• Korean Journal of Radiology
• /
• v.21 no.1
• /
• pp.88-100
• /
• 2020

Objective: We aimed to develop and validate a deep learning system for fully automated segmentation of abdominal muscle and fat areas on computed tomography (CT) images. Materials and Methods: A fully convolutional network-based segmentation system was developed using a training dataset of 883 CT scans from 467 subjects. Axial CT images obtained at the inferior endplate level of the 3rd lumbar vertebra were used for the analysis. Manually drawn segmentation maps of the skeletal muscle, visceral fat, and subcutaneous fat were created to serve as ground truth data. The performance of the fully convolutional network-based segmentation system was evaluated using the Dice similarity coefficient and cross-sectional area error, for both a separate internal validation dataset (426 CT scans from 308 subjects) and an external validation dataset (171 CT scans from 171 subjects from two outside hospitals). Results: The mean Dice similarity coefficients for muscle, subcutaneous fat, and visceral fat were high for both the internal (0.96, 0.97, and 0.97, respectively) and external (0.97, 0.97, and 0.97, respectively) validation datasets, while the mean cross-sectional area errors for muscle, subcutaneous fat, and visceral fat were low for both internal (2.1%, 3.8%, and 1.8%, respectively) and external (2.7%, 4.6%, and 2.3%, respectively) validation datasets. Conclusion: The fully convolutional network-based segmentation system exhibited high performance and accuracy in the automatic segmentation of abdominal muscle and fat on CT images.

• The Transactions of the Korea Information Processing Society
• /
• v.1 no.1
• /
• pp.114-128
• /
• 1994

It is very difficult for a nonprofessional programmer in Koera to write a program with very High Level Language such as, V,REFINE, GIST, and SETL, because the semantic primitives of these languages are based on predicate calculus, set, mapping, or testricted natural language. And it takes time to be familiar with these language. In this paper, we suggest a method to reduce such difficulties by programming with the declarative, procedural constructs, and aggregate constructs. And we design and implement an experimental knowledge-based automatic programming system. called HAPS(Hangul Automatic Program System). HAPS, whose input is specification such as Hangul abstract algorithm and datatype or Hangul procedural constructs, and whose output is C program. The method of operation is based on rule-based and program transformation technique, and the problem transformation technique. The problem area is general problem. The control structure of HAPS accepts the program specification, transforms this specification according to the proper rule in the rule-base, and stores the transformed program specification on the global data base. HAPS repeats these procedures until the target C program is fully constructed.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.75-81
• /
• 2010

The segmentation of cartilage is crucial for the diagnose and treatment of osteoarthritis (OA), and has mostly been done manually by an expert, requiring a considerable amount of time and effort due to the thin shape and vague boundaries of the cartilage in MR (magnetic resonance) images. In this paper, we propose a fully automatic method to segment cartilage in a knee joint on MR images. The proposed method is based on a small number of manually segmented images as the training set and comprised of an initial per patch segmentation process and a global refinement process on the cumulative per patch results. Each patch for per patch segmentation is positioned by classifying the bone-cartilage interface on the pre-segmented bone surface. Next, the shape and intensity priors are constructed for each patch based on information extracted from reference patches in the training set. The ratio of influence between the shape and intensity priors is adaptively determined per patch. Each patch is segmented by graph cuts, where energy is defined based on constructed priors. Finally, global refinement is conducted on the global cartilage using the results of per patch segmentation as the shape prior. Experimental evaluation shows that the proposed framework provide accurate and clinically useful segmentation results.

• Korean Journal of Radiology
• /
• v.22 no.11
• /
• pp.1764-1776
• /
• 2021

Objective: This study aimed to validate a deep learning-based fully automatic calcium scoring (coronary artery calcium [CAC]_auto) system using previously published cardiac computed tomography (CT) cohort data with the manually segmented coronary calcium scoring (CAC_hand) system as the reference standard. Materials and Methods: We developed the CAC_auto system using 100 co-registered, non-enhanced and contrast-enhanced CT scans. For the validation of the CAC_auto system, three previously published CT cohorts (n = 2985) were chosen to represent different clinical scenarios (i.e., 2647 asymptomatic, 220 symptomatic, 118 valve disease) and four CT models. The performance of the CAC_auto system in detecting coronary calcium was determined. The reliability of the system in measuring the Agatston score as compared with CAC_hand was also evaluated per vessel and per patient using intraclass correlation coefficients (ICCs) and Bland-Altman analysis. The agreement between CAC_auto and CAC_hand based on the cardiovascular risk stratification categories (Agatston score: 0, 1-10, 11-100, 101-400, > 400) was evaluated. Results: In 2985 patients, 6218 coronary calcium lesions were identified using CAC_hand. The per-lesion sensitivity and false-positive rate of the CAC_auto system in detecting coronary calcium were 93.3% (5800 of 6218) and 0.11 false-positive lesions per patient, respectively. The CAC_auto system, in measuring the Agatston score, yielded ICCs of 0.99 for all the vessels (left main 0.91, left anterior descending 0.99, left circumflex 0.96, right coronary 0.99). The limits of agreement between CAC_auto and CAC_hand were 1.6 ± 52.2. The linearly weighted kappa value for the Agatston score categorization was 0.94. The main causes of false-positive results were image noise (29.1%, 97/333 lesions), aortic wall calcification (25.5%, 85/333 lesions), and pericardial calcification (24.3%, 81/333 lesions). Conclusion: The atlas-based CAC_auto empowered by deep learning provided accurate calcium score measurement as compared with manual method and risk category classification, which could potentially streamline CAC imaging workflows.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.9
• /
• pp.57-64
• /
• 2021

In this paper, we propose a device that can control the primary control surface of an aircraft by recognizing speech commands. The speech command consists of 19 commands, and a learning model is constructed based on a total of 2,500 datasets. The training model is composed of a CNN model using the Sequential library of the TensorFlow-based Keras model, and the speech file used for training uses the MFCC algorithm to extract features. The learning model consists of two convolution layers for feature recognition and Fully Connected Layer for classification consists of two dense layers. The accuracy of the validation dataset was 98.4%, and the performance evaluation of the test dataset showed an accuracy of 97.6%. In addition, it was confirmed that the operation was performed normally by designing and implementing a Raspberry Pi-based control device. In the future, it can be used as a virtual training environment in the field of voice recognition automatic flight and aviation maintenance.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.14 no.2
• /
• pp.38-43
• /
• 2015

Tungsten inert gas (TIG) welding is commonly used in industries that require airtightness, watertightness, oiltightness, and precision. It is a non-consumable welding method that is commonly used for the welding of non-ferrous metals, but it can be used to weld most metals. The methods of TIG welding can be divided into three types. The first, manual welding is done directly on the metal by a welder with a torch. The second, semi-automatic welding, gets help from a material supplying machine, but it is conducted by a welder. Lastly, automated welding is conducted fully by a machine during its process and operation. Depending on the selection of electrode, the amount of heat that is applied to the base material and the electrode rod changes and makes the shape of welded parts different. A direct-current positive electrode was used for this study. Through the change of shielding gas type on a structural steel (SS-400) that is commonly used in industry, the composition and shape changes in welded parts were detected after welding. The heat-affected area, hardness value, and tensile strength were also identified through hardness testing and tensile testing. In this study, it was found that the higher hardness value of the heat-affected is, the weaker the tensile strength becomes.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1845-1848
• /
• 2005

The automatic $CO_2$ welding is a manufacturing process to produce high quality joints for metal and it could provide a capability of full automation to enhance productivity. Despite the widespread use in the various manufacturing industries, the full automation of the robotic $CO_2$ welding has not yet been achieved partly because the mathematical model for the process parameters of a given welding task is not fully understood and quantified. Several mathematical models to control welding quality, productivity, microstructure and weld properties in arc welding processes have been studied. However, it is not an easy task to apply them to the various practical situations because the relationship between the process parameters and the bead geometry is non-linear and also they are usually dependent on the specific experimental results. Practically, it is difficult, but important to know how to establish a mathematical model that can predict the result of the actual welding process and how to select the optimum welding condition under a certain constraint. In this research, an attempt has been made to develop an intelligent algorithm to predict the weld geometry (top-bead width, top-bead height, back-bead width and back-bead height) as a function of key process parameters in the robotic $CO_2$welding. A sensitivity analysis has been conducted and compared the relative impact of three process parameters on bead geometry in order to verify the measurement errors on the values of the uncertainty in estimated parameters.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1993.10a
• /
• pp.1243-1253
• /
• 1993

Visual features of a mushroom(Lentinus Edodes L) are critical in sorting and grading as most agricultural products are. Because of its complex and various visual features, grading and sorting of mushrooms have been done manually by the human expert. Though actions involved in human grading looks simple, a decision making undereath the simple action comes form the results of the complex neural processing of the visual image. And processing details involved in the visual recognition of the human brain has not been fully investigated yet. Recently, however, an artificial neural network has drawn a great attention because of its functional capability as a partial substitute of the human brain. Since most agricultural products are not uniquely defined in its physical properties and do not have a well defined job structure, a research of the neuro-net based human like information processing toward the agricultural product and processing are widely open and promising. In this pape , neuro-net based grading and sorting system was developed for a mushroom . A computer vision system was utilized for extracting and quantifying the qualitative visual features of sampled mushrooms. The extracted visual features and their corresponding grades were used as input/output pairs for training the neural network and the trained results of the network were presented . The computer vision system used is composed of the IBM PC compatible 386DX, ITEX PFG frame grabber, B/W CCD camera , VGA color graphic monitor , and image output RGB monitor.

• The KIPS Transactions:PartB
• /
• v.9B no.2
• /
• pp.215-222
• /
• 2002

This paper presents a reliable fully automatic labeling system which fits well with languages having well-developed syllables such as in Korean. The ASL System utilize DAC (Divide and Conquer), a control mechanism, based segmentation algorithm to use phonetic and acoustic information with greater efficiency. The segmentation algorithm is to devide speech signals into speechlets which is localized speech signal pieces and to segment each speechlet for speech boundaries. While HMM method has uniform and definite efficiencies, the suggested method gives framework to steadily develope and improve specified acoustic knowledges as a component. Without using statistical method such as HMM, this new method use only phonetic-acoustic information. Therefore, this method has high speed performance, is consistent extending the specific acoustic knowledge component, and can be applied in efficient way. we show experiment result to verify suggested method at the end.

• The Journal of Korean Association of Computer Education
• /
• v.6 no.2
• /
• pp.9-21
• /
• 2003

This paper proposes an automatic construction system for transliteration dictionary from English-Korean parallel corpus. The system works in 3 steps: it extracts all nouns from Korean documents as the first step, filters transliterated foreign word nouns out of them with the language identification method as the second step, and extracts the corresponding English words by using a probabilistic alignment method as the final step. Specially, the fact that there is a corresponding English word in most cases, is utilized to extract the purely transliterated part from a Koreans word phrase, which is usually used in combined forms with Korean endings(Eomi) or particles(Josa). Moreover, the direct phonetic comparison is done to the words in two different alphabet systems without converting them to the same alphabet system. The experiment showed that the performance was influenced by the first and the second preprocessing steps; the most efficient model among manually preprocessed ones showed 85.4% recall, 91.0% precision and the most efficient model among fully automated ones got 68.3% recall, 89.2% precision.