• Journal of the Korean Society of Radiology
• /
• v.83 no.2
• /
• pp.344-359
• /
• 2022

Purpose To develop a denoising convolutional neural network-based image processing technique and investigate its efficacy in diagnosing breast cancer using low-dose mammography imaging. Materials and Methods A total of 6 breast radiologists were included in this prospective study. All radiologists independently evaluated low-dose images for lesion detection and rated them for diagnostic quality using a qualitative scale. After application of the denoising network, the same radiologists evaluated lesion detectability and image quality. For clinical application, a consensus on lesion type and localization on preoperative mammographic examinations of breast cancer patients was reached after discussion. Thereafter, coded low-dose, reconstructed full-dose, and full-dose images were presented and assessed in a random order. Results Lesions on 40% reconstructed full-dose images were better perceived when compared with low-dose images of mastectomy specimens as a reference. In clinical application, as compared to 40% reconstructed images, higher values were given on full-dose images for resolution (p < 0.001); diagnostic quality for calcifications (p < 0.001); and for masses, asymmetry, or architectural distortion (p = 0.037). The 40% reconstructed images showed comparable values to 100% full-dose images for overall quality (p = 0.547), lesion visibility (p = 0.120), and contrast (p = 0.083), without significant differences. Conclusion Effective denoising and image reconstruction processing techniques can enable breast cancer diagnosis with substantial radiation dose reduction.

• Journal of Broadcast Engineering
• /
• v.11 no.1 s.30
• /
• pp.3-14
• /
• 2006

Music summarization means a technique which automatically generates the most importantand representative a part or parts ill music content. The techniques of music summarization have been studied with two categories according to summary characteristics. The first one is that the repeated part is provided as music summary and the second provides the combined segments which consist of segments with different characteristics as music summary in music content In this paper, we propose and evaluate two kinds of music summarization techniques. The algorithm using multi-level vector quantization which provides a repeated part as music summary gives fixed-length music summary is evaluated by overlapping ration between hand-made repeated parts and automatically generated summary. As results, the overlapping ratios of conventional methods are 42.2% and 47.4%, but that of proposed method with fixed-length summary is 67.1%. Optimal length music summary is evaluated by the portion of overlapping between summary and repeated part which is different length according to music content and the result shows that automatically-generated summary expresses more effective part than fixed-length summary with optimal length. The cluster-based algorithm using 2-D similarity matrix and k-means algorithm provides the combined segments as music summary. In order to evaluate this algorithm, we use MOS test consisting of two questions(How many similar segments are in summarized music? How many segments are included in same structure?) and the results show good performance.

• Journal of the Korean Society of Radiology
• /
• v.83 no.6
• /
• pp.1327-1341
• /
• 2022

Purpose To evaluate the pattern of use and the perception of digital breast tomosynthesis (DBT) among Korean breast radiologists. Materials and Methods From March 22 to 29, 2021, an online survey comprising 27 questions was sent to members of the Korean Society of Breast Imaging. Questions related to practice characteristics, utilization and perception of DBT, and research interests. Results were analyzed based on factors using logistic regression. Results Overall, 120 of 257 members responded to the survey (response rate, 46.7%), 67 (55.8%) of whom reported using DBT. The overall satisfaction with DBT was 3.31 (1-5 scale). The most-cited DBT advantages were decreased recall rate (55.8%), increased lesion conspicuity (48.3%), and increased cancer detection (45.8%). The most-cited DBT disadvantages were extra cost for patients (46.7%), insufficient calcification characterization (43.3%), insufficient improvement in diagnostic performance (39.2%), and radiation dose (35.8%). Radiologists reported increased storage requirements and interpretation time for barriers to implementing DBT. Conclusion Further improvement of DBT techniques reflecting feedback from the user's perspective will help increase the acceptance of DBT in Korea.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.14 no.2
• /
• pp.45-49
• /
• 2010

Purpose: Several radiopharmaceuticals were used for cisternography. But recently, due to more short acquisition time, high resolution than other radiopharmaceuticals like In-111 DTPA, we were using Tc-99m DTPA in cisternography. Using of Tc-99m DTPA for intrathecal, was not officially recognised by the FDA. And there are matters of aseptic meningitis, muscular tetany, seizures by inappropriate radiopharmaceuticals handling. So, it is necessary to prevent adverse reactions while handling the radiopharmaceuticals using in cisternography. Therefore, this study aims to evaluation of usefulness and procedures for safety of radiopharmaceuticals in cisternography. Materials and Methods: Subjects were 12radioactive tracer vials using in cisternography in 2008 Dec. 16 - 2009 Dec. 30. (1) Radioactive tracer Vial test - We were measured NaPertechnetate radiation dose and volume, normal saline volume for dilution, source volume and dose activity for patient injection. And then, calculated mass of pure DTPA. (2) Bacterial endotoxin test - We performed pyrogen test using by negative/positive control vials which was added normal saline 0.2 mL and added normal saline 0.1 mL, Tc-99m DTPA 0.1 mL in test control vial. And then, reacted by digital hot plate in $37.5^{\circ}C$ for 1 hour and compared of gel-clot in each control vials. (3) Compliance safety procedure - We were checked safety issues and wrote out a safety procedure exam sheet. Results: (1) Radioactive tracer Vial test - Mass of DTPA per dose for patient injection (mg) was 0.88 (mg) on average, and Mass of DTPA per volume for patient injection (mg) was 0.74 (mg) on average. (2) Bacterial endotoxin test - All control test vials showed negative reactions. (3) Compliance safety procedure - We were checked safety issues and wrote out a safety exam sheet in all the exams. So, there were no adverse reactions. Conclusion: We could examine easier to safety tests using by Techscan - DTPA (Mallinckrodt): CaNa3. Each test results were passed the safety tests and there are no adverse reactions. The use of Tc-99m DTPA for cisternography, always has been become an issue. Since it has occur adverse reaction while examine the cisternography using by Tc-99m DTPA, it needs to set up the 'Standard Operating procedures'.

• Progress in Medical Physics
• /
• v.27 no.2
• /
• pp.86-92
• /
• 2016

An absorbed dose calculation method using a digital phantom is implemented in normal organs. This method cannot be employed for calculating the absorbed dose of tumor. In this study, we measure the S-value for calculating the absorbed dose of each organ and tumor. We inject a radioisotope into a torso phantom and perform Monte Carlo simulation based on the CT data. The torso phantom has lung, liver, spinal, cylinder, and tumor simulated using a spherical phantom. The radioactivity of the actual absorbed dose is measured using the injected dose of the radioisotope, which is Cu-64 73.85 MBq, and detected using a glass dosimeter in the torso phantom. To perform the Monte Carlo simulation, the information on each organ and tumor acquired using the PET/CT and CT data provides anatomical information. The anatomical information is offered above mean value and manually segmented for each organ and tumor. The residence time of the radioisotope in each organ and tumor is calculated using the time activity curve of Cu-64 radioactivity. The S-values of each organ and tumor are calculated based on the Monte Carlo simulation data using the spatial coordinate, voxel size, and density information. The absorbed dose is evaluated using that obtained through the Monte Carlo simulation and the S-value and the residence time in each organ and tumor. The absorbed dose in liver, tumor1, and tumor2 is 4.52E-02, 4.61E-02, and 5.98E-02 mGy/MBq, respectively. The difference in the absorbed dose measured using the glass dosimeter and that obtained through the Monte Carlo simulation data is within 12.3%. The result of this study is that the absorbed dose obtained using an image can evaluate each difference region and size of a region of interest.

• Progress in Medical Physics
• /
• v.22 no.4
• /
• pp.184-189
• /
• 2011

The purpose of this study was to compare the image between DSA and MDCT Angiography and to examine whether MDCT Angiography could be useful as a screening test for the diagnosis of cerebral aneurysm in patients who were diagnosed with cerebral aneurysm on DSA. Of patients who were diagnosed with cerebral aneurysm DSA at University Hospital, 194 patients who concomitantly underwent MDCT Angiography were enrolled in the current retrospective study. The methods for analyzing cerebral aneurysm were to analyze the presence of cerebral aneurysm on DSA and MDCT Angiography. In cases in which it exceeded 1, the corresponding cases were classified as narrow-neck aneurysms. In otherwise cases, they were classified as wide-neck aneurysms. Thus, a comparative analysis could be performed to ascertain if cases were narrow-neck or wide-neck aneurysms. As compared with DSA, the sensitivity of MDCT Angiography for cerebral aneurysm was measured to be 97.4%. The degree of consistency between narrow-neck and wide-neck aneurysms was 90.2% and the proportion of undetectable an at MDCT Angiography was 2.54%. mean size was 2.4 mm. It is expected that a non-invasive diagnostic modality for a screening test for cerebral aneurysm, MDCT Angiography might be a very useful regimen as compared with an invasive one, DSA.

• The Journal of the Korea Contents Association
• /
• v.12 no.10
• /
• pp.340-348
• /
• 2012

To evaluate the usefulness of tomosynthesis in the chest area, simple radiograph, low-dose CT, and tomosynthesis examinations were performed, and their absorbed doses were compared, and finally the images were evaluated. The absorbed dose recorded with the simple Radiograph examination was $0.33{\pm}0.27$ mGy, that of low-dose CT $1.26{\pm}0.56$ mGy, and that of tomosynthesis $0.55{\pm}0.02$ mGy, which indicate significance differences in absorbed doses among the examinations(p<0.001). Based on the evaluations of the images, The simple radiograph scores were $1.66{\pm}0.72$, $1.61{\pm}0.63$, and $1.57{\pm}0.73$, respectively; low-dose CT scores were $2.92{\pm}0.26$, $2.91{\pm}0.29$, and $2.88{\pm}0.32$, respectively; and tomosynthesis scores were $2.69{\pm}0.51$, $2.76{\pm}0.43$, and $2.66{\pm}0.61$, respectively. That is, there were statistically significant differences among the examinations(p<0.001), although there was no significant difference between low-dose CT and tomosynthesis examinations. Therefore, tomosynthesis is judged to be a useful examination that can minimize radiation doses to patients during chest examinations and enhance diagnostic efficacy.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_4
• /
• pp.1179-1194
• /
• 2020

This study investigates the feasibility of three algorithms, K-Nearest Neighbors (K-NN), Random Forest (RF) and Neural Network (NN), for estimating the air temperature of an unobserved area where the weather station is not installed. The satellite image were obtained from Landsat-8 and MODIS Aqua/Terra acquired in 2019, and the meteorological ground weather data were from AWS/ASOS data of Korea Meteorological Administration and Korea Forest Service. In addition, in order to improve the estimation accuracy, a digital surface model, solar radiation, aspect and slope were used. The accuracy assessment of machine learning methods was performed by calculating the statistics of R² (determination coefficient) and Root Mean Square Error (RMSE) through 10-fold cross-validation and the estimated values were compared for each target area. As a result, the neural network algorithm showed the most stable result among the three algorithms with R² = 0.805 and RMSE = 0.508. The neural network algorithm was applied to each data set on Landsat imagery scene. It was possible to generate an mean air temperature map from June to September 2019 and confirmed that detailed air temperature information could be estimated. The result is expected to be utilized for national disaster safety management such as heat wave response policies and heat island mitigation research.

• Journal of the Korean Society of Radiology
• /
• v.3 no.2
• /
• pp.11-16
• /
• 2009

Mercury Iodide as good sensitivity at radiation and has an easy peculiarity that operates at low voltage for other photoconductors(a-Se, a-Si, Ge, etc) Based on this characteristic, we studied about an efficiency of the digital x-ray detector in acccordance with the thickness of photoconductor. To solve the problem that is difficult to make a large area film using PVD(Physical Vapor Deposition)method, we used a PIB(Particle In Binder)method. To make a binder paste, we used a PVB(Polyvinylbutyral) as a binder and a DGME(Diethylene Glycol Monobutyl Ether), DGMEA(Diethylene Glycol Monobutyl Ether Acetate) as a solvent. Using this binder paste, we made a polycrystal mercury iodide film that has an each thickness. To evaluate the electrical properties of this films, we measured a darkcurrent, sensitivity and SNR(Signal to Noise Ratio). Mercury iodide film of the 200um thickness has good electrical properties as a result of the measurement. From this result there is a good chance that replace the existing a-Se(Amnorphous seleinum; a-se) with the mercury iodide.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.8 no.4
• /
• pp.242-249
• /
• 2006

Potential impacts of the future climate change on crop production can be inferred by crop simulations at a landscape scale, if the climate data may be provided at appropriate spatial scales. Northern Gyunggi Province is one of the few prospective regions in South Korea for growing quality soybeans. Any geographical shift of production areas under the changing climate may influence the current land planning policy in this region. A soybean growth simulation was performed at 342 land units in northern Gyunggi province to test the potential geographical shift of the current production areas for quality soybeans in the near future (form 2011 to 2100). The land units for soybean cultivation were selected by the land use, the soil characteristics, and the minimum arable land area. Daily maximum and minimum temperature, precipitation, the number of rain days and solar radiation were extracted for each land unit from the future digital climate models (DCM, 2011-2040, 2041-2070, 2071-2100). Daily weather data for 30 years were randomly generated for each land unit for each normal year by using a well-known statistical method. They were used to run CROPGRO-Soybean model to simulate the growth, phonology, and yields of 3 cultivars representing different maturity groups grown at 342 land units. According to the model calculations, the warming trend in this region will accelerate the flowering and physiological maturity of all cultivars, resulting in a 7 to 9 days reduction in overall growing season and a 1 to 15% reduction in grain yield of early to medium maturity cultivars. There was a slight increase in grain yield of the late maturing cultivar under the projected climate by 2070, but a decreasing tend was dominant by the year 2100.