• Journal of the Korean Society of Radiology
• /
• v.16 no.1
• /
• pp.1-6
• /
• 2022

Deep learning is a collection of algorithms that enable learning by summarizing the key contents of large amounts of data; it is being developed to diagnose lesions in the medical imaging field. To evaluate the accuracy of the cerebral hemorrhage diagnosis, we used a convolutional neural network (CNN) to derive the diagnostic accuracy of cerebral parenchyma computed tomography (CT) images and the cerebral parenchyma CT images of areas where cerebral hemorrhages are suspected of having occurred. We compared the accuracy of CNN with different numbers of hidden layers and discovered that CNN with more hidden layers resulted in higher accuracy. The analysis results of the derived CT images used in this study to determine the presence of cerebral hemorrhages are expected to be used as foundation data in studies related to the application of artificial intelligence in the medical imaging industry.

• Journal of the Korean Society of Radiology
• /
• v.16 no.3
• /
• pp.203-209
• /
• 2022

One of the typical methods for lowering radiation dose while maintaining image quality of computed tomography (CT) is the use of model-based iterative reconstruction (MBIR). This study is to evaluate the image quality by adjusting the strength of the advanced modeled iterative reconstruction (ADMIRE), which is well known as a representative model of MBIR. The study was conducted using phantom, and CT images were obtained while adjusting the strength of ADMIRE in units of 1 to 5. Quantitative evaluation includes noise levels using coefficient of variation (COV) and contrast to noise ratio (CNR), as well as natural image quality evaluation (NIQE) and blind/referenceless image spatial quality evaluator (BRISQUE). As a result, in both noise level and blind quality evaluation results, the higher the strength of ADMIRE, the better the results were derived. In particular, it was confirmed that COV and CNR were improved 1.89 and 1.75 times at ADMIRE 5 compared to ADMIRE 1, respectively, and NIQE and BRISQUE were proved to be improved 1.35 and 1.22 times at ADMIRE 5 compared to ADMIRE 1, respectively. In conclusion, this study was proved that the reconstruction strength of ADMIRE had a great influence on the noise level and overall image quality evaluation of CT images.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.5
• /
• pp.2278-2284
• /
• 2012

Exposure-dose reducing effect was measured by using bolus, a tissue-equivalent material as a shield to obtain useful diagnostic images while minimizing the radiation exposure of thyroid which is highly sensitive to radiation during panoramic radiography. The experiment was performed within the period of 1 June 2001 through 30 June 2011 by measuring entrance surface dose and deep dose at the thyroid-corresponding site of a head and neck phantom. As a result, the entrance surface dose in the thyroid for using no shield was 43.84 ${\mu}Gy$ on the average, and the thyroid shield of bolus 10 mm in thickness reduced the dose by 15.45 ${\mu}Gy$(35.24%) to 28.39 ${\mu}Gy$ on the average. The use of a 20 mm thyroid shield resulted in the dose of 25.38 ${\mu}Gy$ on the average, a 18.46 ${\mu}Gy$(42.10%) drop from 43.84 ${\mu}Gy$ for using no shield. On the site 20 mm below the surface, a thyroid shield 10 mm in thickness had no dose-reducing effect, while a 20 mm thyroid shield reduced the dose by 0.06 mSv(20%).

• Imaging Science in Dentistry
• /
• v.54 no.3
• /
• pp.271-275
• /
• 2024

Purpose: Recent advancements in artificial intelligence (AI), particularly tools such as ChatGPT developed by OpenAI, a U.S.-based AI research organization, have transformed the healthcare and education sectors. This study investigated the effectiveness of ChatGPT in answering dentistry exam questions, demonstrating its potential to enhance professional practice and patient care. Materials and Methods: This study assessed the performance of ChatGPT 3.5 and 4 on U.S. dental exams - specifically, the Integrated National Board Dental Examination (INBDE), Dental Admission Test (DAT), and Advanced Dental Admission Test (ADAT) - excluding image-based questions. Using customized prompts, ChatGPT's answers were evaluated against official answer sheets. Results: ChatGPT 3.5 and 4 were tested with 253 questions from the INBDE, ADAT, and DAT exams. For the INBDE, both versions achieved 80% accuracy in knowledge-based questions and 66-69% in case history questions. In ADAT, they scored 66-83% in knowledge-based and 76% in case history questions. ChatGPT 4 excelled on the DAT, with 94% accuracy in knowledge-based questions, 57% in mathematical analysis items, and 100% in comprehension questions, surpassing ChatGPT 3.5's rates of 83%, 31%, and 82%, respectively. The difference was significant for knowledge-based questions(P=0.009). Both versions showed similar patterns in incorrect responses. Conclusion: Both ChatGPT 3.5 and 4 effectively handled knowledge-based, case history, and comprehension questions, with ChatGPT 4 being more reliable and surpassing the performance of 3.5. ChatGPT 4's perfect score in comprehension questions underscores its trainability in specific subjects. However, both versions exhibited weaker performance in mathematical analysis, suggesting this as an area for improvement.

• Journal of the Korean Society of Radiology
• /
• v.10 no.2
• /
• pp.139-144
• /
• 2016

The study of radiation-hazard in the human skin tissue is carried out by direct irradiating to experimental animals. The influences of a radiation to the animal's skin tissue are analyzed from this experiment. However, this also accompanies losses in terms of both time and economy. In this study, we simulated human tissue by using a swine skin tissue. The depth of the swine skin tissue for the experiment is determined, and the amount of the direct radiation below this skin depth is analyzed numerically. The amount of the radiation occurred by exposure below the skin tissue can be inferred. Moreover, it is possible to use only cells effectively and animal experiments to analyze the body-hazard by radiation.

• Journal of the Korean Society of Radiology
• /
• v.11 no.5
• /
• pp.313-319
• /
• 2017

The purpose of this study is to confirm the double of optic foramen in Korean and apply it to the Rhese method. First, the angle between the right optic foramen and the MSP was measured on the axial image using MPR technique of the 3D CT. Second, we measured the angle between the right optic foramen and OML in sagittal of MPR images. As a result, the angle between the optic foramen and the MSP was $39.9{\pm}4.63^{\circ}$ on average, which was different from the $53^{\circ}$ presented by Rhese method(p<0.05). The angle between optic foramen and OML was $40.8{\pm}6.6^{\circ}$. In conclusion, this study confirms that the standard of the Rhese method proposed in current textbook is difficult to apply to Koreans. Therefore, it is necessary to study angle of Korean standard in various general x-ray technique.

• Journal of radiological science and technology
• /
• v.41 no.5
• /
• pp.463-469
• /
• 2018

Morphological information such as shape and margin of micro lesion is important information for diagnosis of disease in clinical imaging. In this study, we investigated the morphological changes of the micro lesions by comparing the contrast and area in grid suppressed DR images according to grid frequency. In the profile analysis of the image, the mass showed an average intensity variation of 8.6 ~ 72.4 after suppression, The higher the grid frequency, the more the contrast was increased. However, in the images obtained using 103 lp / inch, which is a grid frequency less than the sampling frequency, the contrast of the mass in the vertical direction decreased after suppression. In the binary image, the area change of the mass was also large. As a result, the shape, size, and margin of the mass changed. In the case of very small calcification, the higher the grid frequency is the larger the change in contrast, so that a clear image can be obtained in the post-suppression image. However, we could confirm that the margin of the lesion was blurred and the lesion was lost in some of the images using the 103 lp / inch grid. The higher the frequency of the grid, The change of the contrast of fiber occurred largely and clear boundary was confirmed. The decrease of the number of pixels was small and morphological change was small. In conclusion, when using a grid frequency that is not suitable for the sample frequency, morphological changes or lesion loss of micro lesions in the post- suppression image may give the possibility of misdiagnosis in diagnosis and differentiation of the image.

• Journal of the Korean Society of Radiology
• /
• v.7 no.2
• /
• pp.137-143
• /
• 2013

This study presents comparison results between axial and spiral scanning in the head and chest region with 64 MDCT to evaluate organ doses in infants and toddlers, who are more radiosensitive to radiation than adults and rise in the number of CT examinations, during CT scanning. Organ doses were significantly lower in spiral scanning than axial scanning regardless of scanned regions. The average organ dose for the chest scan using pitch of 1.355 was found to be significantly higher(average -12.03%) than for the other two pitch settings(0.525 and 0.988) in the spiral scanning mode compared with the axial one. Organ doses in the spiral scanning mode were lower by average 20.54% than the axial scanning mode. The results of the study that evaluated organ doses with an anthropomorphic phantom will help to demonstrate the result values of Monte Carlo simulations and make a contribution to more accurate evaluations of organ doses in toddlers undergoing a CT examination.

• Journal of the Korean Society of Radiology
• /
• v.12 no.3
• /
• pp.289-295
• /
• 2018

Most diagnostic devices in the medical field use X-ray sources, which emit energy spectra. In radiological diagnosis, the quantitative and qualitative analyses of X-rays are essential for maintaining the image quality and minimizing the radiation dose to patients. This work aims to obtain the X-ray energy spectra used in diagnostic imaging by Monte Carlo simulation. Various X-ray spectra are simulated using a Monte Carlo simulation tool. These spectra are then compared to the reference data obtained with a tungsten anode spectral model using the interpolating polynomial (TASMIP) code. The X-ray tube voltages used are 50, 60, 80, 100, and 110 kV, respectively. CdTe and a-Se detector are used as the detectors for obtaining the X-ray spectra. Simulation results demonstrate that the various X-ray spectra are well matched with the reference data. Based on the simulation results, an appropriate X-ray spectrum, in accordance with the tube voltage, can be selected when generating an image for diagnostic imaging. The dose to be delivered to the patient can be predicted prior to examination in the diagnostic field.

• Journal of the Korean Society of Radiology
• /
• v.12 no.4
• /
• pp.511-518
• /
• 2018

The purpose of this study was to measured the diameter, maximum diameter, maximum area and volume of the cerebral aneurysm in 53 patients who underwent three-dimensional digital angiography and three-dimensional digital subtraction angiography, which were used for the clinical diagnosis of cerebral aneurysm, image noise and radiation exposure dose of each test method were analyzed to compare clinical diagnosis differences in the cerebral aneurysm diagnosis. Three-dimensional digital angiography and three-dimensional digital subtraction angiography showed that the neck diameter, maximum diameter, maximum area, volume, and noise of the cerebral aneurysm were identical or very small. However, the three-dimensional digital angiography significantly decreased the radiation exposure dose compared to three-dimensional digital subtraction angiography. Therefore, in case of clinical diagnosis of cerebral aneurysm, three-dimensional digital angiography should be preferentially used to reduce radiation exposure dose of patient.