• International journal of advanced smart convergence
• /
• v.11 no.1
• /
• pp.19-27
• /
• 2022

Across the world, 'housing' comprises a significant portion of wealth and assets. For this reason, fluctuations in real estate prices are highly sensitive issues to individual households. In Korea, housing prices have steadily increased over the years, and thus many Koreans view the real estate market as an effective channel for their investments. However, if one purchases a real estate property for the purpose of investing, then there are several risks involved when prices begin to fluctuate. The purpose of this study is to design a real estate price 'return rate' prediction model to help mitigate the risks involved with real estate investments and promote reasonable real estate purchases. Various approaches are explored to develop a model capable of predicting real estate prices based on an understanding of the immovability of the real estate market. This study employs the LSTM method, which is based on artificial intelligence and deep learning, to predict real estate prices and validate the model. LSTM networks are based on recurrent neural networks (RNN) but add cell states (which act as a type of conveyer belt) to the hidden states. LSTM networks are able to obtain cell states and hidden states in a recursive manner. Data on the actual trading prices of apartments in autonomous districts between January 2006 and December 2019 are collected from the Actual Trading Price Disclosure System of the Ministry of Land, Infrastructure and Transport (MOLIT). Additionally, basic data on apartments and commercial buildings are collected from the Public Data Portal and Seoul Metropolitan Government's data portal. The collected actual trading price data are scaled to monthly average trading amounts, and each data entry is pre-processed according to address to produce 168 data entries. An LSTM model for return rate prediction is prepared based on a time series dataset where the training period is set as April 2015~August 2017 (29 months), the validation period is set as September 2017~September 2018 (13 months), and the test period is set as December 2018~December 2019 (13 months). The results of the return rate prediction study are as follows. First, the model achieved a prediction similarity level of almost 76%. After collecting time series data and preparing the final prediction model, it was confirmed that 76% of models could be achieved. All in all, the results demonstrate the reliability of the LSTM-based model for return rate prediction.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.11 no.1
• /
• pp.89-96
• /
• 2023

In this paper, a ferrosilicon by-product was evaluated to confirm the feasibility of recycling it as supplementary cementitious material of ordinary Portland cement in concrete. Three different levels of replacement ratio (10 %, 20 % and 30 % of total binder) were applied to find which is the most beneficial to be used as a binder. Ferrosilicon concrete was initially assessed at setting time and compressive strength. Durability was evaluated by the resistance to chloride penetration test(RCPT) and alkali-silica reaction(ASR) with a comparison to silica fume concrete due to their similarity in chemical composition. The porosimetry and X-ray diffraction analysis along with energy dispersive X-ray spectroscopy give information on the microstructural characteristics of the ferrosilicon concrete. It was found that 10 % ferrosilicon concrete has higher strength while 20 %, 30 % have lower strength than OPC concrete. However, chemical resistance to chloride attack is higher when replacement is increased. Compared to silica fume, the durability of ferrosilicon might be less efficient however, it is obviously beneficial than OPC. High SiO₂ content in ferrosilicon results in producing more C-S-H gel which could make denser pore structure. Most of the risk of alkali silica reaction to silicate binders through length change tests was less than 0.2 %, and both mortar using ferrosilicon and silica fume showed better resistance to alkali silica reaction as the substitution rate increased.Reuse of industrial waste rather than producing highly refined additives might reduce environmental load during manufacture and save costs.

• Korean Journal of Radiology
• /
• v.22 no.6
• /
• pp.983-993
• /
• 2021

Objective: To investigate the image quality of ultralow-dose CT (ULDCT) of the chest reconstructed using a cycle-consistent generative adversarial network (CycleGAN)-based deep learning method in the evaluation of pulmonary tuberculosis. Materials and Methods: Between June 2019 and November 2019, 103 patients (mean age, 40.8 ± 13.6 years; 61 men and 42 women) with pulmonary tuberculosis were prospectively enrolled to undergo standard-dose CT (120 kVp with automated exposure control), followed immediately by ULDCT (80 kVp and 10 mAs). The images of the two successive scans were used to train the CycleGAN framework for image-to-image translation. The denoising efficacy of the CycleGAN algorithm was compared with that of hybrid and model-based iterative reconstruction. Repeated-measures analysis of variance and Wilcoxon signed-rank test were performed to compare the objective measurements and the subjective image quality scores, respectively. Results: With the optimized CycleGAN denoising model, using the ULDCT images as input, the peak signal-to-noise ratio and structural similarity index improved by 2.0 dB and 0.21, respectively. The CycleGAN-generated denoised ULDCT images typically provided satisfactory image quality for optimal visibility of anatomic structures and pathological findings, with a lower level of image noise (mean ± standard deviation [SD], 19.5 ± 3.0 Hounsfield unit [HU]) than that of the hybrid (66.3 ± 10.5 HU, p < 0.001) and a similar noise level to model-based iterative reconstruction (19.6 ± 2.6 HU, p > 0.908). The CycleGAN-generated images showed the highest contrast-to-noise ratios for the pulmonary lesions, followed by the model-based and hybrid iterative reconstruction. The mean effective radiation dose of ULDCT was 0.12 mSv with a mean 93.9% reduction compared to standard-dose CT. Conclusion: The optimized CycleGAN technique may allow the synthesis of diagnostically acceptable images from ULDCT of the chest for the evaluation of pulmonary tuberculosis.

• Korean Journal of Radiology
• /
• v.23 no.12
• /
• pp.1251-1259
• /
• 2022

Objective: T1 mapping provides valuable information regarding cardiomyopathies. Manual drawing is time consuming and prone to subjective errors. Therefore, this study aimed to test a DL algorithm for the automated measurement of native T1 and extracellular volume (ECV) fractions in cardiac magnetic resonance (CMR) imaging with a temporally separated dataset. Materials and Methods: CMR images obtained for 95 participants (mean age ± standard deviation, 54.5 ± 15.2 years), including 36 left ventricular hypertrophy (12 hypertrophic cardiomyopathy, 12 Fabry disease, and 12 amyloidosis), 32 dilated cardiomyopathy, and 27 healthy volunteers, were included. A commercial deep learning (DL) algorithm based on 2D U-net (Myomics-T1 software, version 1.0.0) was used for the automated analysis of T1 maps. Four radiologists, as study readers, performed manual analysis. The reference standard was the consensus result of the manual analysis by two additional expert readers. The segmentation performance of the DL algorithm and the correlation and agreement between the automated measurement and the reference standard were assessed. Interobserver agreement among the four radiologists was analyzed. Results: DL successfully segmented the myocardium in 99.3% of slices in the native T1 map and 89.8% of slices in the post-T1 map with Dice similarity coefficients of 0.86 ± 0.05 and 0.74 ± 0.17, respectively. Native T1 and ECV showed strong correlation and agreement between DL and the reference: for T1, r = 0.967 (95% confidence interval [CI], 0.951-0.978) and bias of 9.5 msec (95% limits of agreement [LOA], -23.6-42.6 msec); for ECV, r = 0.987 (95% CI, 0.980-0.991) and bias of 0.7% (95% LOA, -2.8%-4.2%) on per-subject basis. Agreements between DL and each of the four radiologists were excellent (intraclass correlation coefficient [ICC] of 0.98-0.99 for both native T1 and ECV), comparable to the pairwise agreement between the radiologists (ICC of 0.97-1.00 and 0.99-1.00 for native T1 and ECV, respectively). Conclusion: The DL algorithm allowed automated T1 and ECV measurements comparable to those of radiologists.

• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Korean Journal of Environment and Ecology
• /
• v.21 no.6
• /
• pp.536-543
• /
• 2007

This research was conducted every month from June 1994 until August 1996 with the aim to understand the ecosystem structure through the analysis of oribatid mite community structure in soil subsequent to environmental difference of its habitats located at northward & southward slopes adjacent to each other at an altitude of 1,000 meters of Mt.Jumbong, which is a natural reserved forest, remaining intact. There appeared a significant difference [t-test, p<0.06] in comparison of the number of the species and individuals of Oribatid mite species which were collected and identified at two survey areas. The mean density and the number of the species collected and identified at the northward slopes, and southward slopes were $99.2{\pm}17.6,\;234.2{\pm}62.6$ and $24.7{\pm}3.0,\;40.8{\pm}5.8$, respectively. Species diversity index(H') was higher at the southward slopes($3.09{\pm}0.11$) than at the northward slopes($2.71{\pm}0.13$). The population size of Oribatid mite species was found by the percentage of each species density as against the whole density and classified into dominant species, influent species, and recessive species according to the percentage; as a result, O. nova and Suctobelbella naginata was found to be a dominant species at both survey slopes while Trichogalumna nipponica was found to be a dominant species, at southward but it wasn't collected at the northward slopes at all. The feeding habit of the dominant species at two survey slopes was found to be microphytophagous- eating soil microbe. There appeared a conspicuous difference in compositions of the number of the species, individuals and dominant species at the southward/northward slopes adjoining each other at an attitude of 1,000 meters and less similarity between the two survey slopes. Conclusively, It was found that the heterogeneity of microhabitat has a great effect on Oribatid mite's community characteristics.

• The Korean Journal of Pesticide Science
• /
• v.17 no.4
• /
• pp.350-358
• /
• 2013

Both 13-week and 1-year studies in dog were required for pesticide registration in domestic pesticide control authority. It is raising issue up whether to request 1-year dog study of pesticides using non-food crop. So at this investigation, relevant toxicity test to establish acceptable daily intake (ADI), target organs, difference of no-observed adverse-effect levels (NOAELs) in 13-week and 1-year of 166 active ingredients are analyzed. The data were evaluated to determine if the 13-week dog study and the long term studies in two rodent species (mice and rats) without 1-year dog study were sufficient for the identification of NOAELs and lowest observed adverse effect levels (LOAELs) for the derivation of ADI. Toxicity end points and dose response data from 13 week and 1-year studies were compared. The analysis showed that 68 ADIs of the 166 pesticides were established from dog studies. Major target organs of dog studies were liver in 49 cases, body weight change in 21 cases, cholinesterase inhibition in 16 cases, and alteration in hematology in 14 cases. Similarity of target organ in 13-week and 1-year was 73%. 22 of 40 pesticides had similar critical effects regardless of duration and had NOAELs within a difference of 1.5-fold of each other. For the remaining 18 pesticides, 14 items had lower NOAELs in the 1-year study than 13-week study primarily due to dose selection and spacing. In only 10% of the cases were additional toxic effects identified in the 1-year study that were not observed in the 13-week study.

• The korean journal of orthodontics
• /
• v.36 no.1 s.114
• /
• pp.74-83
• /
• 2006

Analysis of lateral cephalometric radiograph (cephalogram) has been used routinely to evaluate skeletal and dental relationships, but analysis of the lateral facial photograph has not been used frequently for evaluation of skeletal relationships. As concerns about harm of X-ray irradiation increases, this study was planned to evaluate the possibility of substituting analysis of the lateral cephalogram with analysis of the lateral facial photograph by comparing these two analyses. According to the ANB values from cephalometric analysis, subjects were divided into three groups: Class I malocclusion group (n=32). Class II malocclusion group (n=32), and Class III malocclusion group (n=31). After measurements of angles indicating horizontal and vertical relationships of the maxilla and mandible on the lateral cephalograms and photographs, differences between Class I, II and III groups were evaluated. To evaluate the similarity between two similar values in the cephalograms and photographs, t-test using standardized variable Z and correlation analysis were performed in the Class I malocclusion group. The results showed that 1) SnN'Pg' on the photograph can be used to evaluate the antero-posterior relationship of the maxilla and mandible (ANB), 2) N'-Sn/Sn-Pg' on the photograph can be used to evaluate facial convexity (NA/APg), 3) Sn-Tra-Me' on the photograph can be used as a measurement similar to FMA. In conclusion, partly substituting lateral cephalogram analysis with lateral facial photograph analysis was possible in the evaluation of the maxilla and mandible.

• Tuberculosis and Respiratory Diseases
• /
• v.59 no.2
• /
• pp.170-178
• /
• 2005

Background : Idiopathic pulmonary arterial hypertension (IPAH) and chronic thromboembolic pulmonary hypertension (CTEPH) are rare but significantly imperative in inducing chronic pulmonary hypertension. Clinically, it is difficult to distinguish between IPAH and CTEPH. However, the treatment of pulmonary hypertension is different depending on the disease. The present study was performed to analyze the similarities and differences in clinical features between IPAH and CTEPH. Methods : During a nine-year period, thirty-three patients with IPAH and twenty-two patients with CTEPH were enrolled. Symptoms, physical findings, chest radiograph, electrocardiograph, pulmonary function test, echocardiograph, perfusion lung scan, right heart catheterization results were analyzed between both the groups. Results : The median age of IPAH group was 33 (6~70) years that was lower than that (52(27~80) years) of CTEPH group. Amongst the IPAH patients, there was female predominance (76 %) and there was no sex difference between the patients with CTEPH. Both the groups exhibited similarity in the results of chest radiograph, electrocardiograph, pulmonary function test, and echocardiograph. In the perfusion lung scan, all IPAH patients exhibited findings with normal (28%) or low probability (72%) of pulmonary embolism and all CTEPH patients exhibited findings with high probability of pulmonary embolism. Conclusion : Although IPAH and CTEPH bear similarities in terms of symptoms, physical signs and general investigation results, there were differences in age distribution, sex predominance and results of perfusion lung scan.

• Korean Journal of Agricultural Science
• /
• v.5 no.2
• /
• pp.127-135
• /
• 1978

For improvement and new design of tillage equipments, indoor test is very useful and more desirable than outdoor because the experiment of outdoor is very difficult and its cost is expensive. This study was carried out to determine the physical properties of artificial soil suitable for the indoor test with the soil bin manufactured at the workshop of the Dept. of Agricultural Machinery Engineering. The artificial soil being studied was made with very similarity to the natural soil of the experimental plots of Chungnam National University, and it consist of 39.35 percent, by weight of bentonite and 48.10 percent of sand with 12.55 percent of SAE 10W oil. The results are summarized as follows: 1. Bulk density increased with increasing number of rolling, and its relationship could be expressed. $y=1.073200+0.070780x-0.002263x^2$ where, y=bulk density ($g/cm^3$), x=number of rolling. These results could be explained that the effect of rolling velocity on the bulk density was not singnificant in the range of 4.5~10.4 em/sec. 2. The absolute soil hardness depended directly upon number of rolling, and their relationship could be expressed by the equation. $y=37.74(0.64 +0.17x-0.0054x^2)/(3.36-0.17x-0.0054x^2)^3$. where, y=absolute soil hardness($kg/cm^3$), x=number of rolling. 3. Relationship between the bulk density and absolute soil hardness could be expressed by the equation; $y=37.74(2.46x-2.02)/(6.02-2.46x)^3$. where, y=absolute soil hardness, x=bulk density. 4. The cohesion and the angle of internal friction of artificial soil were increased with increasing its bulk density. According to the cohesion and angle of internal friction, at the range of 1.60~1.75 ($g/cm^3$) of bulk density, this artificial soil was similar with sandy loam of 29.5% moisture content of natural soil. 5. Sliding-fricfion coefficient of steel plate on the artificial soil was 0.3~0.4 and rubber plate on it is 0.64~0.72. Those values were very similar with those of natural soil being studies by many others.