• Design & Manufacturing
• /
• v.15 no.3
• /
• pp.1-6
• /
• 2021

An artificial neural network model based on a deep learning algorithm is known to be more accurate than humans in image classification, but there is still a limit in the sense that there needs to be a lot of training data that can be called big data. Therefore, various techniques are being studied to build an artificial neural network model with high precision, even with small data. The transfer learning technique is assessed as an excellent alternative. As a result, the purpose of this study is to develop an artificial neural network system that can classify burr images of light guide plate products with 99% accuracy using transfer learning technique. Specifically, for the light guide plate product, 150 images of the normal product and the burr were taken at various angles, heights, positions, etc., respectively. Then, after the preprocessing of images such as thresholding and image augmentation, for a total of 3,300 images were generated. 2,970 images were separated for training, while the remaining 330 images were separated for model accuracy testing. For the transfer learning, a base model was developed using the NASNet-Large model that pre-trained 14 million ImageNet data. According to the final model accuracy test, the 99% accuracy in the image classification for training and test images was confirmed. Consequently, based on the results of this study, it is expected to help develop an integrated AI production management system by training not only the burr but also various defective images.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.302-310
• /
• 2022

Efficient use of limited blood products is becoming very important in terms of socioeconomic status and patient recovery. To predict the appropriateness of patient-specific transfusions for the intensive care unit (ICU) patients who require real-time monitoring, we evaluated a model to predict the possibility of transfusion dynamically by using the Medical Information Mart for Intensive Care III (MIMIC-III), an ICU admission record at Harvard Medical School. In this study, we developed an explainable machine learning to predict the possibility of red blood cell transfusion for major medical diseases in the ICU. Target disease groups that received packed red blood cell transfusions at high frequency were selected and 16,222 patients were finally extracted. The prediction model achieved an area under the ROC curve of 0.9070 and an F1-score of 0.8166 (LightGBM). To explain the performance of the machine learning model, feature importance analysis and a partial dependence plot were used. The results of our study can be used as basic data for recommendations related to the adequacy of blood transfusions and are expected to ultimately contribute to the recovery of patients and prevention of excessive consumption of blood products.

• Asia pacific journal of information systems
• /
• v.30 no.2
• /
• pp.439-456
• /
• 2020

In online shopping, "out-of-stock" and "sold-out" are used to indicate product unavailability, and this unavailability and its effects on consumers' behaviors have been studied with great interest for practical purposes. However, few studies have specifically discussed out-of-stock and sold-out products in the same paper. We hypothesized that consumers might cognitively interpret items marked out-of-stock and sold-out differently, and in this paper, we studied these potential differences from the perspectives of consumers' emotions, behaviors, and loyalty based on the stimulus-organism-response framework. In order to explore the differences, we used a multi-method approach that consisted of experiments, surveys, and interviews. Specifically, we built an experimental website on which the same products were categorized as either out-of-stock or sold-out, and we measured the participants' emotions, attitudes, and intentions after the experiment. After two weeks, we conducted interviews to confirm our results and to learn more about consumers' everyday behavior. In the results, males and females demonstrated differences in emotion, behaviors, and loyalty with the interaction effects of an item's being marked out-of-stock versus sold-out. We found that the consumers demonstrated different levels of loyalty based on whether the item was marked out-of-stock or sold-out. We discuss the strategic implications of our findings.

• Food Science of Animal Resources
• /
• v.28 no.3
• /
• pp.319-326
• /
• 2008

This study evaluated the effects of adding levels of rice bran fiber on the chemical compositions, cooking characteristics and sensory properties of ground pork meat products. Meat products were produced with products containing 0% (control), 1%, 2%, 3%, 4%, and 5% and rice bran fiber. The control had the highest protein and fat contents, cooking loss, reduction in diameter, CIE L value and CIE a value of uncooked product, and CIE L value of cooked product. The meat product containing rice bran fiber had the higher ash, pH, and CIE b value than the controls. The addition of 5% rice bran fiber resulted in the lowest cooking loss and L value of cooked meat products. Meat product with 2% rice bran had the highest water content, water holding capacity, hardness, color, tenderness, juiciness values. There was a significant difference among the ground meat products with respect to sensory properties, and meat products containing 1% and 2% rice bran had higher overall acceptability than the other meat products.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.10
• /
• pp.239-246
• /
• 2020

In the medical field where artificial intelligence technology is introduced, research related to clinical decision support system(CDSS) in relation to diagnosis and prediction is actively being conducted. In particular, medical imaging-based disease diagnosis area applied AI technologies at various products. However, medical imaging data consists of inconsistent data, and it is a reality that it takes considerable time to prepare and use it for research. This paper describes a one-stop AI learning platform for converting to medical image standard R_CDM(Radiology Common Data Model) and supporting AI algorithm development research based on the dataset. To this, the focus is on linking with the existing CDM(common data model) and model the system, including the schema of the medical imaging standard model and report information for multi-center research based on DICOM(Digital Imaging and Communications in Medicine) tag information. And also, we show the execution results based on generated datasets through the AI learning platform. As a proposed platform, it is expected to be used for various image-based artificial intelligence researches.

• Journal of Nutrition and Health
• /
• v.56 no.5
• /
• pp.537-553
• /
• 2023

Purpose: The purpose of this study was to evaluate the nutritional quality of complementary baby food products sold in Korea according to the baby food stages and food composition. Methods: A total of 1,587 complementary food products sold online and offline between March and December 2021 were investigated. They ranged from liquid meals to solid rice for babies aged 5 to 36 months. Results: The number of intakes per packaged volume was 2.8 in Stage 1, 1.9 in Stage 2, 1.4 in Stage 3, and 1.1 in Stage 4 (p < 0.0001). The dietary variety scores (DVS) of the complementary baby food products were 3.4 in Stage 1, 5.5 in Stage 2, 7.1 in Stage 3, and 9.7 in Stage 4 (p < 0.0001) and showed a significant increase in the later stages. The Korean dietary diversity score (KDDS) significantly increased from 2.3 in Stage 1, to 2.8 in Stage 2, 3.0 in Stage 3, and 3.4 in Stage 4 (p < 0.0001). The higher the baby food stage, the higher the proportion of grains/meat/vegetable ingredients. The ratio of protein intake to Adequate Intake (AI) or Recommended Nutrition Intake (RNI) was higher in products with a KDDS of 3 points or more, or in products with 2 points or fewer in Stages 1 and 2 (p < 0.0001, respectively). The ratio of protein intake to RNI increased as the KDDS score increased in Stages 3 and 4 (p < 0.0001, respectively). For all stages of baby foods, the ratio of protein intake to AI or RNI was high in products that included the meat group (beans, nuts, meat, eggs, fish, and shellfish) (p < 0.0001, respectively). Conclusion: Continuous research and nutritional evaluation are required for establishing nutrient content standards for commercially available baby foods, considering breast milk intake.

• Nutrition Research and Practice
• /
• v.5 no.5
• /
• pp.443-449
• /
• 2011

Instant noodles are widely consumed in Asian countries. The Korean population consumed the largest quantity of instant noodles in the world in 2008, However, few studies have investigated the relationship between instant noodles and nutritional status in Koreans. The objective of this study was to examine the association between instant noodle consumption and food and nutrient intake in Korean adults. We used dietary data of 6,440 subjects aged 20 years and older who participated in the Korean National Health and Nutrition Examination Survey III. The average age of the instant noodle consumers (INC) was 36.2 and that of the non-instant noodle consumers (non-INC) was 44.9; men consumed more instant noodles than women (P<0.001), With the exception of cereals and grain products, legumes, seaweeds, eggs, and milk and dairy products, INC consumed significantly fewer potatoes and starches, sugars, seeds and nuts, vegetables, mushrooms, fruits, seasonings, beverages, meals, fishes, and oils and fats compared with those in the non-INC group. The INC group showed significantly higher nutrient intake of energy, fat, sodium, thiamine, and riboflavin; however, the INC group showed a significantly lower intake of protein, calcium, phosphorus, iron, potassium, vitamin A, niacin, and vitamin C compared with those in the non-INC group. This study revealed that consuming instant noodles may lead to excessive intake of energy, fats, and sodium but may also cause increased intake of thiamine and riboflavin. Therefore, nutritional education helping adults to choose a balanced meal while consuming instant noodles should be implemented, Additionally, instant noodle manufacturers should consider nutritional aspects when developing new products.

• Journal of Environmental Science International
• /
• v.5 no.2
• /
• pp.153-170
• /
• 1996

Photosynthetic pigments and their degradation products in suspended and sinking particles collected from the Gulf of Mexico waters 1987~88, were measured using High Performance liquid Chromatography (HPIC). The short term variations in flux rates of chlorophylls and carotenoids as well as-their degradation products were compared at the mesoscale cyclonic and anticyclonic circulation features (cold joie rink and warm core ring). Chlorophyll a was the predominant porphyrin of suspended particulate matter at both CCR -and WCR. Among carotenoid pigments, 19'-hexanoyloxyfucoxanthin, which is a biomarker of prymnesiophytes, was dominant pigment at both rinds. Phaeophorbide a, which is produced through the Brazing processes of grazers, was the predominant degraded pigment in sinking particles ai the study aiea. Total pigment flux in CCR was an order of magnitude higher than that in WCR. Less than l% of the standing stock of the pigments measured sank out of the upper 200m of the WCR on any given day. Thus, suspended particulate matter in Gulf of Medico was not recycled rapidly.

• Journal of Korea Foundry Society
• /
• v.24 no.3
• /
• pp.165-174
• /
• 2004

The pore formation characteristic of Mg alloy during Lost Foam Casting(LFC) was investigated with reduced pressure test and real casting, which was compared with the results of previous work for Al alloy. Cast Mg alloys in LFC had much lower porosities in comparison with those of Al alloys. Also, the proper pouring temperature gave the minimum porosity like Al alloy although it was higher than that of Al alloys due to the worse fluidity of Mg alloy. The pore formation mechanism of Mg alloy in LFC was similar to that of AI alloy but the critical temperature showing the different mechanism is higher than that of Al alloy as much as $30{\sim}50^{\circ}C$. The result that Mg alloy in LFC had the lower porosity comparing with Al alloy was due to the extra solubility of hydrogen gas although the solubility of Al alloy was easily exceeded by the external sources like pyrolyzed polystyrene products. The mold evacuation gave the lower porosity due to the removal of polystyrene pyrolysis products, and reduced shrinkage defects. Also, there was a proper evacuation pressure that gave a porosity of almost 0vol%. But much higher vacuum degree than this proper pressure caused the severe entrapment of polymer pyrolysis products that gave the large porosity.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.45 no.1
• /
• pp.28-45
• /
• 2021

The importance of smart clothing as a product is increasingly emphasized as further growth in the potential of the smart market is expected. There is a high understanding and sympathy for the potential of smart clothing in the mass consumer market; therefore, commercialization is not actively carried out. This study enhances the understanding of the development direction of products with a focus on technical benefits, in order for smart clothing to gain access to customers as wearable devices. This study identifies major technologies used in smart clothing through an analysis of the patent technology status of smart clothing in Korea. Smart clothing is divided into three types: passive smart, active smart and advanced smart clothing based on a reaction mechanism and functional scope. We present the smart clothing and discuss the product features for three types. According to research, smart clothing products were equipped with passive, active, and advanced smart systems as well as provided new services by converging big data and AI technologies, rather than only using technologies such as sensors, controls, and actuators. Future directions for new smart clothing product development is also discussed in the conclusion.