• The Journal of the Acoustical Society of Korea
• /
• 제28권3E호
• /
• pp.109-117
• /
• 2009

Identity recognition in real environment with a reliable mode is a key issue in human computer interaction (HCI). In this paper, we present a robust person identification system considering score-based optimal reliability measure of audio-visual modalities. We propose an extension of the modified reliability function by introducing optimizing parameters for both of audio and visual modalities. For degradation of visual signals, we have applied JPEG compression to test images. In addition, for creating mismatch in between enrollment and test session, acoustic Babble noises and artificial illumination have been added to test audio and visual signals, respectively. Local PCA has been used on both modalities to reduce the dimension of feature vector. We have applied a swarm intelligence algorithm, i.e., particle swarm optimization for optimizing the modified convection function's optimizing parameters. The overall person identification experiments are performed using VidTimit DB. Experimental results show that our proposed optimal reliability measures have effectively enhanced the identification accuracy of 7.73% and 8.18% at different illumination direction to visual signal and consequent Babble noises to audio signal, respectively, in comparison with the best classifier system in the fusion system and maintained the modality reliability statistics in terms of its performance; it thus verified the consistency of the proposed extension.

• 정보과학회 논문지
• /
• 제43권11호
• /
• pp.1281-1297
• /
• 2016

최근 스마트 기기의 보급으로 자유롭게 비디오 컨텐츠를 생성하고 이를 빠르고 편리하게 공유할 수 있는 네트워크 환경이 갖추어지면서, 퍼스널 비디오가 급증하고 있다. 그러나, 퍼스널 비디오는 비디오라는 특성 상 멀티 모달리티로 구성되어 있으면서 데이터가 시간의 흐름에 따라 변화하기 때문에 이벤트 분류를 할 때 이에 대한 고려가 필요하다. 본 논문에서는 비디오 내의 멀티 모달리티들로부터 고수준의 특징을 추출하여 시간 순으로 재배열한 것을 바탕으로 모달리티 사이의 연관관계를 Deep Neural Network(DNN)으로 학습하여 퍼스널 비디오 이벤트를 분류하는 방법을 제안한다. 제안하는 방법은 비디오에 내포된 이미지와 오디오를 시간적으로 동기화하여 추출한 후 GoogLeNet과 Multi-Layer Perceptron(MLP)을 이용하여 각각 고수준 정보를 추출한다. 그리고 이들을 비디오에 표현된 시간순으로 재 배열하여 비디오 한 편당 하나의 특징으로 재 생성하고 이를 바탕으로 학습한 DNN을 이용하여 퍼스널 비디오 이벤트를 분류한다.

• 정보처리학회논문지A
• /
• 제11A권5호
• /
• pp.391-400
• /
• 2004

서로 다른 종류의 영상을 정확하게 연관시켜 복합적인 정보를 제공하는 다중모달리티 의료 영상정합기법 중 표면정보 기반 영상정합에서는 일반적으로 동일 대상에 대한 서로 다른 모달리티에서 추출된 표면 윤곽정보 사이의 거리를 최소화함으로써 매칭이 이루어진다. 그런데 동일대상에 대해 취득되는 서로 다른 두 모달리티는 관심 영역 상의 표면 특성이 서로 유사하다. 그러므로 다중모달리티 영상정합에서 표면거리와 함께 표면의 형태 특성을 고려하여 두 영상을 매칭하는 방법이 정합결과의 정확도를 향상시킬 수 있다. 본 연구에서는 동일 대상의 서로 다른 두 모달리티 뇌영상 간의 표면거리와 표면곡률을 최적화하는 정합기법을 제안한다. 영상정합은 참조영상과 테스트영상에 대한 표면정보 생성과 이 두 개의 표면정보를 최적화하는 단계로 구성된다. 표면정보 생성 단계에서는 두 모달리티로부터 관심영역의 윤곽선을 추출하고, 이 중 참조 볼륨의 윤곽선에 대해서는 표면거리맵과 표면곡률맵을 구성하게 된다. 최적화 단계에서는 표면거리맵과 표면곡률맵을 참조하는 최적화 평가함수(cost function)에 의해 두 객체의 표면거리 차이와 표면곡률 차이를 최소화하는 정합 변환 값이 결정되고, 이것이 테스트영상의 변환에 적용되어 결과적으로 두 영상이 정합 되게 된다. 제안된 최적화 평가함수는 표면거리 정보만을 사용하는 평가함수에 비해 보다 견고한 정합 정확도를 보였으며 또한 본 연구는 정합결과의 볼륨 가시화를 통해 효율적인 영상 분석 수단을 제공하고자 하였다.

• Nuclear Medicine and Molecular Imaging
• /
• 제42권sup1호
• /
• pp.157-161
• /
• 2008

Malignant pleural mesothelioma (MPM) has a poor prognosis and a strong association with exposure to asbestos. Although there are not generally accepted guidelines for treatment of MPM, recent reports suggest that multi modality therapy combining chemotherapy, radiotherapy, and surgery can improve the survival of patients with MPM. Therefore exact staging is required to decide the best treatment option. However, it is well known that there are many difficulties in determining precise preoperative stage, predicting prognosis, and monitoring response to therapy with conventional imaging modalities such as CT and MRI in MPM. Recently PET with $^{18}F-FDG$ comes into the spotlight as an important staging method. There is increasing evidence that PET is superior to other conventional imaging modalities in diagnosis and staging of MPM. Particularly PET/CT improves the diagnostic and staging accuracy over PET or CT alone in MPM because it provides anatomic imaging data as well as functional information. PET and PET/CT are also useful for monitoring response to therapy and SUV is reported as a prognostic factor in MPM.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2003년도 10월 학술대회지
• /
• pp.33-36
• /
• 2003

Two experiments were conducted to support the suggestion that the same information processing was used in both input modalities, visual and auditory modality in Wonil Choi & Kichun Nam(2003)'s paper. The primed lexical decision task was performed and pseudoword prime stimuli were used. The result was that priming effect did not occur in any experimental condition. This result might be interpreted visual facilitative information and phonological inhibitory information cancelled each other.

• Journal of information and communication convergence engineering
• /
• 제2권3호
• /
• pp.167-171
• /
• 2004

The evolution of modern imaging modalities, followed by the rapid development of computer technology has introduced many new features in the communication networks used in medical facilities. Since it is very important to keep patient's record accurately, the ability to exchange medical data securely over the communication network is essential for any medical information. In this paper, therefore, we introduce some problems which occur from digitizing medical images such as MRI (Magnetic Resonance Imaging), CT (Computed Tomography), CR(Computed Radiography), etc., and then we propose a authentication mechanism for medical image verification using secret watermark images.

• 한국간호교육학회지
• /
• 제9권2호
• /
• pp.244-252
• /
• 2003

This study was intented to identify the differences of fatigue and quality of life in patients with breast cancer according to the different treatment modalities. Ninety-one subjects were recruited from a medical unit and a therapeutic radiologic clinic at a university hospital in Chungnam, Korea. Fatigue was measured using Lee's scale(1999) translated from Piper's Fatigue Scale and quality of life was measured using Yang's scale(2002). The women with breast cancer receiving chemotherapy were less fatigued and had better quality of life than those receiving radiotherapy(t=-2.914, p=.005; t=2.565, p=.012, respectively). The fatigue and quality of life in the women with breast cancer undergoing cancer treatments were influenced by the weight change during the past 3 months(F=24.70, p=.039). Cancer stage, metastasis and weight were not associated with fatigue and quality of life. Health professionals can help cancer patients undergoing treatments prepare to cope with the expected side effects by providing a specific treatment-related information to them. The findings of this study can contribute the knowledge of these information.

• 통상정보연구
• /
• 제9권1호
• /
• pp.277-295
• /
• 2007

This paper presents the ODR phenomenon with a specific focus on regulatory problems. It argues that the current regulatory framework for online dispute resolution is, to a large extent, defective. Existing deficiencies result not only from a lack of comprehensive ODR law, but also from the weakness of the other modalities of regulation : norms and market. The current today's approach to regulating ODR has been unsuccessful, and it is time to re-examine that position. I would like to say that suggesting details of the optimal ODR regulatory framework is beyond the scope of this paper. From a broader perspective, the field of ODR seems to be not sufficiently regulated. Neither by law, nor by the restrains set by other norms or market. Given the current weakness of the modalities of regulation, online dispute resolution requires greater institutional and international support. It appears that adoption of internationally harmonized ODR law would not hamper the development of norms and market, but rather support their proper growth. As a result, the co-regulation approach recommended in this paper is different from both traditional state-oriented models of regulation and self-regulatory systems. Arguably, this co-regulation approach is the optimal model for regulation of online dispute resolution under current circumstances.

• Nuclear Medicine and Molecular Imaging
• /
• 제42권sup1호
• /
• pp.101-109
• /
• 2008

Cervix cancer is one of common gynecological cancers in the world, and staged with FIGO or TNM system. However, these clinical staging systems lack information about lymph node or distant metastases, thus imaging modalities are considered to make an appropriate therapeutic plan and enhance overall survival rate. In this context, FDG PET is recommended to pre-treatment stating and prognosis prediction, for it could noninvasively evaluate the status of lymph nodes, especially abdominal paraaortic nodes which are closely related with prognosis. Moreover, the degree of FDG uptake is correlated with prognosis. Although there is no consistent method for surveillance of cervix cancer, FDG PET seems a very important tool in detecting tumor recurrence because it is much more advantageous than conventional imaging modalities such as MRI for discerning recurrent tumor from fibrosis caused by radiation or surgery. Furthermore, FDG PET could be used to evaluate treatment response. On the other hand, recently introduced PET/CT is expected to play an ancillary role to FIGO staging by adding anatomical information, and enhance diagnostic performance of PET by decreasing false positive findings.

• Biomolecules & Therapeutics
• /
• 제26권1호
• /
• pp.81-92
• /
• 2018

It is widely accepted that altered metabolism contributes to cancer growth and has been described as a hallmark of cancer. Our view and understanding of cancer metabolism has expanded at a rapid pace, however, there remains a need to study metabolic dependencies of human cancer in vivo. Recent studies have sought to utilize multi-modality imaging (MMI) techniques in order to build a more detailed and comprehensive understanding of cancer metabolism. MMI combines several in vivo techniques that can provide complementary information related to cancer metabolism. We describe several non-invasive imaging techniques that provide both anatomical and functional information related to tumor metabolism. These imaging modalities include: positron emission tomography (PET), computed tomography (CT), magnetic resonance imaging (MRI), magnetic resonance spectroscopy (MRS) that uses hyperpolarized probes and optical imaging utilizing bioluminescence and quantification of light emitted. We describe how these imaging modalities can be combined with mass spectrometry and quantitative immunochemistry to obtain more complete picture of cancer metabolism. In vivo studies of tumor metabolism are emerging in the field and represent an important component to our understanding of how metabolism shapes and defines cancer initiation, progression and response to treatment. In this review we describe in vivo based studies of cancer metabolism that have taken advantage of MMI in both pre-clinical and clinical studies. MMI promises to advance our understanding of cancer metabolism in both basic research and clinical settings with the ultimate goal of improving detection, diagnosis and treatment of cancer patients.