• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.06a
• /
• pp.204-207
• /
• 2001

뇌 MR 영상에서 질환을 자동적으로 진단하고 판별하는 작업은 정상인의 뇌 영상과의 비교를 통해서 가능하다. 정상인과의 뇌 영상 비교를 통하여 보다 정확하게 질병에 대한 근거를 제시할 수가 있기 때문에 이러한 접근 방법들이 여러 의료영상 연구 분야에서 시도되고 있다. 정상인의 뇌 영상과의 비교를 위해서는 우선적으로 해결되어야 하는 것이 현재의 대상 영상이 정상인 뇌의 어느 위치의 영상과 일치하는 지를 판별하는 문제이다. 따라서 본 연구는 이러한 뇌 매핑에 사용될 수 있는 특징들을 추출하기 위한 것으로, 뇌 매핑에 사용되는 특징들을 추출하기 위해서 뇌 MR 영상으로부터 대리영역, 뇌영역, 뇌척수액영역 그리고 눈영역을 분할한 후 이들의 윤곽선, 최소사각형과 각 영역들의 픽셀 정보들을 찾아낸다. 이는 추후 연구할 뇌 매핑을 위한 대분류에 사용될 수 있다.

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.35-45
• /
• 2022

Even in recent years, treatment of first-aid patients is still often delayed due to a shortage of medical resources in marginalized areas. Research on automating the analysis of medical data to solve the problems of inaccessibility for medical services and shortage of medical personnel is ongoing. Computer vision-based medical inspection automation requires a lot of cost in data collection and labeling for training purposes. These problems stand out in the works of classifying lesion that are rare, or pathological features and pathogenesis that are difficult to clearly define visually. Anomaly detection is attracting as a method that can significantly reduce the cost of data collection by adopting an unsupervised learning strategy. In this paper, we propose methods for detecting abnormal images on chest X-RAY images as follows based on existing anomaly detection techniques. (1) Normalize the brightness range of medical images resampled as optimal resolution. (2) Some feature vectors with high representative power are selected in set of patch features extracted as intermediate-level from lesion-free images. (3) Measure the difference from the feature vectors of lesion-free data selected based on the nearest neighbor search algorithm. The proposed system can simultaneously perform anomaly classification and localization for each image. In this paper, the anomaly detection performance of the proposed system for chest X-RAY images of PA projection is measured and presented by detailed conditions. We demonstrate effect of anomaly detection for medical images by showing 0.705 classification AUROC for random subset extracted from the PadChest dataset. The proposed system can be usefully used to improve the clinical diagnosis workflow of medical institutions, and can effectively support early diagnosis in medically poor area.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.9 no.1
• /
• pp.61-66
• /
• 2015

Chest X-ray images is the most common and widely used in clinical findings for a wide range of anatomical information about the prognosis of the disease in patients with cardiopulmonary rehabilitation. Many analysis algorithm was developed by a number of studies regarding the region segmentation and image analysis, there are specific differences due to the complexity and diversity of the image. In this paper, a diagnosis support system of the chest X-ray image based on image processing and analysis methods to detect the cardiopulmonary disease. The threshold value and morphological method was applied to segment the pulmonary region in a chest X-ray image. Anatomical measurements and texture analysis was performed on the segmented regions. The effectiveness of the proposed method is shown through experiments and comparison with diagnosis results by clinical experts to show that the proposed method can be used for decision support system.

• Journal of the Korean Society of Radiology
• /
• v.13 no.4
• /
• pp.589-596
• /
• 2019

X-rays used for diagnosis have a continuous energy distribution. However, photons with low energy not only reduce image contrast, but also contribute to the patient's radiation exposure. Therefore, clinics currently use filters made of aluminum. Such filters are advantageous because they can reduce the exposure of the patient to radiation. However, they may have negative effects on imaging quality, as they lead to increases in the scattered dose. In this study, we investigated the effects of the scattered dose generated by an aluminum filter on medical image quality. We used the relative standard deviation and the scatter degradation factor as evaluation indices, as they can be used to quantitatively express the decrease in the degree of contrast in imaging. We verified that the scattered dose generated by the increase in the thickness of the aluminum filter causes degradation of the quality of medical images.

• Electronics and Telecommunications Trends
• /
• v.36 no.5
• /
• pp.1-8
• /
• 2021

Over the last three decades, the development of Ti:sapphire femtosecond lasers has led to advancements in scientific and industrial fields. In particular, these advanced lasers show great potential for applications with bio-imaging and medical surgery, such as two-photon microscopy, nonlinear Raman microscopy, optical coherence tomography, and ophthalmic surgery. Herein, we present a detailed description of the theoretical and experimental physics of Kerr-lens mode-locked femtosecond Ti:sapphire lasers and its two-photon microscopy.

• Journal of the Korean Society of Radiology
• /
• v.15 no.1
• /
• pp.15-20
• /
• 2021

CT is important role in the medical field, such as disease diagnosis, but the number of examination and CT images are increasing. Recently, deep learning has been actively used in the medical field, and it has been used to diagnose auxiliary disease through object detection during deep learning using medical images. The purpose of study to evaluate accuracy by detecting kidney and vertebrae during abdominal CT using object detection deep learning in YOLOv3. As a results of the study, the detection accuracy of the kidney and vertebrae was 83.00%, 82.45%, and can be used as basic data for the object detection of medical images using deep learning.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.8
• /
• pp.504-508
• /
• 2010

In conventional diagnostic ultrasound strain imaging, when displaying strain image on a monitor, human visual characteristics are utilized such that hard regions are displayed as dark and soft regions are displayed as bright. Thus, hard regions representing tumor or cancer are displayed as dark, decreasing the contrast inside the lesion. Because the lesion area is stiff and thus displayed as dark, a method of inverting the image brightness and thereby increasing the contrast in the lesion for better diagnostic purposes is proposed wherein a postcompression signal is extended in the time domain by a factor corresponding to the reciprocal of the amount of the applied compression using a technique termed globally uniform stretching. Experiments were carried out to verify the proposed method on an ultrasound elasticity phantom with radio-frequency data acquired from a diagnostic ultrasound clinical scanner. It is found that the new method improves the contrast-to-noise ratio by a factor of up to about 1.8 compared to a conventional strain imaging method that employs a reversed gray color map without globally uniform stretching.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.5
• /
• pp.3308-3314
• /
• 2015

The purpose of this study is to investigate the casual relationships among purchase decision factors of medical devices, satisfaction and repurchasing intention. Data of this study is a 2012 survey on medical institutions using medical devices of MW(ministry of health and welfare) and We analyzed 116 medical institutions having ultrasound imaging system lager than general hospitals among data. For empirical analysis, we carried out confirmative factor analysis and path-analysis using AMOS 21.0 package. Main results of this study are as follows: First, brand has positive influence on satisfaction and satisfaction has positive impaction on repurchase intention. Second, although not statistically significant, performance and service have positive impaction on satisfaction and price has negative influence on satisfaction. The significance of this study is to provide empirical basis on establishment of efficient marketing strategy and enhancement of competitiveness for the medical device industry.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.101-104
• /
• 2006

본 논문에서는 재택형 한방 진단기기를 개발하기 위한 단계별 연구과정 중 혀의 영역 추출과 색상 분석을 통해 질환을 진단할 수 있는 설진 시스템 개발 기법에 대해 제안하고자 한다. 우리나라는 초고령화 사회를 맞아 의료비 증가로 인한 사회적 부담과 대체의학에 대한 의존도가 증가되고 있으며 이를 해결하기 위해 영상 처리 기술과 한방 진단 기법의 연계를 통한 설진 시스템을 구축하고자 한다. 통상 인체의 생체 신호를 반영하여 나타내어 주는 곳은 홍채나 혀, 오관 등이 있다. 본 논문에서는 이 중 혀를 통해 인간의 생체 신호에 대한 결과를 반영하여 건강 상태에 대한 정보를 제공해 주는 설진(舌診)을 네트워크 기반의 재택형 한방 진단 시스템으로 개발하고자 한다. 특히 본 논문은 한방 유비쿼터스 의료 시스템 개발을 위한 전체 시스템 중 혀 영역 추출과 색상으로부터 질환에 대한 정보를 제공하여 주는 방법에 대해 제안하고자 한다. 끝으로 실험에 의해 제안한 방법의 유용성을 입증하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.475-478
• /
• 2023

컴퓨터 단층촬영(CT)을 활용한 골격근 단면적은 근감소증과 관련된 기능을 평가하는 데 사용된다. 일반적인 근감소증 연구는 요추 3번의 골격근량을 주로 보지만 암 또는 폐절제술과의 상관관계를 예측하기 위한 다양한 연구에서는 흉추 4번, 7번, 8번, 10번, 12번 다양한 수준의 골격근량으로 연구를 진행하고 있음을 알 수 있다. 본 논문에서는 흉부와 복부 CT 영상에서 근감소증 진단을 위해서 흉추와 요추의 영역별 슬라이스를 검출하기 위해서 CNN 구조의 EfficientNetV2를 전이학습하여 인공지능 모듈을 개발하였다. 인공지능 모듈은 전체 흉부 및 복부 CT 영상에서 Cervical, T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11, T12, L1, L2, L3, L4, L5, Sacral 총 19 클래스를 검출하도록 하였다. Test 데이터셋을 사용하여 Confusion Matrix와 Grad-CAM으로 모델의 정확도를 시각화하여 보였으며 검증으로 인공지능 모듈의 정확성을 측정하였다. 끝으로 우리가 개발한 다기관 공동연구 지원플랫폼에 적용하여 시각화된 결과를 보였다.