• Journal of the Korean Society of Radiology
• /
• v.14 no.6
• /
• pp.773-780
• /
• 2020

With the development of the 4th industrial, research is being conducted to prevent diseases and reduce damage in various fields of science and technology such as medicine, health, and bio. As a result, artificial intelligence technology has been introduced and researched for image analysis of radiological examinations. In this paper, we will directly apply a deep learning model for classification and detection of pneumonia using chest X-ray images, and evaluate whether the deep learning model of the Inception series is a useful model for detecting pneumonia. As the experimental material, a chest X-ray image data set provided and shared free of charge by Kaggle was used, and out of the total 3,470 chest X-ray image data, it was classified into 1,870 training data sets, 1,100 validation data sets, and 500 test data sets. I did. As a result of the experiment, the result of metric evaluation of the Inception V3 deep learning model was 94.80% for accuracy, 97.24% for precision, 94.00% for recall, and 95.59 for F1 score. In addition, the accuracy of the final epoch for Inception V3 deep learning modeling was 94.91% for learning modeling and 89.68% for verification modeling for pneumonia detection and classification of chest X-ray images. For the evaluation of the loss function value, the learning modeling was 1.127% and the validation modeling was 4.603%. As a result, it was evaluated that the Inception V3 deep learning model is a very excellent deep learning model in extracting and classifying features of chest image data, and its learning state is also very good. As a result of matrix accuracy evaluation for test modeling, the accuracy of 96% for normal chest X-ray image data and 97% for pneumonia chest X-ray image data was proven. The deep learning model of the Inception series is considered to be a useful deep learning model for classification of chest diseases, and it is expected that it can also play an auxiliary role of human resources, so it is considered that it will be a solution to the problem of insufficient medical personnel. In the future, this study is expected to be presented as basic data for similar studies in the case of similar studies on the diagnosis of pneumonia using deep learning.

• Journal of the Korean Society of Radiology
• /
• v.82 no.6
• /
• pp.1524-1533
• /
• 2021

Purpose To investigate the incidence of tuberculosis (TB) in healthcare workers (HCWs) with positive interferon-gamma release assay (IGRA) results based on chest X-ray (CXR) and CT findings and determine the role of imaging in the diagnosis of TB. Materials and Methods Among 1976 hospital personnel screened for TB using IGRA, IGRApositive subjects were retrospectively investigated. Clustered nodular and/or linear streaky opacities in the upper lung zone were considered positive on CXR. The CT findings were classified as active, indeterminate, inactive, or normal. The active or indeterminate class was considered CT-positive. Results IGRA was positive in 255 subjects (12.9%). CXR and CT were performed in 249 (99.2%) and 113 subjects (45.0%), respectively. CXR- and CT-positive findings were found in 7 of 249 (2.8%) and 9 of 113 (8.0%) patients, respectively. Among the nine CT-positive subjects, active and indeterminate TB findings were found in 6 (5.3%) and 3 (2.7%) patients, respectively. Microbiological tests, including acid-fast bacilli staining, culture, and polymerase chain reaction for TB, were negative in all nine CT-positive subjects. Empirical anti-TB medications were administered to 9 CT-positive subjects, and 3 of these nine subjects were CXR-negative for pulmonary TB.

• The Journal of the Korea Contents Association
• /
• v.15 no.11
• /
• pp.298-306
• /
• 2015

This study is the material of the additional filter(Cu, Ni, CaWO4, Gd+Ba) being used when the diagnosis X-ray was varied to evaluate the spatial dose distribution accordingly. And it suggest to find a suitable material. Experiments using MCNPX program based on the Monte Carlo simulation method was carried out by selecting the chest and abdomen taken. As a result, each material per dose, the average scatter dose is approximately 62%, 100 cm radius of the point of the simulated body surface exposure dose and 50 cm radius centered on the point average about 47%. It is determined that an Al material is currently available in accordance with the result to be replaced by Cu, Ni material is sufficient. With just the thickness due to the difference in the atomic number and density adjusted to be about one-tenth of the Al it will be suitable.

• Tuberculosis and Respiratory Diseases
• /
• v.39 no.1
• /
• pp.55-61
• /
• 1992

Background: The lung is the most common site of metastasis and usually it manifests as a single or multiple nodules in chest X-ray. But less commonly the cancer spreads through the lymphatics and X-ray shows diffuse reticulonodular densities. Sometimes, patient is presented with respiratory symptoms only with interstitial lung infiltration before the signs of primary tumor and in that cases, the differential diagnosis with other interstitial lung disease is required. We have experienced 5 such cases, who were diagnosed as lymphangitic carcinomatosis by transbronchial lung biopsy. Methods: Clinical manifestation, pulmonary function test, modified thin section CT, bronchoalveolar lavage and transbronchial lung biopsy were done. Results: The primary tumor was gastric cancer in 3, lung cancer in 2. Pulmonary function test showed restrictive pattern with low DLco in 2 patients and obstructive pattern in one. Bronchoalveolar lavage showed lymphocytosis in 4 patients and malignant cells were found in one patient. Transbronchial lung biopsy revealed malignant cells localized to the lymphatics (peribronchial, perivascular and perialveolar). Cell type was adenocarcinoma in 4 and squamous cell carcinoma in one. Conclusion: Rarely lymphangitic carcinomatosis can be presented as diffuse interstitial lung disease and easily diagnosed by transbronchial lung biopsy.

• Journal of the Korean Society of Radiology
• /
• v.14 no.5
• /
• pp.661-668
• /
• 2020

One of the causes of death for pregnant female is embolism, when a chest PA examination is performed. In addition, due to small doses, the examinations are performed for the purpose of preparing for pre-delivery emergency surgery or basic examination for pregnant female. Evaluating fetal doses through actual measurements is subject to ethical problems, Monte Carlo simulations assesses the organ and fetal doses of pregnant females according to week of pregnancy. The results of the simulations showed that the fetal dose decreased according to weeks of pregnancy and it showed a dose of about 0.1 mGy. The higher the density and thickness of the shielding material, the better the shielding effect. In addition, the dose reduction effect for each shielding material is between 40 and 98%. Afterwards, it is deemed necessary to study the reduction of fetal doses through various shielding characteristics and methods.

• Journal of the Korean Society of Radiology
• /
• v.9 no.7
• /
• pp.473-478
• /
• 2015

Abdominal obesity is one of the most influential index to predict of insulin resistance syndrome/metabolic syndrome in social demographic characteristics. It is matter of fact that radiation dose are increasing with development of medical treatment and device. In this study, we estimated distortion between reference image and entrance surface dose when take a chest radiography forward chest phantom assumed abdominal obesity. When angle of chest phantom incline $5^{\circ}$ forward, thoracic transverse and longitudinal diameter increase 1.22% and 0.44% each. Also cardiac transverse diameter increase 1.01% and cardio-throracic ratio (CTR) decrease 0.27% in the same situation of incline to $5^{\circ}$ forward. Thoracic transverse diameter shows the largest increase, and CTR was decreased. But entrance surface dose to phantom increase significantly 6.12% when angle of chest phantom incline $5^{\circ}$ forward. In conclusion, we have to pay attention to accurate positioning, to prevent a distortion of image through incline, and make patients not to expose to additional radiation.

• Journal of the Korean Society of Radiology
• /
• v.12 no.7
• /
• pp.827-832
• /
• 2018

Research on effective dose analysis of actual conditions of use based on large data is scarce. In this study, the exposure conditions of Chest X-Ray examinations used by 324 medical institutions in Korea were calculated and evaluated using computer simulations. As a result of the experiment, the effective dose in the low energy parameter bands was 0.024 mSv, followed by spleen, adrenal glands, and lung. The effective dose in the high-energy exposure parameter band was 0.123 mSv, followed by height, spleen and adrenal glands. The effective dose was 0.017 mSv when the optimal conditions considered the quality and exposure proposed in Park's study were used. The results of the study will be a reference for chest X-rays and will help reduce patient exposure.

• Proceedings of the KOR-BRONCHOESO Conference
• /
• 1982.05a
• /
• pp.7.2-7
• /
• 1982

Foreign body in the esophagus is not uncommom in the otolaryngological field, but esophageal perforation followed by pneumothrax due to esophageal foreign body is very rare. Authors recently experienced such a case developed in 1 year old male baby. This baby had been treated at local clinic for 2 weeks prior to admission under the impression of U.R.I.. Thereafter foreign body in the esophagus (fine wire pin) with left pneumothorax was detected by chest X-ray and the body was transfered to our hospital. Closed chest tube was inserted on left and under the general anesthesia, foreign body was removed by esophagoscopy. He was discharged on 16th postoperative day uneventfully.

• Journal of radiological science and technology
• /
• v.37 no.3
• /
• pp.177-185
• /
• 2014

This study focused on effects of patient exposure dose reduction with AEC (Auto Exposure Control) marker that is designed for showing location of AEC in X-ray Chest radiography. It included 880 adults who have to use Chest X-ray Digital Radiography system (DRS, LISTEM, Korea). AEC (Ion chambers are posited in top of both sides) are used to every adult and set X-ray system as Field size $17{\times}17inch$, 120kVp, FFD 180cm. 440 people of control group are posited on detector to include both sides of lung field and the other 440 people of experimental group are set to contact their lung directly to Ion chamber (making marker to shows location). Then, measured every DAP and, estimated patient effective dose by using PCXMC 2.0. The average age of control group (M:F=245:195) is 53.9 and the average BMI is 23.4. BMI ranges from under weight: 35, normal range: 279, over weight: 106 to obese: 20 and average DAP is 223.56mGycm2, Mean effective dose is 0.045mSv. The average age of experimental group (M:F=197:243) is 53.7 and the average BMI is 22.7. BMI ranges from under weight: 34, normal range: 315, over weight: 85 to obese: 6 and average DAP is 207.36mGycm2, Mean effective dose is 0.041mSv. Experimental group shows less Mean effective dose as 0.004mSv (9.7%) than control group. Also, patient numbers who got over exposure more than 0.056mSv (limit point to know efficiency of AEC marker) is 65 in control group (14.7%), 19 in experimental group (4.3%) and take statistics with t-Test. The statistical difference between two groups is 0.006. In order to use proper amount of X-ray in auto exposure controlled chest X-ray system, matching location between ion chamber and body part is needed, and using AEC marker (designed for showing location of ion chamber) is a way to reduce unnecessary patient exposure dose.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.725-733
• /
• 2023

The purpose of this study is to evaluate the effectiveness of virtual grid software (VGS). The purpose of this study is to evaluate the changes in energy and object thickness by dividing the use of VGS into two cases (Without-VGS) without using a movable grid. We attempted to determine the effectiveness of VGS by acquiring images using a chest phantom and a thigh phantom and analyzing SNR and CNR. In the chest phantom and femoral phantom, the tube flow was fixed at 2.5 mAs, and the tube voltage was changed by 10 kVp from 60 to 100 kVp to measure SNR and CNR, and SNR was about 1.09 to 8.86% higher in the chest phantom than in Without-VGS, and CNR was 4.18 to 14.56% higher in the VGS than in Without-VGS. And in the femoral phantom, SNR was about 9.78 to 18.05% higher in VGS than in Without-VGS, and CNR was 21.07 to 44.44% higher in VGS than in Without-VGS. The tube voltage was fixed at 70 kVp in the chest phantom and the femoral phantom, and the amount of tube current was changed at 2.5 to 16 mAs, respectively, and after X-ray irradiation, SNR and CNR were measured in the chest phantom, which was about 1.49 to 11.11% higher in VGS than in Without-VGS, and CNR was 4.76 to 13.40% higher in VGS than in Without-VGS. And in the femoral phantom, SNR was about 2.22 to 17.38% higher in VGS than in Without-VGS, and CNR was 13.85 to 40.46% higher in VGS than in Without-VGS. Therefore, if an inspection is required with a mobile X-ray imaging device, it is believed that good image quality can be obtained by using VGS in an environment where it is difficult to use a mobile grid, and it is believed that the use of mobile X-ray devices can be increased.