• The Korean Journal of Nuclear Medicine Technology
• /
• v.24 no.1
• /
• pp.15-19
• /
• 2020

Purpose Generative Adversarial Network(GAN) is one of deep learning technologies. This is a way to create a real fake image after learning the real image. In this study, after acquiring artificial intelligence images through GAN, We were compared and evaluated with real scan time images. We want to see if these technologies are potentially useful. Materials and Methods 30 patients who underwent ¹⁸F-FDG Brain PET/CT scanning at Severance Hospital, were acquired in 15-minute List mode and reconstructed into 1,2,3,4,5 and 15minute images, respectively. 25 out of 30 patients were used as learning images for learning of GAN and 5 patients used as verification images for confirming the learning model. The program was implemented using the Python and Tensorflow frameworks. After learning using the Pix2Pix model of GAN technology, this learning model generated artificial intelligence images. The artificial intelligence image generated in this way were evaluated as Mean Square Error(MSE), Peak Signal to Noise Ratio(PSNR), and Structural Similarity Index(SSIM) with real scan time image. Results The trained model was evaluated with the verification image. As a result, The 15-minute image created by the 5-minute image rather than 1-minute after the start of the scan showed a smaller MSE, and the PSNR and SSIM increased. Conclusion Through this study, it was confirmed that AI imaging technology is applicable. In the future, if these artificial intelligence imaging technologies are applied to nuclear medicine imaging, it will be possible to acquire images even with a short scan time, which can be expected to reduce artifacts caused by patient movement and increase the efficiency of the scanning room.

• Convergence Security Journal
• /
• v.21 no.2
• /
• pp.57-66
• /
• 2021

Deep neural networks (DNNs) provide excellent performance for image, speech, and pattern recognition. However, DNNs sometimes misrecognize certain adversarial examples. An adversarial example is a sample that adds optimized noise to the original data, which makes the DNN erroneously misclassified, although there is nothing wrong with the human eye. Therefore studies on defense against adversarial example attacks are required. In this paper, we have experimentally analyzed the success rate of detection for adversarial examples by adjusting various parameters. The performance of the ensemble defense method was analyzed using fast gradient sign method, DeepFool method, Carlini & Wanger method, which are adversarial example attack methods. Moreover, we used MNIST as experimental data and Tensorflow as a machine learning library. As an experimental method, we carried out performance analysis based on three adversarial example attack methods, threshold, number of models, and random noise. As a result, when there were 7 models and a threshold of 1, the detection rate for adversarial example is 98.3%, and the accuracy of 99.2% of the original sample is maintained.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.6A
• /
• pp.139-149
• /
• 2008

The amount of personal information collected by organizations and government agencies is continuously increasing. When a data collector publishes personal information for research and other purposes, individuals' sensitive information should not be revealed. On the other hand, published data is also required to provide accurate statistical information for analysis. k-Anonymity and ${\iota}$-diversity models are popular approaches for privacy preserving data publication. However, they are limited to static data release. After a dataset is updated with insertions and deletions, a data collector cannot safely release up-to-date information. Recently, the m-invariance model has been proposed to support re-publication of dynamic datasets. However, the m-invariant generalization can cause high information loss. In addition, if the adversary already obtained sensitive values of some individuals before accessing released information, the m-invariance leads to severe privacy disclosure. In this paper, we propose a novel technique for safely releasing dynamic datasets. The proposed technique offers a simple and effective method for handling inserted and deleted records without generalization. It also gives equivalent degree of privacy preservation to the m-invariance model.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.3
• /
• pp.77-84
• /
• 2021

In this study, flexural tests of prestressed high strength spun concrete (PHC) piles reinforced with infilled concrete and longitudinal rebars were conducted, where the longitudinal rebar ratio and the presence of sludge formed on the inner surface of PHC pile were set as key test variables. A total of six PHC pile specimens were manufactured, and their flexural behaviors including failure mode, crack pattern, longitudinal strain distribution in a section and end slip between external PHC pile and infilled concrete were measured and discussed in detail. The test results revealed that the flexural stiffness and strength increased as the longitudinal rebar ratio became larger, and that the sludge formed on the inner surface of PHC pile did not show any detrimental effect on the flexural performance. In addition to the experimental approach, this study presents a nonlinear flexural analysis model considering compatibility conditions and strain and stress distributions of the PHC piles and infilled concrete. The rationality of the nonlinear flexural analysis model was verified by comparing it with test results, and it appeared that the proposed model well evaluated the flexural behavior of PHC piles reinforced with infilled concrete and longitudinal rebars with a good accuracy.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.6
• /
• pp.33-45
• /
• 2020

This paper aims to look at the perspective that the latest cutting-edge technologies are predicting individual diseases in the actual medical environment in a situation where various types of wearable devices are rapidly increasing and used in the healthcare domain. Through the process of collecting, processing, and transmitting data by merging clinical data, genetic data, and life log data through a user-participating wearable device, it presents the process of connecting the learning model and the feedback model in the environment of the Deep Neural Network. In the case of the actual field that has undergone clinical trial procedures of medical IT occurring in such a high-tech medical field, the effect of a specific gene caused by metabolic syndrome on the disease is measured, and clinical information and life log data are merged to process different heterogeneous data. That is, it proves the objective suitability and certainty of the deep neural network of heterogeneous data, and through this, the performance evaluation according to the noise in the actual deep learning environment is performed. In the case of the automatic encoder, we proved that the accuracy and predicted value varying per 1,000 EPOCH are linearly changed several times with the increasing value of the variable.

• Journal of IKEEE
• /
• v.28 no.1
• /
• pp.97-103
• /
• 2024

Some drugs can offer far better medical effectiveness as it is injected through the intradermal layer of the skin, known as a needle-free injection. However, conventional needle-free devices might deliver a relatively large amount of drug in a just single spot of skin, splitting open the tissue layer structure, which might cause bruising and bleeding. By injecting the small volume with a fast repetition rate in a large surface area of skin, the patient may get much fewer injuries and pain. To achieve that specification, the driving force must be instantaneous and short-pulsed. Such a form of an injection device has been developed but the efficacy of those devices has been rarely examined. Therefore, this study developed the laser-induced microjet device that ejects microjet whose speed is ~310 m/s, during the 400~800 ㎲ of pulse time. The device can eject ~1 µL of the drug at the rate at which each shot repeated 10 shots per second. Using this specification, we evaluated the efficacy of drug injection onto mouse models. After injecting the insulin solution into the mouse model, the blood insulin level is detected, resulting in 20 % of blood insulin level with the ordinary needle syringe injection method.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.12 no.1
• /
• pp.19-26
• /
• 2008

Purpose: In general, We discharged radioactive wastewater and sewages less than $8.1{\times}10^{-13}$ Ci/ml in a exclusive water-purifier tank. Our hospital operating three exclusive water-purifier tank for radioactive wastewater and sewages of 60 tons capacity respectively. In order to meet the criteria it need a enough decay more than 125 days per each exclusive tank. However, recently we fell into the serious situation that decay period was decreased remarkably, owing to the wastewater amount increased rapidly by enlarge the therapy ward. For that reason, in this article, I'd like to say the way that reducing of radioactive wastewater and sewages rationally. Materials and Methods: From January, 2006 to October, four hundred and two cases were analyzed. They were all hospitalized during 3 days and 2 nights. We calculated the average amount of water used (include toilet water used, shower water used, washstand water used, $\cdots$), each exclusive water-purifier tank's decay period, as well as try to search the increased factors about water-purifier tank inflow flux by re-analysis of the procedure of radioisotope therapy step by step. Results: We could increase each exclusive water-purifier tank's decay period from 84 days to 130 days through the improvement about following cause: (1) Improvement of conventional toilet stool for excessive water waste $\rightarrow$ Replacement of water saving style toilet stool (2) Prevention of unnecessary shower and wash (3) Stop the diuretics taking during hospitalization (4) Analysis of relationship between water intakes and residual dose of body (5) Education about outside toilet utilization before the administration (6) Changed each water-purifier tank's maximum level from85% to 90% Conclusion: The originality of our efforts are not only software but hardware performance improvements. Incidentally the side of software's are change of therapy procedures and protocols, the side of hardware's are replacement of water saving style toilet stool and change of each water-purifier tank's maximum level. Thus even if a long lapse of time, problem such as return to the former conditions may not happen. Besides, We expect that our trials become a new reasonable model in similar situation.

• The Journal of Korean Academy of Prosthodontics
• /
• v.57 no.2
• /
• pp.102-109
• /
• 2019

Purpose: The purpose of this study was to evaluate the accuracy of three types of intraoral scanners and the accuracy of the single abutment and bridge abutment model. Materials and methods: In this study, a single abutment, and a bridge abutment with missing first molar was fabricated and set as the reference model. The reference model was scanned with an industrial three-dimensional scanner and set as reference scan data. The reference model was scanned five times using the three intraoral scanners (CS3600, CS3500, and EZIS PO). This was set as the evaluation scan data. In the three-dimensional analysis (Geomagic control X), the divided abutment region was selected and analyzed to verify the scan accuracy of the abutment. Statistical analysis was performed using SPSS software (${\alpha}=.05$). The accuracy of intraoral scanners was compared using the Kruskal-Wallis test and post-test was performed using the Pairwise test. The accuracy difference between the single abutment model and the bridge abutment model was analyzed by the Mann-Whitney U test. Results: The accuracy according to the intraoral scanner was significantly different (P < .05). The trueness of the single abutment model and the bridge abutment model showed a statistically significant difference and showed better trueness in the single abutment (P < .05). There was no significant difference in the precision (P = .616). Conclusion: As a result of comparing the accuracy of single and bridge abutments, the error of abutment scan increased with increasing scan area, and the accuracy of bridge abutment model was clinically acceptable in three types of intraoral scanners.

• Tuberculosis and Respiratory Diseases
• /
• v.57 no.6
• /
• pp.535-542
• /
• 2004

Background : Despite the clinical clues of bronchial asthma, some chronic coughers fail to be diagnosed due to negative test results. This study was aimed at evaluating the diagnostic performance of routine objective tests and identifying a cost-effective approach for asthmatics with a chronic cough. Methods : Patients with a chronic cough of more than 3 weeks duration, and showing normal chest radiograph and spirometry were enrolled. On the first visit, objective tests, composed of serum total IgE, peripheral blood eosinophil count, spontaneous sputum eosinophil count, methacholine bronchial provocation test (MBPT) and paranasal sinus radiograph, were performed, with the simultaneous administration of oral prednisolone (0.5mg/kg) for one week. The final diagnoses were made on the basis of the test results, and the patients grouped according to their steroid responsiveness. The role of the etiologic diagnosis tests was evaluated, and the medical costs of the final management plan simulated with respect to three assumed models. Results : Sixty chronic coughers were finally analyzed. The final diagnoses were as follows: bronchial asthma 21.7%, eosinophilic bronchitis 6.7%, paranasal sinusitis 18.3%, presumptive allergy 8.3% and non-diagnostic case 45.0%. Ninety percent were steroid responder. With the bronchial asthma cases, the positive rate of MBPT was 38.5%, with sputum eosinophil count in 84.6%, serum total IgE in 38.5%, and a peripheral blood eosinophil count rate of 30.8%. When the test results and steroid responsiveness data were applied to the 3 models, the chest radiograph, spirometry, sputum eosinophil count and paranasal sinus radiograph test results, and simultaneous short term steroid treatment seemed to have acceptable diagnostic performances, which could be used as a further guide to cost-effective planning. Conclusion : Objective tests, composed of chest radiograph, spirometry, paranasal sinus radiograph and sputum eosinophil count, with simultaneous short term steroid treatment, are suggested as cost-effective approaches for asthmatics with a chronic cough.

• Journal of radiological science and technology
• /
• v.28 no.1
• /
• pp.33-44
• /
• 2005

The objective of this study is to supply the good quality of experts the radiation industries in Korea and develop the major subject matter needed in the radiation industries and the curriculum in order to execute it for the variation of fields of employment at the department of radiation in the junior college and the development of the percentage of employment. In addition, this study is to improve the level of radiation experts engaged in the industries in quality, and it is to improve the social recognition of radiation rather negative now because of the development of radiation industry. As for the core results of this research, it was to suggest the detailed choice method curriculum proper to the service fields of radiation industries, but it may be subject to change due to each college's property and the educational objectives. From the result of this research above, it may be summed up as follows. First, as for the detailed curriculum by the service field, this study was to organize two subject matters: 1. the subject matter proper to the field of using the radiation, and 2. the subject matters proper to the safety control field of radiation. Second, as for the detailed curriculum by the pattern of industries, this study was to organize the four subject matters: 1. the subject matter needed in the manufactures, 2. the subject matter needed in the nondestructive testing industries, 3. the subject matter needed in the sales agencies, and 4. the subject matter needed in the laboratories. This study was to suggest the operational model about the curriculum in order to execute these subject matters. It could be executed as two methods below. First, one method is to execute the major systems by the medical field and industrial field in the third course at the department of radiation in the junior college now. Second, the other method is to make them specialize the industrial radiation in the Advanced Course(one year course) after the graduation of junior college. To operate these curricula successively it needs to assume the deeper research and the development of materials about the subject matters related to the nuclear radiation industries hereafter. In addition, it needs to solve the security of finance like the manpower of professor, space for practice, and the educational appliances, etc. needed in the operation of subject matters. Finally, the effect and result from the development or revision of college curriculum did not come out in a short time. It will require considerable time until the undergraduates at the department in the junior college finish a set of curriculum newly developed, and graduate the university, and can get the results while they engage in their works in the industrial sites. Accordingly, all the interested parties have to anticipate the results of this research with the patience in long-standing point of view. Also, this researcher considers it as it is willing to give them the continuous interest and support.