector on the spatial resolution of the reconstructed image under the parallel beam geometry. The reconstruction program was written in Visual C++, and the matrix size of the reconstructed image was $512{\times}512$. The numerical bar phantom was constructed and the Min-Max method was introduced to evaluate the spatial resolution on the reconstructed image. When the number of projections used in reconstruction process was small, artifact like streak appeared and Min-Max was also low. The Min-Max showed upper saturation when the number of projections is increased. If the pixel size of the detector is reduced to 50% of the pixel size of the reconstructed image, the reconstructed image was perfectly recovered as the original phantom and the Min-Max decreased as increasing the detector pixel size. This study will be useful in determining the detector and the accuracy of rotation stage needed to achieve the spatial resolution required in the CT system.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.14 no.2
• /
• pp.21-25
• /
• 2010

Purpose: For better PET imaging with accuracy the transmission scanning is inevitably required for attenuation correction. The attenuation is affected by condition of acquisition and patient position, consequently quantitative accuracy may be decreased in emission scan imaging. In this paper, the present study aims at providing the measurement for attenuation varying with the positions of the patient's arm in whole body PET/CT, further performing the comparative analysis over its SUV changes. Materials and Methods: NEMA 1994 PET phantom was filled with $^{18}F$-FDG and the concentration ratio of insert cylinder and background water fit to 4:1. Phantom images were acquired through emission scanning for 4min after conducting transmission scanning by using CT. In an attempt to acquire image at the state that the arm of the patient was positioned at the lower of ahead, image was acquired in away that two pieces of Teflon inserts were used additionally by fixing phantoms at both sides of phantom. The acquired imaged at a were reconstructed by applying the iterative reconstruction method (iteration: 2, subset: 28) as well as attenuation correction using the CT, and then VOI was drawn on each image plane so as to measure CT number and SUV and comparatively analyze axial uniformity (A.U=Standard deviation/Average SUV) of PET images. Results: It was found from the above phantom test that, when comparing two cases of whether Teflon insert was fixed or removed, the CT number of cylinder increased from -5.76 HU to 0 HU, while SUV decreased from 24.64 to 24.29 and A.U from 0.064 to 0.052. And the CT number of background water was identified to increase from -6.14 HU to -0.43 HU, whereas SUV decreased from 6.3 to 5.6 and A.U also decreased from 0.12 to 0.10. In addition, as for the patient image, CT number was verified to increase from 53.09 HU to 58.31 HU and SUV decreased from 24.96 to 21.81 when the patient's arm was positioned over the head rather than when it was lowered. Conclusion: When arms up protocol was applied, the SUV of phantom and patient image was decreased by 1.4% and 9.2% respectively. With the present study it was concluded that in case of PET/CT scanning against the whole body of a patient the position of patient's arm was not so much significant. Especially, the scanning under the condition that the arm is raised over to the head gives rise to more probability that the patient is likely to move due to long scanning time that causes the increase of uptake of $^{18}F$-FDG of brown fat at the shoulder part together with increased pain imposing to the shoulder and discomfort to a patient. As regarding consideration all of such factors, it could be rationally drawn that PET/CT scanning could be made with the arm of the subject lowered.

• The Journal of Korea Robotics Society
• /
• v.15 no.4
• /
• pp.316-322
• /
• 2020

Endoscopic spine surgery is an advanced surgical technique for spinal surgery since it minimizes skin incision, muscle damage, and blood loss compared to open surgery. It requires, however, accurate positioning of an endoscope to avoid spinal nerves and to locate the endoscope near the target disk. Before the insertion of the endoscope, a guide needle is inserted to guide it. Also, the result of the surgery highly depends on the surgeons' experience and the patients' CT or MRI images. Thus, for the training, a number of haptic simulators for spinal needle insertion have been developed. But, still, it is difficult to be used in the medical field practically because previous studies require manual segmentation of vertebrae from CT images, and interaction force between the needle and soft tissue has not been considered carefully. This paper proposes AI-based automatic vertebrae CT-image segmentation and haptic rendering method using the proposed need-tissue interaction model. For the segmentation, U-net structure was implemented and the accuracy was 93% in pixel and 88% in IoU. The needle-tissue interaction model including puncture force and friction force was implemented for haptic rendering in the proposed spinal needle insertion simulator.

• Journal of radiological science and technology
• /
• v.32 no.1
• /
• pp.69-78
• /
• 2009

The cardiovascular disease has been known as a common cause of death for a long time in the west. The eating habits of Asia, including Korea, have changed recently, so that this disease is also a problem in Asia now. Annual Report on the Cause of Death Statistics from 1996 to 2006 reported that the cardiovascular disease would become the number one cause of death in the next $5{\sim}10$ years. Therefore we realize that more accurate examination is required. The aim of this study was to investigate the accuracy of Calcium-scoring CT and the relationship between risk factor and quantitative scores of Calcium-scoring CT. Through this study we expect that the national public health will be improved. Seventy patients with chest pain were chosen at random. The patients were undergone both coronary CT antigraphy and Calcium - scoring CT at G hospital in Incheon from February 1 to June 30, 2008. The result of the Calcium-scoring CT showed its usefulness for Ischemic cardiovascular disease, with an accuracy similar to that of exercise/pharmacologic stress or ECG when it is difficult for a patient to exercise due to joint problems, aging or for other reasons.

• Korean Journal of Computational Design and Engineering
• /
• v.15 no.4
• /
• pp.297-305
• /
• 2010

Medical image acquisition techniques such as CT and MRI have disadvantages in that the numerous time and efforts are needed. Furthermore, a great amount of radiation exposure is an inherent proberty of the CT imaging technique, a number of side-effects are expected from such method. To improve such conventional methods, a number of novel methods that can obtain 3D medical images from a few X-ray images, such as algebraic reconstruction technique (ART), have been developed. Such methods deform a generic model of the internal body part and fit them into the X-ray images to obtain the 3D model; the initial shape, therefore, affects the entire fitting process in a great deal. From this fact, we propose a novel method that can generate a 3D vertebraic generic model based on the statistical database of CT scans in this study. Moreover, we also discuss a method to generate patient-tailored generic model using the facts obtained from the statistical analysis. To do so, the mesh topologies of CT-scanned 3D vertebra models are modified to be identical to each other, and the database is constructed based on them. Furthermore, from the results of a statistical analysis on the database, the tendency of shape distribution is characterized, and the modeling parameters are extracted. By using these modeling parameters for generating the patient-tailored generic model, the computational speed and accuracy of ART can greatly be improved. Furthermore, although this study only includes an application to the C1 (Atlas) vertebra, the entire framework of our method can be applied to other body parts generally. Therefore, it is expected that the proposed method can benefit the various medical imaging applications.

• Imaging Science in Dentistry
• /
• v.39 no.1
• /
• pp.13-18
• /
• 2009

Purpose: This study was designed to determine a proper threshold value and opacity in three-dimensional CT volume rendering of oral and maxillofacial area. Materials and Methods: Three-dimensional CT data obtained from 50 persons who were done orthognatic surgery in department of oral and maxillofacial radiology of Seoul National University retrospectively. 12 volume rendering post-processing protocols of combination of threshold(100HU, 150HU, 221HU, 270HU) and opacity (58%, 80%, 90%) were applied. Five observers independently evaluated image quality using a five-point range scale. The results were analyzed by receiver operating characteristic curves, ANOVA and Kappa value. And three oromaxillofacial surgeons chose the all images that they thought proper clinically in the all of images. Results: Analysis using ROC curves revealed the area under each curve which indicated a diagnostic accuracy. The highest diagnostic accuracy appear with 100HU and 58% opacity. and the lowest diagnostic accuracy appear with 221HU and 58% opacity that are being used protocol in department of oral and maxillofacial radiology of Seoul National University. But, no statistically significant difference was noted between any of the protocols. And the number of proper images clinically that chosen by three oromaxillofacial surgeons is the largest in the cases of protocol 8 (221HU, opacity 80%) and protocol 11 (270HU, opacity 80%) in one after the other. Conclusion: Threshold and opacity in volume rendering can be controled easily and these can be causes of making an diagnostic accuracy. So we need to select proper values of these factors.

• Journal of Biomedical Engineering Research
• /
• v.43 no.5
• /
• pp.341-352
• /
• 2022

Medical image segmentation is the most important task in radiation therapy. Especially, when segmenting medical images, the liver is one of the most difficult organs to segment because it has various shapes and is close to other organs. Therefore, automatic segmentation of the liver in computed tomography (CT) images is a difficult task. Since tumors also have low contrast in surrounding tissues, and the shape, location, size, and number of tumors vary from patient to patient, accurate tumor segmentation takes a long time. In this study, we propose a method algorithm for automatically segmenting the liver and tumor for this purpose. As an advantage of setting the boundaries of the tumor, the liver and tumor were automatically segmented from the CT image using the 2D CoordConv DeepLab V3+ model using the CoordConv layer. For tumors, only cropped liver images were used to improve accuracy. Additionally, to increase the segmentation accuracy, augmentation, preprocess, loss function, and hyperparameter were used to find optimal values. We compared the CoordConv DeepLab v3+ model using the CoordConv layer and the DeepLab V3+ model without the CoordConv layer to determine whether they affected the segmentation accuracy. The data sets used included 131 hepatic tumor segmentation (LiTS) challenge data sets (100 train sets, 16 validation sets, and 15 test sets). Additional learned data were tested using 15 clinical data from Seoul St. Mary's Hospital. The evaluation was compared with the study results learned with a two-dimensional deep learning-based model. Dice values without the CoordConv layer achieved 0.965 ± 0.01 for liver segmentation and 0.925 ± 0.04 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.927 ± 0.02 for liver division and 0.903 ± 0.05 for tumor division. The dice values using the CoordConv layer achieved 0.989 ± 0.02 for liver segmentation and 0.937 ± 0.07 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.944 ± 0.02 for liver division and 0.916 ± 0.18 for tumor division. The use of CoordConv layers improves the segmentation accuracy. The highest of the most recently published values were 0.960 and 0.749 for liver and tumor division, respectively. However, better performance was achieved with 0.989 and 0.937 results for liver and tumor, which would have been used with the algorithm proposed in this study. The algorithm proposed in this study can play a useful role in treatment planning by improving contouring accuracy and reducing time when segmentation evaluation of liver and tumor is performed. And accurate identification of liver anatomy in medical imaging applications, such as surgical planning, as well as radiotherapy, which can leverage the findings of this study, can help clinical evaluation of the risks and benefits of liver intervention.

• Imaging Science in Dentistry
• /
• v.50 no.4
• /
• pp.291-298
• /
• 2020

Purpose: The detection and exact localization of penetrating foreign bodies are crucial for the appropriate management of patients with dentoalveolar trauma. This study compared the efficacy of cone-beam computed tomography (CBCT) and spiral computed tomography (CT) scans for the detection of different foreign bodies composed of 5 frequently encountered materials in 2 sizes. The effect of the location of the foreign bodies on their visibility was also analyzed. Materials and Methods: In this in vitro study, metal, tooth, stone, glass, and plastic particles measuring 1×1×1 mm and 2×2×2 mm were prepared. They were implanted in a sheep's head in the tongue muscle, nasal cavity, and at the interface of the mandibular cortex and soft tissue. CBCT and spiral CT scans were taken and the visibility of foreign bodies was scored by 4 skilled maxillofacial radiologists who were blinded to the location and number of foreign bodies. Results: CT and CBCT were equally accurate in visualizing metal, stone, and tooth particles of both sizes. However, CBCT was better for detecting glass particles in the periosteum. Although both imaging modalities visualized plastic particles poorly, CT was slightly better for detecting plastic particles, especially the smaller ones. Conclusion: Considering the lower patient radiation dose and cost, CBCT can be used with almost equal accuracy as CT for detecting foreign bodies of different compositions and sizes in multiple maxillofacial regions. However, CT performed better for detecting plastic particles.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.794-798
• /
• 2009

We introduce an implementation of plug-ins on PLUTO. These plug-ins discriminate inflammatory nodules from other types of nodules in chest X-ray CT images. The PLUTO is a common platform for computer-aided diagnosis systems on Microsoft Windows series and it is easy to add new functions as plug-ins. We coded two plug-ins. One of the them calculates features based on medical knowledge. The other plug-in calculates parameters to classify the type of nodules, and it also classifies nodules into inflammatory nodules and others using SVM. These plug-ins are coded using MIST library which is produced at Nagoya University, Japan. In our previous study, the MIST library was parallelized, so that we can utilize a number of CPUs to calculate features and SVM learning/classifying depending on the amount of computation. Using these plug-ins, it became easy to extract features to discriminate inflammatory nodules from other types of nodules and to change parameters for feature extraction and SVM learning/classifying with GUI interface. The accuracy of the classifying result is 100% with 78 solid nodules which contains 43 inflammatory nodules and 35 other type of nodules.