• Proceedings of the KOSOMBE Conference
• /
• v.1989 no.05
• /
• pp.23-26
• /
• 1989

This paper describes the automatic recognition technique for heart enlargement diagnosis with the chest X-ray films enhanced by digital image processing. The technique measures a maximum transverse of the daiphragm and a maximum transverse of the heart by means of direction signature algorithms, then calculates the desired ratio of these two measured values. This is often refered to a cadiothoracic ratio which gives as useful clue for heart enlargement symptom. The experiment results in this paper show presented method is more efficient than the manual diagnosis method of the symptom.

• Journal of the Korean Society of Radiology
• /
• v.17 no.7
• /
• pp.1157-1164
• /
• 2023

This study measures the additional dose for each treatment area using kV X-ray based OBI (On-Board Imager) and CBCT (Cone-Beam CT), which have excellent spatial resolution and contrast, and evaluates the adequacy and stability of radiation management aspects of IGRT. The subjects of the experiment were examined with OBI and CBCT attached to a linear accelerator (Clinac IX), and ring-shaped Halcyon CBCT under imaging conditions for each treatment area, and the dose at the center was measured using an ion chamber. OBI single fraction dose was measured as 0.77 mGy in the head area, 3.04 mGy in the chest area, and 7.19 mGy in the pelvic area. The absorbed doses from the two devices, Clinac IX CBCT and Halcyon CBCT, were measured to be similar in the pelvic area, at 70.04 mGy and 70.45 mGy. and in chest CBCT, the Clinac IX absorbed dose (70.05 mGy) was higher than the Halcyon absorbed dose (21.01 mGy). The absorbed dose to the head area was also higher than that of Clinac IX (9.08 mGy) and Halcyon (5.44 mGy). In kV X-ray-based IGRT, additional radiation exposure due to photoelectric absorption may affect the overall volume of the treatment area, and caution is required.

• Journal of the Korean Society of Radiology
• /
• v.9 no.7
• /
• pp.467-471
• /
• 2015

This study's purpose is improve image quality to keep accurate tube angle in order to recognize distortion degree conditions by patient's position or tube angle and to provide exact clinical informations when taking chest AP projection for patient which have L-tube in stomach. The experimental equipment was ELMO-T6S by SHIMADZU corporation, then we put L-tube which attached 1 mm gap scales ruler on chest phantom surface. The experiment set by 90 kVp, 4 mAs, 120 cm distance. Each phantom position which changed supine, 30degree, 45degree, 60degree on the table exposured direct, ${\pm}5degree$, ${\pm}10degree$, ${\pm}15degree$ to head and feet directions. As a result, L-tube tip's position was changed by patient's position and tube angle. When patient's position is supine, tip's position change was lower than 30degree, 45degree, 60degree. We have to adjust patient's position or tube angle in order to occur image distortion by fault tube angle when confirming correct position L-tube tip through chest x-ray. Also, Radiological technologist try to make accurate evaluation index for satisfied L-tube insertion.

• Journal of Broadcast Engineering
• /
• v.25 no.1
• /
• pp.1-12
• /
• 2020

In the medical field, numerical imbalance of data due to differences in disease prevalence is a common problem. It reduces the performance of a artificial intelligence network, leading to difficulties in learning a network with good performance. Recently, generative adversarial network (GAN) technology has been introduced as a way to address this problem, and its ability has been demonstrated by successful applications in various fields. However, it is still difficult to achieve good results in solving problems with performance degraded by numerical imbalances because the image resolution of the previous studies is not yet good enough and the structure in the image is modeled locally. In this paper, we propose a multi-scale conditional generative adversarial network based on attention mechanism, which can produce high resolution images to solve the numerical imbalance problem of chest X-ray image data. The network was able to produce images for various diseases by controlling condition variables with only one network. It's efficient and effective in that the network don't need to be learned independently for all disease classes and solves the problem of long distance dependency in image generation with self-attention mechanism.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1104-1120
• /
• 2020

The chest X-rays are a common way to diagnose lung cancer or pneumonia. In particular, the finding of a lung nodule is the most important problem in the early detection of lung cancer. Recently, a lot of automatic diagnosis algorithms have been studied to find the lung nodules missed by doctors. The algorithms are typically based on segmentation network like U-Net. However, the occurrence of false positives that similar to lung nodules present outside the lungs can severely degrade performance. In this study, we propose a multi-task learning method that simultaneously learns the lung region and nodule-labeled data based on the prior knowledge that lung nodules exist only in the lung. The proposed method significantly reduces false positives outside the lung and improves the recognition rate of lung nodules to 83.8 F1 score compared to 66.6 F1 score of single task learning with U-net model. The experimental results on the JSRT public dataset demonstrate the effectiveness of the proposed method compared with other baseline methods.

• Journal of radiological science and technology
• /
• v.40 no.1
• /
• pp.1-6
• /
• 2017

Visual evaluation of chest radiograph images is the most practical and effective method. This study compared the Body Mass Index, waist circumference, and mAs with chest radiographs of 351 women. The Bureau of Radiological Health method was used to evaluate the image quality of chest X-ray images by anatomical and physical methods. The average age of the subjects was $30.17{\pm}4.73$ and the average waist circumference was $66.91{\pm}4.67cm$. The mean Body Mass Index value was $20.21{\pm}2.23$, the mean value of mAs was $3.04{\pm}0.78$, and the mean value of Bureau of Radiological Health was $79.83{\pm}8.45$. When the Body Mass Index value increased, waist circumference and mAs mean value increased. The mean value of Body Mass Index was statistically significant(p<0.05) in Group 4 compared to Groups 1 and 2, with increasing Body Mass Index. Exposure control of the automatic exposure control system is considered to be well performed according to body thickness or Body Mass Index at the time of chest radiography. As the Body Mass Index increases, the thickness of the body increases and the breast thickness of the woman also increases. Therefore, it is considered that the exposure amount is changed by the automatic exposure control device to affect the image quality.

• Journal of the Korean Society of Radiology
• /
• v.13 no.1
• /
• pp.111-118
• /
• 2019

Chest digital tomosynthesis has become a practical imaging modality because it can solve the problem of anatomy overlapping in conventional chest radiography. However, because of both limited scan angle and finite-size detector, a portion of chest cannot be represented in some or all of the projection. These bring a discontinuity in intensity across the field of view boundaries in the reconstructed slices, which we refer to as the truncation artifacts. The purpose of this study was to reduce truncation artifacts using a weighted normalization approach and to investigate the performance of this approach for our prototype chest digital tomosynthesis system. The system source-to-image distance was 1100 mm, and the center of rotation of X-ray source was located on 100 mm above the detector surface. After obtaining 41 projection views with ${\pm}20^{\circ}$ degrees, tomosynthesis slices were reconstructed with the filtered back projection algorithm. For quantitative evaluation, peak signal to noise ratio and structure similarity index values were evaluated after reconstructing reference image using simulation, and mean value of specific direction values was evaluated using real data. Simulation results showed that the peak signal to noise ratio and structure similarity index was improved respectively. In the case of the experimental results showed that the effect of artifact in the mean value of specific direction of the reconstructed image was reduced. In conclusion, the weighted normalization method improves the quality of image by reducing truncation artifacts. These results suggested that weighted normalization method could improve the image quality of chest digital tomosynthesis.

• Korean Journal of Digital Imaging in Medicine
• /
• v.13 no.3
• /
• pp.139-144
• /
• 2011

The purpose of this report is recommending a standard indicator which reflects the radiation exposure that is incident on a detector after every exposure event and that reflects the noise levels present in the image data. The experiment was performed with mobile digital X-ray unit and used a acrylic phantom for exposure index measurement. Exposure modality was kVp, mAs, SID. After every exposure, make a data sheet for characteristic curve of detector response. The equipment performed Mobile digital X-ray unit provide the user with values ralated to the incident exposure(air kerma)to the digital detector. They are showed as a logarithmic function shaped. As a result, DEI means a relative measure of exposure to the detector, as compared to the expected exposure for a particular anatomical view. Radiographic technique is the combination of factors used to exposure an anatomical part to produce a high quality radiography and technique charts used most commonly by radiographers to produce consistently exposure level which patient dose can be kept acceptably low.

• Journal of the Korean Society of Radiology
• /
• v.12 no.5
• /
• pp.549-555
• /
• 2018

The medical imaging technique images the contrast formed based on the difference in absorption coefficient of X-rays which changes according to the composition and thickness of the object. At this time, not only primary rays entering the image detector but also scattered rays greatly affect the image quality. Therefore, in this paper, Forward scattering rate and Scattered to primary ratio analysis were performed through Monte Carlo simulation in order to consider influence of scattered ray generated according to object thickness and radiation exposure area change on image quality. In the study, the Forward scattering rate corresponding to the thickness of the object was analyzed at a maximum of 15.3%p and the Scattered to primary ratio was analyzed at 2.00 to 4.54, but it was analyzed as maintaining a constant value for radiation exposure area change. Based on these results, the thickness of the object should be considered as a factor influencing the quality of the image, but radiation exposure area verified that it is a factor that does not affect the image quality. We believe that the results of this research can be utilized as basic information of scattered radiation to improve image quality.

• Journal of radiological science and technology
• /
• v.31 no.2
• /
• pp.115-121
• /
• 2008

Purpose: Aortic Dissection is very dangerous, prognostic disease, which the bloodstream flow out of the true lumen of the aorta by the bursting of aortic intima resulting in a rapid dissociation of inner and outer layer from the media. It is difficult to diagnose aortic dissection clinically by normal X-ray. This study was to investigate the occurrence frequency by age and number of patients who are identified to be aortic dissection by CT (Computed Tomography) scan. Materials and methods: We investigated the trend of yearly fluctuation, gender, age, and department of clinical research of the 112 patients who conducted CT scan in C- University Hospital for two years from January 2005 to December 2006. The MIP and SSD which reconstructed CT image and the VRT image were obtained for the accurate observation. The result was investigated by comparing normal X-ray and CT scan. Results and Conclusion: 1. The yearly check of 112 patients conducted CT scan showed 37 people (41.9%) in 2005, and it was increased to 65 (58.1%) in 2006 by 1.4 times. 2. The gender distribution of patients given a CT scan showed 45 males (40.1%), and female 67 (59.9 %). The aortic dissection patients were 9 (20%) out of 45 males, 21 (31.3%) out of 67 females and women were 1.6 times more than men. Women are also 1.5 times more than men in the number of examinee. 3. The age distribution of patient's who conducted CT scan revealed that there was no patient under 30 years old while 88.3% of all patients were through 41 to 80 years old. The higher the age was, the higher the occurrence of aortic dissection was. The difference in the occurrence frequency of age was statistically significant (p<0.01). 4. The departments that requested CT scan were the emergency department 46 (41.1%), circulatory internal medicine 37 (33.0%), chest surgery 13 (11.6%), and others 6 (14.3%). The combined ratio of emergency medicine and circulatory internal medicine was 74.1% of all. The results show that the aortic dissection is a very dangerous disease whose patients visit mainly via the emergency room. 5. The aortic dissection patients had normal X-ray readings in 22 (73.3%) out of 30, and only 8 (26.7 percent) are abnormal in the X-ray diagnosis. Therefore, the CT scan needs to be enforced in order to assess accurately the disease of aortic dissection.