• The Research Journal of the Costume Culture
• /
• v.14 no.2
• /
• pp.320-333
• /
• 2006

The purpose of this study was to find out differences of visual evaluation according to perceiver's gender, clothing silhouette, and body exposure of the lingerie look. Subjects were 246 college males and females in Seoul. The visual evaluation of the lingerie look was divided into four image dimensions: elegance, individuality, fascination, and activity. Silhouette had significant influences on the perception of elegance and activity. The hourglass silhouettes were evaluated more elegant and active than the tubular silhouettes. The body exposure had significant influences on the evaluation of elegance, individuality, fascination, and activity. The shoulder and the back exposure were estimated high in fascination, the breast exposure low in elegance, and the waist exposure high in individuality and activity. Individuality, fascination, and activity had interaction effects by perceiver's gender and body exposure. Males estimated the waist exposure to be more fascinating than females, and females estimated the breast exposure to be less active than males. There were significant interaction effects in evaluating the 4 image dimensions according to the silhouette and body exposure. The shoulder and the leg exposure of the hourglass silhouettes were estimated more elegant than those of the tubular silhouettes. Dimensions of clothing image which influenced on preference of lingerie look were different between males and females.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.399-401
• /
• 2022

A definitive diagnosis to classify the breast malignancy status may be achieved by microscopic analysis using surgical open biopsy. However, this procedure requires experts in the specializing of histopathological image analysis directing to time-consuming and high cost. To overcome these issues, deep learning is considered practically efficient to categorize breast cancer into benign and malignant from histopathological images in order to assist pathologists. This study presents a pre-trained convolutional neural network model architecture with a 100% fine-tuning scheme and Adagrad optimizer to classify the breast cancer histopathological images into benign and malignant using a 40× magnification BreaKHis dataset. The pre-trained architecture was constructed using the InceptionResNetV2 model to generate a modified InceptionResNetV2 by substituting the last layer with dense and dropout layers. The results by demonstrating training loss of 0.25%, training accuracy of 99.96%, validation loss of 3.10%, validation accuracy of 99.41%, test loss of 8.46%, and test accuracy of 98.75% indicated that the modified InceptionResNetV2 model is reliable to predict the breast malignancy type from histopathological images. Future works are necessary to focus on k-fold cross-validation, optimizer, model, hyperparameter optimization, and classification on 100×, 200×, and 400× magnification.

• Journal of radiological science and technology
• /
• v.23 no.1
• /
• pp.55-61
• /
• 2000

This report was the results of an investigation based on the status of the mammography in the 45 medical facilities in the areas of Seoul and Kyong-Gi Do. In regard to mammography we were able to understand the rectification method of the generator, the functions, the radiographic techniques, the patient exposure dose, etc. Recently, the occurrence of breast cancer has rapidly increased and has lead to increased interest in the early discovery of breast cancer. However, mammography has not kept up with the publics interests and its demand. The main problem is thought to be a wide great difference in the quality of the facilities, especially in the techniques of radiograpic and the capabilities of the generator, which would have a major effect on the grade management of the image quality. In order to be ready and keep up with the high rate of increase of breast cancer, the standardization of mammography and the grade management is urgently required as a solution to the problems in the increase of image qualify and the decrease of dose. It is regarded that the basic guide of the techniques of photography in the mammography, which is being used in all developed countries due to the influence of the USA, should be presented, especially in Korea, and for this, more active and enthusiastic education and enlightenment should be needed.

• Archives of Plastic Surgery
• /
• v.51 no.1
• /
• pp.30-35
• /
• 2024

Background Breast aesthetics evaluation often relies on subjective assessments, leading to the need for objective, automated tools. We developed the Seoul Breast Esthetic Scoring Tool (S-BEST), a photometric analysis software that utilizes a DenseNet-264 deep learning model to automatically evaluate breast landmarks and asymmetry indices. Methods S-BEST was trained on a dataset of frontal breast photographs annotated with 30 specific landmarks, divided into an 80-20 training-validation split. The software requires the distances of sternal notch to nipple or nipple-to-nipple as input and performs image preprocessing steps, including ratio correction and 8-bit normalization. Breast asymmetry indices and centimeter-based measurements are provided as the output. The accuracy of S-BEST was validated using a paired t-test and Bland-Altman plots, comparing its measurements to those obtained from physical examinations of 100 females diagnosed with breast cancer. Results S-BEST demonstrated high accuracy in automatic landmark localization, with most distances showing no statistically significant difference compared with physical measurements. However, the nipple to inframammary fold distance showed a significant bias, with a coefficient of determination ranging from 0.3787 to 0.4234 for the left and right sides, respectively. Conclusion S-BEST provides a fast, reliable, and automated approach for breast aesthetic evaluation based on 2D frontal photographs. While limited by its inability to capture volumetric attributes or multiple viewpoints, it serves as an accessible tool for both clinical and research applications.

• Journal of radiological science and technology
• /
• v.41 no.3
• /
• pp.231-240
• /
• 2018

Automated breast ultrasound (ABUS) equipment is a new innovative technique for 3D automatic breast scanning, but limited for the examination in the concave axillary region. The purpose of this study was to determine feasible candidate materials for the ultrasonic wave propagation media in ABUS, enabling the evaluation of the axillary region. Ultrasonography was performed using an ABUS system ($Invenia^{TM}ABUS$, GE, USA) on the ultrasound-specific phantom (UC-551M-0.5, ATS Laboratories, USA) covered by different candidate materials. The validity of feasible candidate materials was evaluated by image quality. Three independent radiological technologists, with more than 10 years of experience, visually assessed on the images. The inter-observer agreements according to the candidate materials were tested using Cronbach's alpha. Unenveloped solidified carrageenan can be a feasible material for the use of ABUS with excellent test reliability. Therefore, the coverage of the axillary region with carrageenan may be effective for ABUS which was originally developed for the convex anatomic structure as female breast.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.8
• /
• pp.1272-1277
• /
• 2017

To assist radiologists for the characterization of breast masses, Computer-aided Diagnosis(CADx) system has been studied. The CADx system can improve the diagnostic accuracy of radiologists by providing objective information about breast masses. Morphological and texture features were extracted from the breast ultrasound images. Based on extracted features, the CADx system retrieves masses that are similar to a query mass from a reference library using a k-nearest neighbor (k-NN) approach. Eight similarity measures of distance, Euclidean, Chebyshev(Minkowski family), Canberra, Lorentzian($F_2$ family), Wave Hedges, Motyka(Intersection family), and Cosine, Dice(Inner Product family) are evaluated by ROC(Receiver Operating Characteristic) analysis. The Inner Product family measure used with the k-NN classifier provided slightly higher performance for classification of malignant and benign masses than those with the Minkowski, $F_2$, and Intersection family measures.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.898-902
• /
• 2003

In this paper we proposes a new technique for identification of breast cancer by classification of proteome pattern generated from 2-D polyacrylamide gel electrophoresis (2-D PAGE) and development of cancer diagnosis system : HABIT. Proteome patterns reflect the underlying pathological state of a human organ and it is believed that the anomalies or diseases of human organs are identified by the analysis or classification of the patterns. Proteome patterns consist of quantitative information of the spots such as their size, position, and density in the proteome image produced from 2-D PAGE, for the Image mining of proteome pattern, SVM(support vector machine) and GA(genetic algorithm) are used to generate a decision model for the identification of breast cancer The decision model was then used to classify an independent set of test proteome patterns into the affecter and unaffecter classes. The proposed technique was tested by actual clinical test samples and showed a good performance of a hit ratio of 90%.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.4
• /
• pp.1677-1684
• /
• 2016

Background: Among the factors reported to determine the quality of life of breast cancer patients are socio-demographic background, clinical stage, type of treatment received, and the duration since diagnosis. Objective: The objective of this study was to determine the quality of life (QOL) scores among breast cancer patients at a Malaysian public hospital. Materials and Methods: This cross-sectional study of breast cancer patients was conducted between March to June 2013. QOL scores were determined using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) and its breast cancer supplementary measure (QLQ-BR23). Both the QLQ-C30 and QLQ-BR23 assess items from functional and symptom scales. The QLQ-C30 in addition also measures the Global Health Status (GHS). Systematic random sampling was used to recruit patients. Results: 223 breast cancer patients were recruited with a response rate of 92.1%. The mean age of the patients was 52.4 years (95% CI = 51.0, 53.7, SD=10.3). Majority of respondents are Malays (60.5%), followed by Chinese (19.3%), Indians (18.4%), and others (1.8%). More than 50% of respondents are at stage III and stage IV of malignancy. The mean Global Health Status was 65.7 (SD = 21.4). From the QLQ-C30, the mean score in the functioning scale was highest for 'cognitive functioning' (84.1, SD=18.0), while the mean score in the symptom scale was highest for 'financial difficulties' (40.1, SD=31.6). From the QLQ-BR23, the mean score for functioning scale was highest for 'body image' (80.0, SD=24.6) while the mean score in the symptom scale was highest for 'upset by hair loss' (36.2, SD=29.4). Two significant predictors for Global Health Status were age and employment. The predictors explained 10.6% of the variation of global health status ($R^2=0.106$). Conclusions: Age and employment were found to be significant predictors for Global Health Status (GHS). The Quality of Life among breast cancer patients reflected by the GHS improves as age and employment increases.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.5 s.311
• /
• pp.80-88
• /
• 2006

In this paper we developed a novel system for automatic detection of mass type breast cancer on dense digital mammogram images. The new approaches presented in this paper are as follows: 1) we presented a method that stably decides the mass center and radius without being affected by image signal irregularity. 2) We developed a radial directional filter that is suitable to process mass image signal. 3) And we developed the multiple feature function based on mass shape spiculation, mass center homogeneity, and mass eccentricity, so as to determine mass-type breast cancer. When the proposed system is applied to dense mammographic images, the true 기arm rate is improved by 10% over a conventional system while the false alarm is increased by 1 per image.

• Progress in Medical Physics
• /
• v.24 no.3
• /
• pp.183-190
• /
• 2013

Recently, the photoacoustic imaging system has been widely and intensively developed, and has been shown the possibility of diagnosis for early stage cancer. In this study, we developed a photoacoustic tomography imaging system with a commercial ultra sound device and a linear array probe. A tube phantom and a chicken breast phantom was made for the possibility of a system as a breast cancer detection. A moving average filter and a band pass filter with 3~6 MHz bandwidth were developed for background noise elimination before delay-and-sum beamforming algorithm was used for image reconstruction. As a result, we showed that some signal processing procedure before beamforming was effective for the photoacoustic image reconstruction.