• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1140-1146
• /
• 2018

Recently, a methodology for analyzing complex bio-signals using a deep learning model has emerged among studies that recognize human emotions. At this time, the accuracy of emotion classification may be changed depending on the evaluation method and reliability depending on the kind of data to be learned. In the case of biological signals, the reliability of data is determined according to the noise ratio, so that the noise detection method is as important as that. Also, according to the methodology for defining emotions, appropriate emotional evaluation methods will be needed. In this paper, we propose a wavelet -based noise threshold setting algorithm for verifying the reliability of data for multimodal bio-signal data labeled Valence and Arousal and a method for improving the emotion recognition rate by weighting the evaluation data. After extracting the wavelet component of the signal using the wavelet transform, the distortion and kurtosis of the component are obtained, the noise is detected at the threshold calculated by the hampel identifier, and the training data is selected considering the noise ratio of the original signal. In addition, weighting is applied to the overall evaluation of the emotion recognition rate using the euclidean distance from the median value of the Valence-Arousal plane when classifying emotional data. To verify the proposed algorithm, we use ASCERTAIN data set to observe the degree of emotion recognition rate improvement.

• Proceedings of the Korea Association of Crystal Growth Conference
• /
• 2000.06a
• /
• pp.195-214
• /
• 2000

The high-speed steel (shorten as HSS) consists of Fe and several kinds of transition metal carbides. The cutting tools or wear-resistant materials made from HSS experience relatively high thermal shock because a coolant such as water or oil is flowed over the surface of heated HSS. The purpose of this research is to increase the hardness, strength, fracture toughness and thermal shock resistance of HSS. A possible strategy is to incorporate a hard ceramic material with high strength in HSS matrix. This paper describes the processing, microstructure and mechanical properties of the oriented unidirectional mullite fiber/HSS composite. The unidirectional mullite fibers of 10${\mu}{\textrm}{m}$ diameter were dispersed by the ultrasonic irradiation of 38 kHz in an ethylenglycol suspension containing HSS powder of 11${\mu}{\textrm}{m}$ median size. The dried green composites with 4-68 vol% fibers were hot-pressed for 2h at 100$0^{\circ}C$ in Ar atmosphere under a pressure of 39 MPa. The higher density was achieved in the composite with a lower content of fibers. The oriented unidirectional fibers were well dispersed in the HSS matrix. The average distance between the center of fibers in the cross section was close to the value calculated from the fiber fraction. No reaction occurred at the interfaces between HSS and mullite fibers in the composites. The composite with 13.6 vol% fibers showed 100 MPa of four point flexural strength at room temperature. The thermal expansion of composite with heating was influenced by the orientation of mullite fibers.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.3
• /
• pp.83-90
• /
• 2020

Prototype selection offers the advantage of ensuring low learning time and storage space by selecting the minimum data representative of in-class partitions from the training data. This paper designs a new training data generation method using hyper-rectangles that can be applied to general classification algorithms. Hyper-rectangular regions do not contain different class data and divide the same class space. The median value of the data within a hyper-rectangle is selected as a prototype to form new training data, and the size of the hyper-rectangle is adjusted to reflect the data distribution in the class area. A set cover optimization algorithm is proposed to select the minimum prototype set that represents the whole training data. The proposed method reduces the time complexity that requires the polynomial time of the set cover optimization algorithm by using the greedy algorithm and the distance equation without multiplication. In experimented comparison with hyper-sphere prototype selections, the proposed method is superior in terms of prototype rate and generalization performance.

• Archives of Plastic Surgery
• /
• v.45 no.2
• /
• pp.165-170
• /
• 2018

Background To present the outcomes of the tarsal switch procedure using an anterior approach to correct severe ptosis with poor levator muscle function (<4 mm) with absent or poor Bell's phenomenon. Methods This retrospective case series included 11 patients with severe neurogenic or acquired myogenic palpebral ptosis. All patients underwent the tarsal switch procedure through an anterior approach from 2012 to 2015. Margin reflex distance (MRD1 and MRD2) and the palpebral fissure were evaluated preoperatively and postoperatively. Data were compared using the Wilcoxon signed-rank test. P-values <0.05 were considered to indicate statistical significance. Results Surgery was performed on 18 eyelids (11 patients). The median age at surgery was 57 years (range, 29-86 years). Four patients had unilateral ptosis and seven had bilateral ptosis. Nine patients had myogenic ptosis and two had neurogenic ptosis. Postoperatively, the chin-up position improved in all patients. The MRD1 increased statistically significantly, from 0 mm preoperatively to 1.0 mm postoperatively (P=0.001). The MRD2 decreased statistically significantly, from 4.5 mm preoperatively to 3.0 mm postoperatively (P=0.001). The palpebral fissure did not change (4.0 mm preoperatively to 4.0 mm postoperatively) (P=0.13). Conclusions The tarsal switch procedure through an anterior approach is an effective alternative for correcting severe ptosis, especially neurogenic or acquired myogenic ptosis. This procedure can be performed with minimal risk of ocular surface exposure and provides stable outcomes.

• Journal of Periodontal and Implant Science
• /
• v.48 no.4
• /
• pp.236-250
• /
• 2018

Purpose: Resorption of the alveolar bone is an unavoidable consequence of tooth extraction when appropriate alveolar ridge preservation (ARP) measures are not taken. The objective of this trial was to test the hypothesis that dimensional changes in the alveolar bone after tooth extraction would be reduced by inserting an equine collagen membrane and a collagen cone to fill and seal the alveolus (as ARP), in comparison to extraction with untreated alveoli. Methods: In this randomized clinical trial, 31 patients were directly treated with the collagen material after extraction of a tooth from the maxilla (the ARP group). Twenty-nine patients served as the control group. After extraction, no further treatment (i.e., no socket preservation measures) was performed in the control group. Changes in the alveolar process immediately after extraction and after an 8 (${\pm}1$)-week healing period were evaluated 3-dimensionally. Blinded analyses were performed after superimposing the data from the digitalized impressions and surfaces generated by cone-beam computed tomography. Results: Both the ARP and control groups showed a reduction of bone in the alveolar area after tooth extraction. However, significantly less bone resorption was detected in the clinically relevant buccal region in the ARP group. The median bone reduction was 1.18 mm in the ARP group and 5.06 mm in the control group (P=0.03). Conclusions: The proposed hypothesis that inserting a combination material comprising a collagen cone and membrane would lead to a difference in alveolar bone preservation can be accepted for the clinically relevant buccal distance. In this area, implantation of the collagen material led to significantly less alveolar bone resorption. German Clinical Trials Register at www.drks.de, DRKS00004769.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.35 no.6
• /
• pp.485-494
• /
• 2017

This study focuses on automatic image registration between multiple IR images using simple preprocessing method and modified local feature extraction algorithm. The input images were preprocessed by using the median and absolute value after histogram equalization, and it could be effectively applied to reduce the brightness difference value between images by applying the similarity of extracted features to the concept of angle instead of distance. The results were evaluated using visual and inverse RMSE methods. The features that could not be achieved by the existing local feature extraction technique showed high image matching reliability and application convenience. It is expected that this method can be used as one of the automatic registration methods between multi-sensor images under specific conditions.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.7
• /
• pp.1429-1434
• /
• 2017

In automatic defect detection problem for spectacle lenses, it is important to extract lens area accurately. Many existing detection methods fail to do it due to insufficient minute noise removal. In this paper, we propose an automatic defect detection method using Bresenham algorithm and adaptive binarization strategy. After usual average binarization, we apply Bresenham algorithm that has the power in extracting ellipse shape from image. Then, adaptive binarization strategy is applied to the critical minute noise removal inside the lens area. After noise removal, We can also compute the influence factor of the defect based on the fuzzy logic with two membership functions such as the size of the defect and the distance of the defect from the center of the lens. In experiment, our method successfully extracts defects in 10 out of 12 example images that include CHEMI, MID, HL, HM type lenses.

• Journal of Biomedical Engineering Research
• /
• v.28 no.5
• /
• pp.647-656
• /
• 2007

The purpose of this study was to analyze the effects of vibratory stimulus as somatosensory inputs on the postural control in human standing. To study these effects, the center of pressure(COP) was observed while subjects were standing on a stable and an unstable support with co-stimulated mechanical vibrations to flexor ankle muscles(tibialis anterior tendon, achilles tendon) and two plantar zones on both foot. The COP sway measurement was repeated twice in four conditions: (1) with visual cue and vibration, (2) without visual cue and vibration, (3) with visual cue and without vibration, (4) without visual cue and with vibration. The calculated parameters were the COP sway area and the distance, the median frequency and the spectral energy of COP sway in three intervals $0.1{\sim}0.3,\;0.3{\sim}1,\;1{\sim}3Hz$. The results showed that vibratory stimulus affect postural stability. The reduction rate of the COP sway with vibratory stimulus were higher on the unstable support because the effect of postural stability increases when afferent nervous flow is more activated by vibration on unstable support. If unclear visual or vibratory information is received, one type of information is compared with the other type of sensory information. Then the input balance between visual and vibratory information is corrected to maintain postural stability. These findings are important for the rehabilitation system of postural balance control and the use of vibratory information.

• Radiation Oncology Journal
• /
• v.35 no.2
• /
• pp.121-128
• /
• 2017

Purpose: To report the results of a correlation analysis of skin dose assessed by in vivo dosimetry and the incidence of acute toxicity. This is a phase 2 trial evaluating the feasibility of intraoperative radiotherapy (IORT) as a boost for breast cancer patients. Materials and Methods: Eligible patients were treated with IORT of 20 Gy followed by whole breast irradiation (WBI) of 46 Gy. A total of 55 patients with a minimum follow-up of 1 month after WBI were evaluated. Optically stimulated luminescence dosimeter (OSLD) detected radiation dose delivered to the skin during IORT. Acute toxicity was recorded according to the Common Terminology Criteria for Adverse Events v4.0. Clinical parameters were correlated with seroma formation and maximum skin dose. Results: Median follow-up after IORT was 25.9 weeks (range, 12.7 to 50.3 weeks). Prior to WBI, only one patient developed acute toxicity. Following WBI, 30 patients experienced grade 1 skin toxicity and three patients had grade 2 skin toxicity. Skin dose during IORT exceeded 5 Gy in two patients: with grade 2 complications around the surgical scar in one patient who received 8.42 Gy. Breast volume on preoperative images (p = 0.001), ratio of applicator diameter and breast volume (p = 0.002), and distance between skin and tumor (p = 0.003) showed significant correlations with maximum skin dose. Conclusions: IORT as a boost was well-tolerated among Korean women without severe acute complication. In vivo dosimetry with OSLD can help ensure safe delivery of IORT as a boost.

• Journal of Preventive Medicine and Public Health
• /
• v.51 no.5
• /
• pp.227-233
• /
• 2018

Objectives: Chronic diseases, including back pain, result in significant patient morbidity and societal burden. Overall improvement in physical fitness is recommended for prevention and treatment. Walking is a convenient modality for achieving initial gains. Our objective was to determine whether neighbourhood walkability, acting as a surrogate measure of physical fitness, was associated with the presence of chronic disease. Methods: We conducted a cross-sectional study of prospectively collected data from a prior randomized cohort study of 227 patients referred for tertiary assessment of chronic back pain in Ottawa, ON, Canada. The Charlson Comorbidity Index (CCI) was calculated from patient-completed questionnaires and medical record review. Using patients' postal codes, neighbourhood walkability was determined using the Walk Score, which awards points based on the distance to the closest amenities, yielding a score from 0 to 100 (0-50: car-dependent; 50-100: walkable). Results: Based on the Walk Score, 134 patients lived in car-dependent neighborhoods and 93 lived in walkable neighborhoods. A multivariate logistic regression model, adjusted for age, gender, rural postal code, body mass index, smoking, median household income, percent employment, pain, and disability, demonstrated an adjusted odds ratio of 2.75 (95% confidence interval, 1.16 to 6.53) times higher prevalence for having a chronic disease for patients living in a car-dependent neighborhood. There was also a significant dose-related association (p=0.01; Mantel-Haenszel chi-square=6.4) between living in car-dependent neighbourhoods and more severe CCI scores. Conclusions: Our findings suggest that advocating for improved neighbourhood planning to permit greater walkability may help offset the burden of chronic disease.