• International journal of advanced smart convergence
• /
• v.8 no.1
• /
• pp.24-34
• /
• 2019

In this research, a practical deep learning framework to differentiate the lesions and nodules in breast acquired with ultrasound imaging has been proposed. 7408 ultrasound breast images of 5151 patient cases were collected. All cases were biopsy proven and lesions were semi-automatically segmented. To compensate for the shift caused in the segmentation, the boundaries of each lesion were drawn using Fully Convolutional Networks(FCN) segmentation method based on the radiologist's specified point. The data set consists of 4254 benign and 3154 malignant lesions. In 7408 ultrasound breast images, the number of training images is 6579, and the number of test images is 829. The margin between the boundary of each lesion and the boundary of the image itself varied for training image augmentation. The training images were augmented by varying the margin between the boundary of each lesion and the boundary of the image itself. The images were processed through histogram equalization, image cropping, and margin augmentation. The networks trained on the data with augmentation and the data without augmentation all had AUC over 0.95. The network exhibited about 90% accuracy, 0.86 sensitivity and 0.95 specificity. Although the proposed framework still requires to point to the location of the target ROI with the help of radiologists, the result of the suggested framework showed promising results. It supports human radiologist to give successful performance and helps to create a fluent diagnostic workflow that meets the fundamental purpose of CADx.

• Journal of Biosystems Engineering
• /
• v.43 no.4
• /
• pp.426-439
• /
• 2018

Purpose: To continuously monitor soil and climatic properties, a data acquisition system (DAQ) was developed and tested in plum farms (Gyewol-ri and Haechang-ri, Suncheon, Korea). Methods: The DAQ consisted of a Raspberry-Pi processor, a modem, and an ADC board with multiple sensors (soil moisture content (SEN0193), soil temperature (DS18B20), climatic temperature and humidity (DHT22), and rainfall gauge (TR-525M)). In the laboratory, various tests were conducted to calibrate SEN0193 at different soil moistures, soil temperatures, depths, and bulk densities. For performance comparison of the SEN0193 sensor, two commercial moisture sensors (SMS-BTA and WT-1000B) were tested in the field. The collected field data in Raspberry-Pi were transmitted and stored on a web server database through a commercial communications wireless network. Results: In laboratory tests, it was found that the SEN0193 sensor voltage reading increased significantly with an increase in soil bulk density. A linear calibration equation was developed between voltage and soil moisture content depending on the farm soil bulk density. In field tests, the SEN0193 sensor showed linearity (R = 0.76 and 0.73) between output voltage and moisture content; however, the other two sensors showed no linearity, indicating that site-specific calibration is important for accurate sensing. In the long-term monitoring results, it was observed that the measured climate temperature was almost the same as website information. Soil temperature information was higher than the values measured by DS18B20 during spring and summer. However, the local rainfall measured using TR 525M was significantly different from the values on the website. Conclusion: Based on the test results obtained using the developed monitoring system, it is thought that the measurement of various parameters using one device would be helpful in monitoring plum growth. Field data from the local farm monitoring system can be coupled with website information from the weather station and used more efficiently.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.4
• /
• pp.79-86
• /
• 2019

This paper describes a time-series data prediction based on artificial neural networks (ANN). In this study, a batch based ANN model and a stochastic ANN model have been implemented using TensorFlow libraries. Each model are evaluated by comparing training and testing errors that are measured through experiment. To train and test each model, tax dataset was used that are collected from the government website of indiana state budget agency in USA from 2001 to 2018. The dataset includes tax incomes of individual, product sales, company, and total tax incomes. The experimental results show that batch model reveals better performance than stochastic model. Using the batch scheme, we have conducted a prediction experiment. In the experiment, total taxes are predicted during next seven months, and compared with actual collected total taxes. The results shows that predicted data are almost same with the actual data.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.30 no.2
• /
• pp.173-176
• /
• 2019

In this study, a Ka-band waveguide rotary joint that can be applied to a millimeter-wave seeker is designed and fabricated. The proposed rotary joint maintains a low standing-wave ratio and low-loss characteristics, and has two rotary axes designed to enable azimuth and elevation rotation. The rotary joint is designed as a ridge-waveguide-type mode converter and a ${\lambda}/4$ choke structure to match the electromagnetic wave propagation mode between the spherical and circular waveguides. A performance test using a network analyzer and a high-power transmitter to assess vibration and shock were conducted. Results showed that the rotary joint had a very low standing-wave ratio of less than the maximum of 1.19:1 and an insertion loss of less than 0.80 dB at $F_C{\pm}500MHz$.

• Journal of IKEEE
• /
• v.25 no.1
• /
• pp.95-100
• /
• 2021

If we know the location of the hull block and the welding feeder in the shipyard, we can easily obtain the location information of the worker. That data is very useful for implementing a workplace safety monitoring system. However, it is difficult to apply a fixed communication network to the workplace due to the specificity of the hull structure and welding process. In this study, inductive power line communication, which can replace dedicated communication line, was reviewed. A ferrite core was used as an inductive coupler to be installed on the power cable of the welding machine, and a nano-crystalline core was applied as a coupler to be fastened to the support rod of the metal block. In order to visualize the operating principle of the proposed couplers, 3D modeling and finite element analysis were performed with the COMSOL AC/DC module. In the communication performance test using an aluminum profile, when the communication channel was formed by the contact of the welding electrode, the bandwidth was kept above 6 Mbps.

• Journal of dental hygiene science
• /
• v.20 no.4
• /
• pp.206-212
• /
• 2020

Background: A Teachable Machine is a kind of machine learning web-based tool for general persons. In this paper, the feasibility of Google's Teachable Machine (ver. 2.0) was studied in the diagnosis of the tooth-marked tongue. Methods: For machine learning of tooth-marked tongue diagnosis, a total of 1,250 tongue images were used on Kaggle's web site. Ninety percent of the images were used for the training data set, and the remaining 10% were used for the test data set. Using Google's Teachable Machine (ver. 2.0), machine learning was performed using separated images. To optimize the machine learning parameters, I measured the diagnosis accuracies according to the value of epoch, batch size, and learning rate. After hyper-parameter tuning, the ROC (receiver operating characteristic) analysis method determined the sensitivity (true positive rate, TPR) and specificity (false positive rate, FPR) of the machine learning model to diagnose the tooth-marked tongue. Results: To evaluate the usefulness of the Teachable Machine in clinical application, I used 634 tooth-marked tongue images and 491 no-marked tongue images for machine learning. When the epoch, batch size, and learning rate as hyper-parameters were 75, 0.0001, and 128, respectively, the accuracy of the tooth-marked tongue's diagnosis was best. The accuracies for the tooth-marked tongue and the no-marked tongue were 92.1% and 72.6%, respectively. And, the sensitivity (TPR) and specificity (FPR) were 0.92 and 0.28, respectively. Conclusion: These results are more accurate than Li's experimental results calculated with convolution neural network. Google's Teachable Machines show good performance by hyper-parameters tuning in the diagnosis of the tooth-marked tongue. We confirmed that the tool is useful for several clinical applications.

• Journal of Platform Technology
• /
• v.8 no.1
• /
• pp.16-23
• /
• 2020

For protecting copyright for trademark, convolutional neural network can be used to confirm genuine of trademark image. For this, repeated training one trademark image degrades the performance of machine learning because of overfitting problem. Therefore, this type of machine learning application generates training data in various way. But if genuine trademark image is rotated, this image is classified as not genuine trademark. In this paper, we propose the way for extending training data to confirm genuine of trademark image which is rotated. Our proposed way generates rotated image from genuine trademark image as training data. To show effectiveness of our proposed way, we use CNN machine learning model, and evaluate the accuracy with test image. From evaluation result, our way can be used to generate training data for machine learning application which confirms genuine of rotated trademark image.

• Geomechanics and Engineering
• /
• v.28 no.5
• /
• pp.505-520
• /
• 2022

In recent years, geotextile-encased gravel columns (usually called stone columns) have become a popular method to increasing soil shear strength, decreasing the settlement, acceleration of the rate of consolidation, reducing the liquefaction potential and increasing the bearing capacity of foundations. The behavior of improved loose base-soil with gravel columns under shear loading and the shear stress-horizontal displacement curves got from large scale direct shear test are of great importance in understanding the performance of this method. In the present study, by performing 36 large-scale direct shear tests on sandy base-soil with different fine-content of zero to 30% in both not improved and improved with gravel columns, the effect of the presence of gravel columns in the loose soils were investigated. The results were used to predict the shear stress-horizontal displacement curve of these samples using support vector machines (SVM). Variables such as the non-plastic fine content of base-soil (FC), the area replacement ratio of the gravel column (Arr), the geotextile encasement and the normal stress on the sample were effective factors in the shear stress-horizontal displacement curve of the samples. The training and testing data of the model showed higher power of SVM compared to multilayer perceptron (MLP) neural network in predicting shear stress-horizontal displacement curve. After ensuring the accuracy of the model evaluation, by introducing different samples to the model, the effect of different variables on the maximum shear stress of the samples was investigated. The results showed that by adding a gravel column and increasing the Arr, the friction angle (ϕ) and cohesion (c) of the samples increase. This increase is less in base-soil with more FC, and in a proportion of the same Arr, with increasing FC, internal friction angle and cohesion decreases.

• Journal of Biomedical Engineering Research
• /
• v.44 no.2
• /
• pp.139-146
• /
• 2023

Osteoporosis is a disease in which the risk of bone fractures increases due to a decrease in bone density caused by aging. Osteoporosis is diagnosed by measuring bone density in the total hip, femoral neck, and lumbar spine. To accurately measure bone density in the lumbar spine, the vertebral region must be segmented from the lumbar X-ray image. Deep learning-based automatic spinal segmentation methods can provide fast and precise information about the vertebral region. In this study, we used 695 lumbar spine images as training and test datasets for a deep learning segmentation model. We proposed a lumbar automatic segmentation model, CM-Net, which combines the center point of the spine and the modified U-Net network. As a result, the average Dice Similarity Coefficient(DSC) was 0.974, precision was 0.916, recall was 0.906, accuracy was 0.998, and Area under the Precision-Recall Curve (AUPRC) was 0.912. This study demonstrates a high-performance automatic segmentation model for lumbar X-ray images, which overcomes noise such as spinal fractures and implants. Furthermore, we can perform accurate measurement of bone density on lumbar X-ray images using an automatic segmentation methodology for the spine, which can prevent the risk of compression fractures at an early stage and improve the accuracy and efficiency of osteoporosis diagnosis.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.6
• /
• pp.51-57
• /
• 2009

Wireless sensor networks consist of numerous sensor nodes that have inexpensive and limited resources. Generally, most of the sensors are assigned to the hazardous or uncontrollable environments. If the sensor nodes are randomly assigned to the wide target area, it is very hard to see the accurate locations of sensor nodes. Therefore, this study provides an efficient key distribution scheme to solve these problems. Based on the provided scheme, the study enabled the closely neighboring nodes to exchange information with each other after securing safe links by using the pre-distributed keys. At the same time, the provided scheme could increase the probability of multiparty key detection among nodes by using the location information of sensor node. Lastly, the study intended to show the superiority of the limitation method through a performance test.