• Journal of the Korean Society of Radiology
• /
• v.7 no.3
• /
• pp.199-204
• /
• 2013

Despite the suzerain of Korea Taekwondo physical activity and a corresponding lack of basic research on energy consumption and the status of this research is urgent. In this study, the sensor (SenseWear (R) PRO2 Armband), using physical activity were obtained in Taekwondo, and body composition data were obtained by Inbody 520, We were thigh scan using CT scanner and thigh muscle area by CT scan data were acquired. Result of analysis, average thigh muscle area of experimenter 8 people was 132.79 $cm^2$, Of 20 cm above the patella thigh was a 178.79 $cm^2$. Thigh circumference and muscle area showed that the correlation. The average energy consumption per minute was 6.94 calories, and thigh muscle area and average energy consumption per minute also showed that correlation.

• 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 Audiology & Otology
• /
• v.24 no.1
• /
• pp.17-23
• /
• 2020

Background and Objectives: We aimed to measure the head dimensions on computed tomography (CT) images, to compare them to directly measured head dimensions, and to predict a new parameter of bone thickness for aiding bone conduction implant (BCI) placement. Subjects and Methods: We reviewed the facial and mandibular bone CT images of 406 patients. Their head sizes were analyzed using five parameters included in the 6th Size Korea project, and they were divided into age groups (ranging from the 10s to the 80s). We compared the head length, head width, sagittal arc, bitragion arc, and head circumference in the CT and Size Korea groups. We also added the parameter bone thickness for aiding BCI placement. Results: All the head size parameters measured using CT were significantly smaller than those measured directly, with head length showing the smallest difference at 7.85 mm. The differences in the other four parameters between the two groups according to patient age were not statistically significantly different. Bone thickness had the highest value of 4.89±0.93 mm in the 70s and the lowest value of 4.10±0.99 mm in the 10s. Bone thickness also significantly correlated with head width (p=0.038). Conclusions: Our findings suggested that the CT and direct measurements yielded consistent data. Moreover, CT enabled the measurement of bone sizes, including bone thickness, that are impossible to measure directly. CT measurements may complement direct measurements in the Size Korea data when used for developing bone conduction hearing devices (BCIs and headsets) for the Korean population.

• Korean Journal of Audiology
• /
• v.24 no.1
• /
• pp.17-23
• /
• 2020

Background and Objectives: We aimed to measure the head dimensions on computed tomography (CT) images, to compare them to directly measured head dimensions, and to predict a new parameter of bone thickness for aiding bone conduction implant (BCI) placement. Subjects and Methods: We reviewed the facial and mandibular bone CT images of 406 patients. Their head sizes were analyzed using five parameters included in the 6th Size Korea project, and they were divided into age groups (ranging from the 10s to the 80s). We compared the head length, head width, sagittal arc, bitragion arc, and head circumference in the CT and Size Korea groups. We also added the parameter bone thickness for aiding BCI placement. Results: All the head size parameters measured using CT were significantly smaller than those measured directly, with head length showing the smallest difference at 7.85 mm. The differences in the other four parameters between the two groups according to patient age were not statistically significantly different. Bone thickness had the highest value of 4.89±0.93 mm in the 70s and the lowest value of 4.10±0.99 mm in the 10s. Bone thickness also significantly correlated with head width (p=0.038). Conclusions: Our findings suggested that the CT and direct measurements yielded consistent data. Moreover, CT enabled the measurement of bone sizes, including bone thickness, that are impossible to measure directly. CT measurements may complement direct measurements in the Size Korea data when used for developing bone conduction hearing devices (BCIs and headsets) for the Korean population.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.709-715
• /
• 2023

In this study, the performance evaluation of image quality for noise reduction was implemented using the U-net deep learning architecture in computed tomography (CT) images. In order to generate input data, the Gaussian noise was applied to ground truth (GT) data, and datasets were consisted of 8:1:1 ratio of train, validation, and test sets among 1300 CT images. The Adagrad, Adam, and AdamW were used as optimizer function, and 10, 50 and 100 times for number of epochs were applied. In addition, learning rates of 0.01, 0.001, and 0.0001 were applied using the U-net deep learning model to compare the output image quality. To analyze the quantitative values, the peak signal to noise ratio (PSNR) and coefficient of variation (COV) were calculated. Based on the results, deep learning model was useful for noise reduction. We suggested that optimized hyper parameters for noise reduction in CT images were AdamW optimizer function, 100 times number of epochs and 0.0001 learning rates.

• Journal of the Korean Society of Radiology
• /
• v.13 no.7
• /
• pp.933-939
• /
• 2019

The purpose of this study is to provide basic clinical data by evaluating images, measuring absorbed dose and effective dose by using high resolution CT and low dose CT by using anthropomorphic chest phantom and glass dosimeter. Tissue dose was measured by inserting a glass dosimeter into the anthropomorphic chest phantom. A 64-slice CT system (SOMATOM Sensation 64, Siemens AG, Forchheim, Germany) and CARE Dose 4D were used, and the parameters of the high resolution CT were 120 kVp, Eff. Scan parameters of mAs 104, scan time 7.93 s, slice 1.0 mm (Acq. 64 × 0.6 mm), convolution kernel (B60f sharp) were used, and low dose CT was 120 kVp, Eff. mAs 15, scan time 7.41 s, slice 3.0 mm (Acq. 64 × 0.6 mm), scan of convolution kernel B50f medium sharp. CTDIvol was measured at 8.01 mGy for high resolution CT and 1.18 mGy for low dose CT. Low dose CT scans showed 85.49% less absorbed dose than high resolution CT scans.

• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.85-92
• /
• 2023

Deep learning applications are increasingly being leveraged for disease detection tasks in medical imaging modalities such as X-ray, Computed Tomography (CT), and Magnetic Resonance Imaging (MRI). Most data-centric deep learning challenges necessitate the use of supervised learning methodologies to attain high accuracy and to facilitate performance evaluation through comparison with the ground truth. Supervised learning mandates a substantial amount of image and label sets, however, procuring an adequate volume of medical imaging data for training is a formidable task. Various data augmentation strategies can mitigate the underfitting issue inherent in supervised learning-based models that are trained on limited medical image and label sets. This research investigates the enhancement of a deep learning-based rib fracture segmentation model and the efficacy of data augmentation techniques such as left-right flipping, rotation, and scaling. Augmented dataset with L/R flipping and rotations(30°, 60°) increased model performance, however, dataset with rotation(90°) and ⨯0.5 rescaling decreased model performance. This indicates the usage of appropriate data augmentation methods depending on datasets and tasks.

• Journal of Trauma and Injury
• /
• v.28 no.3
• /
• pp.149-157
• /
• 2015

Purpose: Traumatic brain injury (TBI) is the most common cause of pediatric trauma patients came to the emergency department. Without guidelines, many of these children underwent repeat brain computed tomography (CT). The purpose of this study was to evaluate the value of repeat brain CT in children with TBI. Methods: We conducted a retrospective study of TBI in children younger than 19 years of age who visited the emergency department (ED) from January 2011 to December 2012. According to the Glasgow Coma Scale (GCS) and Pediatric Glasgow Coma Scale score of the patients, study population divided in three groups. Clinical data collected included age, mechanism of injury, type of TBI, and outcome. Results: A Total 83 children with TBI received repeat brain CT. There were no need for neurosurgical intervention in mild TBI (GCS score 13-15) group who underwent routine repeat CT. 4 patients of mild TBI group, received repeat brain CT due to neurological deterioration, and one patient underwent neurosurgical intervention. Routine repeat CT identified 12 patients with radiographic progression. One patient underwent neurosurgical intervention based on the second brain CT finding, who belonged to the moderate TBI (GCS score 9-12) group. Conclusion: Our study showed that children with mild TBI can be observed without repeat brain CT when there is no evidence of neurologic deterioration. Further study is needed for establish indication for repetition of CT scan in order to avoid unnecessary radiation exposure of children.

• Journal of Chest Surgery
• /
• v.33 no.10
• /
• pp.798-805
• /
• 2000

Background : the prediction on changes in the lung function after lung surgery would be an important indicator in terms of the operability and postoperative complications. In order to predict the postoperative FEV1 - the commonly used method for measuring changes in lung function- a comparison between the quantitative CT and the perfusion lung scan was made and proved its usefulness. Material and Method : The subjects included 22 patients who received perfusion lung scan and quantitative CT preoperatively and with whom the follow-up of PFT were possibles out of the pool of patients who underwent right lobectomy or right pneumonectomy between June of 1997 and December of 1999. The FEV1 and FVC were calibrated by performing the PFT on each patient and then the predicted FEV1 and FVC were calculated after performing perfusion lung scan and quantitative CT postoperatively. The FEV1 and FVC were calibrated by performing the PFT after 1 week and after 3 momths following the surgery. Results : There was a significant mutual scan and the actual postoperative FEV1 and FVC at 1 week and 3 months. The predicted FEV1 and FVC(pneumonectomy group : r=0.962 and r=0.938 lobectomy group ; r=0.921 and r=913) using quantitative CT at 1 week postoperatively showed a higher mutual relationship than that predicted by perfusion lung scan(pneumonectomy group : r=0.927 and r=0.890 lobectomy group : r=0.910 and r=0.905) The result was likewise at 3 months postoperatively(CT -pneumonectomy group : r=0.799 and r=0.882 lobectomy group : r=0.934 and r=0.932) Conclusion ; In comparison to perfusion lung scan quantitative CT is more accurate in predicting lung function postoperatively and is cost-effective as well. Therefore it can be concluded that the quantitative CT is an effective method of replacing the perfusion lung scan in predicting lung function post-operatively. However it is noted that further comparative analysis using more data and follow-up studies of the patients is required.

• Radiation Oncology Journal
• /
• v.4 no.2
• /
• pp.179-181
• /
• 1986

Computed tomography (CT) adds a new dimension in the study of body contour, organs, and tissues as well as various pathologic conditions. This modality provides a great degree of accuracy in radiation therapy Planning (RTP). However, CT images are usually taken on a small reduced format so that possible errors can be made during inputting the CT data into an automatic planner. Authors have designed a simple inexpensive magnifying device of CT images to obviate errors created by reduced image.