• 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.

• Nuclear Medicine and Molecular Imaging
• /
• v.40 no.6
• /
• pp.309-315
• /
• 2006

Purpose: Accurate evaluation of cervical lymph node (LN) metastasis of head and neck squamous cell canter (SCC) is important to treatment planning. We evaluated the diagnostic accuracy of F-18 fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) for the detection of cervical LN metastasis of head and neck SCC and performed a retrospective comparison with CT/MRI findings. Materials & Methods: Seventeen patients with pathologically proven head and neck SCC underwent F-18 FDG PET/CT and CT/MRI within 4 week before surgery. We recorded lymph node metastases according to the neck level system of imaging-based nodal classification. F-18 FDG PET/CT images were analyzed visually for assessment of regional tracer uptake in LN. We analyzed the differences in sensitivity and specificity between F-18 FDG PET/CT and CT/MRI using the Chi-square test. Results: Among the 17 patients, a total of 123 LN levels were dissected, 29 of which showed metastatic involvement. The sensitivity and specificity of F-18 FDG PET/CT for detecting cervical LN metastasis on a level-by-level basis were 69% (20/29) and 99% (93/94). The sensitivity and specificity of CT/MRI were 62% (18/29) and 96% (90/94). There was no significant difference in diagnostic accuracy between F-18 FDG PET/CT and CT/MRI. Interestingly, F-18 FDG PET/CT detected double primary tumor (hepatocellular carcinoma) and rib metastasis, respectively. Conclusion: There was not statistically significant difference of diagnostic accuracy between F-18 FDG PET/CT and CT/MRI for the detection of cervical LN metastasis of head and neck SCC. The low sensitivity of F-18 FDG PET/CT was due to limited resolution for small metastatic deposits.

• The Korean Journal of Nuclear Medicine
• /
• v.33 no.6
• /
• pp.466-474
• /
• 1999

Purpose: The purpose of this study was to evaluate the diagnostic accuracy of [$^{18}F$]FDG PET in the diagnosis of recurrent head and neck cancer after the completion of surgery and radiotherapy in patients with head and neck cancers. Materials and Methods: In fifty-nine patients with head and neck cancers whole body [$^{18}F$]FDG PET studies were performed. According to the different therapeutic modalities, patients were divided into four groups (Group I; pre-treatment, Group II: surgery, Group III; radiotherapy, Group IV; both surgery and radiotherapy). [$^{18}F$]FDG PET images were compared with clinical, CT and histopathologic findings. Results: for detection of metastatic lymph nodes in 14 patients of pre-treatment group (group I), the sensitivity and specificity of PET were 100% (10/10) and 75% (3/4), and those of CT were 80% (8/10) and 100% (4/4). For detection of recurrence in 45 patients of post-treatment group, overall sensitivity and specificity of PET were 96.2% (25/26) and 78.9% (15/19) [(100% and 75% in group II, 80% and 10% in group III, and 100% and 100% in group IV)] without significant difference from pre-treatment group (P>0.1). In detecting recurrence, the sensitivity and specificity of [$^{18}F$]FDG PET were 90.9% (10/11) and 20% (1/5) in 16 patients who underwent [$^{18}F$]FDG PET within 2 months after the completion of treatment. The specificity of these patients was significantly lower than that of 29 patients (100% of sensitivity and specificity) who underwent [$^{18}F$]FDG PET 2 months after treatment (p<0.05). Conclusion: [$^{18}F$]FDG PET is an accurate diagnostic modality for the detection of recurrence in head and neck cancer. Post-therapy [$^{18}F$]FDG PET should be obtained at least 2 months after the completion of surgery or radiotherapy.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.7
• /
• pp.271-278
• /
• 2021

As the current cyber attacks become more intelligent, the existing Intrusion Detection System is difficult for detecting intelligent attacks that deviate from the existing stored patterns. In an attempt to solve this, a model of a deep learning-based intrusion detection system that analyzes the pattern of intelligent attacks through data learning has emerged. Intrusion detection systems are divided into host-based and network-based depending on the installation location. Unlike network-based intrusion detection systems, host-based intrusion detection systems have the disadvantage of having to observe the inside and outside of the system as a whole. However, it has the advantage of being able to detect intrusions that cannot be detected by a network-based intrusion detection system. Therefore, in this study, we conducted a study on a host-based intrusion detection system. In order to evaluate and improve the performance of the host-based intrusion detection system model, we used the host-based Leipzig Intrusion Detection-Data Set (LID-DS) published in 2018. In the performance evaluation of the model using that data set, in order to confirm the similarity of each data and reconstructed to identify whether it is normal data or abnormal data, 1D vector data is converted to 3D image data. Also, the deep learning model has the drawback of having to re-learn every time a new cyber attack method is seen. In other words, it is not efficient because it takes a long time to learn a large amount of data. To solve this problem, this paper proposes the Siamese Convolutional Neural Network (Siamese-CNN) to use the Few-Shot Learning method that shows excellent performance by learning the little amount of data. Siamese-CNN determines whether the attacks are of the same type by the similarity score of each sample of cyber attacks converted into images. The accuracy was calculated using Few-Shot Learning technique, and the performance of Vanilla Convolutional Neural Network (Vanilla-CNN) and Siamese-CNN was compared to confirm the performance of Siamese-CNN. As a result of measuring Accuracy, Precision, Recall and F1-Score index, it was confirmed that the recall of the Siamese-CNN model proposed in this study was increased by about 6% from the Vanilla-CNN model.

• Journal of Radiation Protection and Research
• /
• v.38 no.4
• /
• pp.214-227
• /
• 2013

Computed tomographic scan as a screening procedures in asymptomatic individuals has seen a steady increase with the introduction of multiple-raw detector CT scanners. This report provides a brief review of the current controversy surrounding CT cancer screening, with a focus on the radiation induced cancer risks and clinical efficacy. 1. A large study of patients at high risk of lung cancer(the National Lung Screening Trial[NLST]) showed that CT screening reduced cancer deaths by 20%(1.33% in those screened compared with 1.67% in those not screened). The rate of positive screening tests was 24.2% and 96.4% of the positive screening results in the low-dose CT group were false-positive. Radiation induced lung cancer risk was estimated the most important in screening population because ERR of radiation induced lung cancer does not show the decrease with increasing age and synergistic connection between smoking and radiation risk. Therefore, the radiation risk may be on the same order of magnitude as the benefit observed in the NLST. Optimal screening strategy remain uncertain, CT lung cancer screening is not yet ready for implementation. 2. Computed tomographic colonography is as good as colonoscopy for detecting colon cancer and is almost as good as colonoscopy for detecting advanced adenomas, but significantly less sensitive and specific for smaller lesions and disadvantageous for subsequent therapeutic optical colonoscopy if polyps are detected. The average effective dose from CT colonography was estimated 8-10 $mS{\nu}$, which could be a significant dose if administered routinely within the population over many years. CT colonography should a) achieve at least 90% sensitivity and specificity in the size category from 6 and 10 mm, b) offer non-cathartic bowl preparation and c) be optimized and standardized CT parameters if it is to be used for mass screening. 3. There is little evidence that demonstrates, for whole-body scanning, the benefit outweighs the detriment. This test found large portion of patient(86~90.8%) had at least one abnormal finding, whereas only 2% were estimated to have clinically significant disease. Annual scans from ages 45 to 75 years would accrue an estimated lifetime cancer mortality risk of 1.9%. There is no group within the medical community that recommends whole-body CT. No good studies indicate the accuracy of screening CT, at this time. The benefit/risk balance for any of the commonly suggested CT screening techniques has yet to be established. These areas need further research. Therefore wild screening should be avoided.

• Geophysics and Geophysical Exploration
• /
• v.10 no.4
• /
• pp.383-392
• /
• 2007

Karstic features and mining-related cavities not only lead to severe restrictions in land utilizations, but also constitute serious concern about geohazard and groundwater contamination. A microgravity survey was applied for detecting, mapping and monitoring karstic cavities in the test site at Muan prepared by KIGAM. The gravity data were collected using an AutoGrav CG-3 gravimeter at about 800 stations by 5 m interval along paddy paths. The density distribution beneath the profiles was drawn by two dimensional inversion based on the minimum support stabilizing functional, which generated better focused images of density discontinuities. We also imaged three dimensional density distribution by growing body inversion with solution from Euler deconvolution as a priori information. The density image showed that the cavities were dissolved, enlarged and connected into a cavity network system, which was supported by drill hole logs. A time-lapse microgravity was executed on the road in the test site for monitoring the change of the subsurface density distribution before and after grouting. The data were adjusted for reducing the effects due to the different condition of each survey, and inverted to density distributions. They show the change of density structure during the lapsed time, which implies the effects of grouting. This case history at the Muan test site showed that the microgravity with accuracy and precision of ${\mu}Gal$ is an effective and practical tool for detecting, mapping and monitoring the subsurface cavities.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.125-141
• /
• 2012

These days, the malicious attacks and hacks on the networked systems are dramatically increasing, and the patterns of them are changing rapidly. Consequently, it becomes more important to appropriately handle these malicious attacks and hacks, and there exist sufficient interests and demand in effective network security systems just like intrusion detection systems. Intrusion detection systems are the network security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. Conventional intrusion detection systems have generally been designed using the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. However, they cannot handle new or unknown patterns of the network attacks, although they perform very well under the normal situation. As a result, recent studies on intrusion detection systems use artificial intelligence techniques, which can proactively respond to the unknown threats. For a long time, researchers have adopted and tested various kinds of artificial intelligence techniques such as artificial neural networks, decision trees, and support vector machines to detect intrusions on the network. However, most of them have just applied these techniques singularly, even though combining the techniques may lead to better detection. With this reason, we propose a new integrated model for intrusion detection. Our model is designed to combine prediction results of four different binary classification models-logistic regression (LOGIT), decision trees (DT), artificial neural networks (ANN), and support vector machines (SVM), which may be complementary to each other. As a tool for finding optimal combining weights, genetic algorithms (GA) are used. Our proposed model is designed to be built in two steps. At the first step, the optimal integration model whose prediction error (i.e. erroneous classification rate) is the least is generated. After that, in the second step, it explores the optimal classification threshold for determining intrusions, which minimizes the total misclassification cost. To calculate the total misclassification cost of intrusion detection system, we need to understand its asymmetric error cost scheme. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, total misclassification cost is more affected by FNE rather than FPE. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 10,000 samples from them by using random sampling method. Also, we compared the results from our model with the results from single techniques to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell R4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on GA outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that the proposed model outperformed all the other comparative models in the total misclassification cost perspective. Consequently, it is expected that our study may contribute to build cost-effective intelligent intrusion detection systems.

• The Korean Journal of Nuclear Medicine
• /
• v.36 no.4
• /
• pp.232-243
• /
• 2002

Purpose: We studied whether brain perfusion SPECT is useful in the psychiatric disability evaluation of patients with chronic traumatic brain injury (TBI). Materials and Methods: Sixty-nine patients (M:F=58:11, age $39{\pm}14$ years) who underwent Tc-99m HMPAO brain SPECT, brain MRI and neuropsychological (NP) tests during hospitalization in psychiatric wards for the psychiatric disability evaluation were included; the severity of injury was mild in 31, moderate in 17 and severe in 21. SPECT, MRI, NP tests were peformed $6{\sim}61$ months (mean 23 months) post-injury. Diagnostic accuracy of SPECT and MRI to show hypoperfusion or abnormal signal intensity in patients with cognitive impairment represented by NP test results were compared. Results: Forty-two patients were considered to have cognitive impairment on NP tests and 27 not. Brain SPECT showed 71% sensitivity and 85% specificity, while brain MRI showed 62% sensitivity and 93% specificity (p>0.05, McNemar test). SPECT found more cortical lesions and MRI was superior in detecting white matter lesions. Sensitivity and specificity of 31 mild TBI patients were 45%, 90% for SPECT and 27%, 100% for MRI (p>0.05, McNemar test). Among 41 patients with normal brain MRI, SPECT showed 63% sensitivity (50% for mild TBI) and 88% specificity (85% for malingerers). Conclusion: Brain SPECT has a supplementary role to neuropsychological tests in the psychiatric disability evaluation of chronic TBI patients by detecting more cortical lesions than MRI.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.3
• /
• pp.82-98
• /
• 2019

The importance of spatial information is rapidly rising. In particular, 3D spatial information construction and modeling for Real World Objects, such as smart cities and digital twins, has become an important core technology. The constructed 3D spatial information is used in various fields such as land management, landscape analysis, environment and welfare service. Three-dimensional modeling with image has the hig visibility and reality of objects by generating texturing. However, some texturing might have occlusion area inevitably generated due to physical deposits such as roadside trees, adjacent objects, vehicles, banners, etc. at the time of acquiring image Such occlusion area is a major cause of the deterioration of reality and accuracy of the constructed 3D modeling. Various studies have been conducted to solve the occlusion area. Recently the researches of deep learning algorithm have been conducted for detecting and resolving the occlusion area. For deep learning algorithm, sufficient training data is required, and the collected training data quality directly affects the performance and the result of the deep learning. Therefore, this study analyzed the ability of detecting the occlusion area of the image using various image quality to verify the performance and the result of deep learning according to the quality of the learning data. An image containing an object that causes occlusion is generated for each artificial and quantified image quality and applied to the implemented deep learning algorithm. The study found that the image quality for adjusting brightness was lower at 0.56 detection ratio for brighter images and that the image quality for pixel size and artificial noise control decreased rapidly from images adjusted from the main image to the middle level. In the F-measure performance evaluation method, the change in noise-controlled image resolution was the highest at 0.53 points. The ability to detect occlusion zones by image quality will be used as a valuable criterion for actual application of deep learning in the future. In the acquiring image, it is expected to contribute a lot to the practical application of deep learning by providing a certain level of image acquisition.

• Journal of the Korean Society of Radiology
• /
• v.83 no.5
• /
• pp.1090-1103
• /
• 2022

Purpose To evaluate the diagnostic performance of digital breast tomosynthesis (DBT) with the two-dimensional synthesized mammogram (2DSM), compared to full-field digital mammography (FFDM), for suspicious microcalcifications in the breast ahead of stereotactic biopsy and to assess the diagnostic image visibility of the images. Materials and Methods This retrospective study involved 189 patients with microcalcifications, which were histopathologically verified by stereotactic breast biopsy, who underwent DBT with 2DSM and FFDM between January 8, 2015, and January 20, 2020. Two radiologists assessed all cases of microcalcifications based on Breast Imaging Reporting and Data System (BI-RADS) independently. They were blinded to the histopathologic outcome and additionally evaluated lesion visibility using a fivepoint scoring scale. Results Overall, the inter-observer agreement was excellent (0.9559). Under the setting of category 4A as negative due to the low possibility of malignancy and to avoid the dilution of malignancy criteria in our study, McNemar tests confirmed no significant difference between the performances of the two modalities in detecting microcalcifications with a high potential for malignancy (4B, 4C, or 5; p = 0.1573); however, the tests showed a significant difference between their performances in detecting microcalcifications with a high potential for benignancy (4A; p = 0.0009). DBT with 2DSM demonstrated superior visibility and diagnostic performance than FFDM in dense breasts. Conclusion DBT with 2DSM is superior to FFDM in terms of total diagnostic accuracy and lesion visibility for benign microcalcifications in dense breasts. This study suggests a promising role for DBT with 2DSM as an accommodating tool for stereotactic biopsy in female with dense breasts and suspicious breast microcalcifications.