• Progress in Medical Physics
• /
• v.25 no.3
• /
• pp.128-138
• /
• 2014

In gated radiation therapy (gRT), due to residual motion, beam delivery is intended to irradiate not only the true extent of disease, but also neighboring normal tissues. It is desired that the delivery covers the true extent (i.e. clinical target volume or CTV) as a minimum, although target moves under dose delivery. The objectives of our study are to validate if the intended dose is surely delivered to the true target in gRT and to quantitatively understand the trend of dose delivery on it and neighboring normal tissues when gating window (GW), motion amplitude (MA), and CTV size changes. To fulfill the objectives, experimental and computational studies have been designed and performed. A custom-made phantom with rectangle- and pyramid-shaped targets (CTVs) on a moving platform was scanned for four-dimensional imaging. Various GWs were selected and image integration was performed to generate targets (internal target volume or ITV) for planning that included the CTVs and internal margins (IM). The planning was done conventionally for the rectangle target and IMRT optimization was done for the pyramid target. Dose evaluation was then performed on a diode array aligned perpendicularly to the gated beams through measurements and computational modeling of dose delivery under motion. This study has quantitatively demonstrated and analytically interpreted the impact of residual motion including penumbral broadening for both targets, perturbed but secured dose coverage on the CTV, and significant doses delivered in the neighboring normal tissues. Dose volume histogram analyses also demonstrated and interpreted the trend of dose coverage: for ITV, it increased as GW or MA decreased or CTV size increased; for IM, it increased as GW or MA decreased; for the neighboring normal tissue, opposite trend to that of IM was observed. This study has provided a clear understanding on the impact of the residual motion and proved that if breathing is reproducible gRT is secure despite discontinuous delivery and target motion. The procedures and computational model can be used for commissioning, routine quality assurance, and patient-specific validation of gRT. More work needs to be done for patient-specific dose reconstruction on CT images.

• Progress in Medical Physics
• /
• v.19 no.2
• /
• pp.89-94
• /
• 2008

TomoTherapy has a merit to treat cancer with Intensity modulated radiation and combines precise 3-D imaging from computerized tomography (CT scanning) with highly targeted radiation beams and rotating beamlets. In this paper, we comparing the dose distribution between TomoTherapy and linear accelerator based intensity modulated radiotherapy (IMRT) for 10 Head & Neck patients using TomoTherapy which is newly installed and operated at National Cancer Center since Sept. 2006. Furthermore, we estimate how the homogeneity and Normal Tissue Complication Probability (NTCP) are changed by motion of target. Inverse planning was carried out using CadPlan planning system (CadPlan R.6.4.7, Varian Medical System Inc. 3100 Hansen Way, Palo Alto, CA 94304-1129, USA). For each patient, an inverse IMRT plan was also made using TomoTherapy Hi-Art System (Hi-Art2_2_4 2.2.4.15, TomoTherapy Incorporated, 1240 Deming Way, Madson, WI 53717-1954, USA) and using the same targets and optimization goals. All TomoTherapy plans compared favorably with the IMRT plans regarding sparing of the organs at risk and keeping an equivalent target dose homogeneity. Our results suggest that TomoTherapy is able to reduce the normal tissue complication probability (NTCP) further, keeping a similar target dose homogeneity.

• Journal of the Korean Society of Radiology
• /
• v.6 no.6
• /
• pp.455-464
• /
• 2012

This study suggested evaluation of approximately exposure to low-dose ionization radiation from medical images using a computed radiography (CR) system in standard X-ray examination and experimental model can compare diagnostic reference level (DRL) will suggest on optimization condition of guard about medical radiation of low dose space. Entrance surface dose (ESD) cross-measuring by standard dosimeter and optically stimulated luminescence dosimeters (OSLDs) in experiment condition about tube voltage and current of X-ray generator. Also, Hounsfield unit (HU) scale measured about each experiment condition in CR system and after character relationship table and graph tabulate about ESD and HU scale, approximately radiation dose about head, neck, thoracic, abdomen, and pelvis draw a measurement. In result measuring head, neck, thoracic, abdomen, and pelvis, average of ESD is 2.10, 2.01, 1.13, 2.97, and 1.95 mGy, respectively. HU scale is $3,276{\pm}3.72$, $3,217{\pm}2.93$, $2,768{\pm}3.13$, $3,782{\pm}5.19$, and $2,318{\pm}4.64$, respectively, in CR image. At this moment, using characteristic relationship table and graph, ESD measured approximately 2.16, 2.06, 1.19, 3.05, and 2.07 mGy, respectively. Average error of measuring value and ESD measured approximately smaller than 3%, this have credibility cover all the bases radiology area of measurement 5%. In its final analysis, this study suggest new experimental model approximately can assess radiation dose of patient in standard X-ray examination and can apply to CR examination, digital radiography and even film-cassette system.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.08a
• /
• pp.88-89
• /
• 2013

A variety of influenza A viruses from animal hosts are continuously prevalent throughout the world which cause human epidemics resulting millions of human infections and enormous industrial and economic damages. Thus, early diagnosis of such pathogen is of paramount importance for biomedical examination and public healthcare screening. To approach this issue, here we propose a fully integrated Rotary genetic analysis system, called Rotary Genetic Analyzer, for on-site detection of influenza A viruses with high speed. The Rotary Genetic Analyzer is made up of four parts including a disposable microchip, a servo motor for precise and high rate spinning of the chip, thermal blocks for temperature control, and a miniaturized optical fluorescence detector as shown Fig. 1. A thermal block made from duralumin is integrated with a film heater at the bottom and a resistance temperature detector (RTD) in the middle. For the efficient performance of RT-PCR, three thermal blocks are placed on the Rotary stage and the temperature of each block is corresponded to the thermal cycling, namely $95^{\circ}C$ (denature), $58^{\circ}C$ (annealing), and $72^{\circ}C$ (extension). Rotary RT-PCR was performed to amplify the target gene which was monitored by an optical fluorescent detector above the extension block. A disposable microdevice (10 cm diameter) consists of a solid-phase extraction based sample pretreatment unit, bead chamber, and 4 ${\mu}L$ of the PCR chamber as shown Fig. 2. The microchip is fabricated using a patterned polycarbonate (PC) sheet with 1 mm thickness and a PC film with 130 ${\mu}m$ thickness, which layers are thermally bonded at $138^{\circ}C$ using acetone vapour. Silicatreated microglass beads with 150~212 ${\mu}L$ diameter are introduced into the sample pretreatment chambers and held in place by weir structure for construction of solid-phase extraction system. Fig. 3 shows strobed images of sequential loading of three samples. Three samples were loaded into the reservoir simultaneously (Fig. 3A), then the influenza A H3N2 viral RNA sample was loaded at 5000 RPM for 10 sec (Fig. 3B). Washing buffer was followed at 5000 RPM for 5 min (Fig. 3C), and angular frequency was decreased to 100 RPM for siphon priming of PCR cocktail to the channel as shown in Figure 3D. Finally the PCR cocktail was loaded to the bead chamber at 2000 RPM for 10 sec, and then RPM was increased up to 5000 RPM for 1 min to obtain the as much as PCR cocktail containing the RNA template (Fig. 3E). In this system, the wastes from RNA samples and washing buffer were transported to the waste chamber, which is fully filled to the chamber with precise optimization. Then, the PCR cocktail was able to transport to the PCR chamber. Fig. 3F shows the final image of the sample pretreatment. PCR cocktail containing RNA template is successfully isolated from waste. To detect the influenza A H3N2 virus, the purified RNA with PCR cocktail in the PCR chamber was amplified by using performed the RNA capture on the proposed microdevice. The fluorescence images were described in Figure 4A at the 0, 40 cycles. The fluorescence signal (40 cycle) was drastically increased confirming the influenza A H3N2 virus. The real-time profiles were successfully obtained using the optical fluorescence detector as shown in Figure 4B. The Rotary PCR and off-chip PCR were compared with same amount of influenza A H3N2 virus. The Ct value of Rotary PCR was smaller than the off-chip PCR without contamination. The whole process of the sample pretreatment and RT-PCR could be accomplished in 30 min on the fully integrated Rotary Genetic Analyzer system. We have demonstrated a fully integrated and portable Rotary Genetic Analyzer for detection of the gene expression of influenza A virus, which has 'Sample-in-answer-out' capability including sample pretreatment, rotary amplification, and optical detection. Target gene amplification was real-time monitored using the integrated Rotary Genetic Analyzer system.

• Journal of the Korean Society of Radiology
• /
• v.13 no.6
• /
• pp.879-887
• /
• 2019

The purpose of this study is to establish a physiological injection protocol according to body weight, in order to minimize amount of contrast medium and optimize contrast enhancement in pediatric patients performing thoracic CT examinations. The 80 pediatric patients under the age of 10 were studied. Intravenous contrast material containing 300 mgI/ml was used. The group A injected with a capacity of 1.5 times its weight, and groups B, C and D added 5 to 15 ml of normal saline with a 10% decrease in each. The physiologic model which can be calculated by weight about amount of injection of contrast medium and normal saline, flow rate and delay time were applied. To assess image quality, measured average HU value and SNR of superior vena cava, pulmonary artery, ascending and descending aorta, right and left atrium, right and left ventricle. CT numbers of subclavian vein and superior vena cava were compared to identify the effects of reducing artifacts due to normal saline. Comparing SNR according to the contrast medium injection protocol, significant differences were found in superior vena cava and pulmonary artery, descending aorta, right and left ventricle, and CT numbers showed significant differences in all organs. In particular, B group with a 10% decrease in contrast medium and an additional injection of saline showed a low degree of contrast enhancement in groups with a decrease of more than 20%. In addition, the group injected with normal saline greatly reduced contrast enhancement of subclavian vein and superior vena cava, and the beam hardening artifact by contrast medium was significantly attenuated. In conclusion, the application of physiological protocol for injection of contrast medium in pediatric thoracic CT examinations was able to reduce artifacts by contrast medium, prevent unnecessary use of contrast medium and improve the effect of contrast enhancement.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.8
• /
• pp.29-37
• /
• 2019

Textile quality management targets the physical properties of fabrics and the subjective discriminations of color and fitting. Color is the most representative quality factor that consumers can use to evaluate quality levels without any instruments. For this reason, quantification using a color discrimination device has been used for statistical quality management in the textile industry. However, small and medium-sized domestic textile manufacturers use only visual inspection for color discrimination. As a result, color discrimination is different based on the inspectors' individual tendencies and work procedures. In this research, we want to develop a textile industry-friendly quality management method, evaluating the possibility of rapid color discrimination using a digital imaging device, which is one of the office-automation instruments. The results show that an imaging process-based color discrimination method is highly correlated with conventional color discrimination instruments ($R^2=0.969$), and is also applicable to field discrimination of the manufacturing process, or for different lots. Moreover, it is possible to recognize quality management factors by analyzing color components, ${\Delta}L$, ${\Delta}a$, ${\Delta}b$. We hope that our rapid discrimination method will be a substitute technique for conventional color discrimination instruments via elaboration and optimization.

• Journal of the Korean Society of Radiology
• /
• v.12 no.7
• /
• pp.895-908
• /
• 2018

As modern science is developed and advanced, examination and number of times using radiation are increasing daily. General diagnostic X-ray generator is installed on stationary form, But X-ray generator was developed because patient who is in the intensive care unit, operation room, emergency room can not move to general x-ray room. What we examine patient by x-ray generator is certainly necessary, So patient exposure is inevitable. but reducing radiation exposure is highly important matter about radiation technology, guardian, patient in the same hospital room, nurse etc. For this reason, rule regarding safety control of diagnostic x-ray generator revised for radiation worker, patient and protector proclaim that mobile diagnostic x-ray shield must placed in case of examine different location excluding operation room, emergency room, intensive care unit. But, radiogical technologist is having a lot of difficulties to examine with mobile x-ray generator, diagnostic x-ray shield partition, image plate and lead apron. So, when we use x-ray generator, we manufacture shield tools can be attached to the mobile x-ray generator On behalf of x-ray shield partition and conduct analysis and in comparison to part of body and distribution of dose rate and find way to reduce radiation exposure through distribution of dose rate of patient within the radiogical technologist, medical team. Mobile x-ray generator aimed at SHIMADZU inc. R-20, We manufactured equipment for shielding x-ray scattered x-ray by installing shielding wall from side to side based on support beam on the mobile x-ray generator. Shielding wall when moving can be folded and designed to expand when examine. Experiment measured five times in each by an angle for dose rate of eyes, thyroid, breast, abdomen and gonad on exposure condition of upper and lower extremity, chest, abdomen which is examined many times by mobile x-ray generator. We used dosimeter RSM-100 made by IJRAD and measured a horizontal dose rate by body part. The result of an experiment, shielding decreasing rate of the front and the rear showed 77 ~ 98.7%. Therefore using self-production shielding wall reduce scattered x-ray occurrence rate and confirm can decrease exposure dose consequently. Therefore, through this study, reduction result which is used shielding wall of self-production will be a role of shielding optimization and it could be answer about reduction of medical exposure recommended by ICRP 103.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.43-62
• /
• 2019

At one time, the anomaly detection sector dominated the method of determining whether there was an abnormality based on the statistics derived from specific data. This methodology was possible because the dimension of the data was simple in the past, so the classical statistical method could work effectively. However, as the characteristics of data have changed complexly in the era of big data, it has become more difficult to accurately analyze and predict the data that occurs throughout the industry in the conventional way. Therefore, SVM and Decision Tree based supervised learning algorithms were used. However, there is peculiarity that supervised learning based model can only accurately predict the test data, when the number of classes is equal to the number of normal classes and most of the data generated in the industry has unbalanced data class. Therefore, the predicted results are not always valid when supervised learning model is applied. In order to overcome these drawbacks, many studies now use the unsupervised learning-based model that is not influenced by class distribution, such as autoencoder or generative adversarial networks. In this paper, we propose a method to detect anomalies using generative adversarial networks. AnoGAN, introduced in the study of Thomas et al (2017), is a classification model that performs abnormal detection of medical images. It was composed of a Convolution Neural Net and was used in the field of detection. On the other hand, sequencing data abnormality detection using generative adversarial network is a lack of research papers compared to image data. Of course, in Li et al (2018), a study by Li et al (LSTM), a type of recurrent neural network, has proposed a model to classify the abnormities of numerical sequence data, but it has not been used for categorical sequence data, as well as feature matching method applied by salans et al.(2016). So it suggests that there are a number of studies to be tried on in the ideal classification of sequence data through a generative adversarial Network. In order to learn the sequence data, the structure of the generative adversarial networks is composed of LSTM, and the 2 stacked-LSTM of the generator is composed of 32-dim hidden unit layers and 64-dim hidden unit layers. The LSTM of the discriminator consists of 64-dim hidden unit layer were used. In the process of deriving abnormal scores from existing paper of Anomaly Detection for Sequence data, entropy values of probability of actual data are used in the process of deriving abnormal scores. but in this paper, as mentioned earlier, abnormal scores have been derived by using feature matching techniques. In addition, the process of optimizing latent variables was designed with LSTM to improve model performance. The modified form of generative adversarial model was more accurate in all experiments than the autoencoder in terms of precision and was approximately 7% higher in accuracy. In terms of Robustness, Generative adversarial networks also performed better than autoencoder. Because generative adversarial networks can learn data distribution from real categorical sequence data, Unaffected by a single normal data. But autoencoder is not. Result of Robustness test showed that he accuracy of the autocoder was 92%, the accuracy of the hostile neural network was 96%, and in terms of sensitivity, the autocoder was 40% and the hostile neural network was 51%. In this paper, experiments have also been conducted to show how much performance changes due to differences in the optimization structure of potential variables. As a result, the level of 1% was improved in terms of sensitivity. These results suggest that it presented a new perspective on optimizing latent variable that were relatively insignificant.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.