• Radiation Oncology Journal
• /
• 제32권3호
• /
• pp.125-131
• /
• 2014

Purpose: We reviewed treatment outcomes and prognostic factors for patients with salivary ductal carcinoma (SDC) treated with surgery and postoperative radiotherapy from 2005 to 2012. Materials and Methods: A total of 16 patients were identified and 15 eligible patients were included in analysis. Median age was 61 years (range, 40 to 71 years) and 12 patients (80%) were men. Twelve patients (80%) had a tumor in the parotid gland, 9 (60%) had T3 or T4 disease, and 9 (60%) had positive nodal disease. All patients underwent surgery and postoperative radiotherapy. Postoperative radiotherapy was delivered using 3-dimensional conformal radiotherapy or intensity-modulated radiotherapy. Locoregional failure-free survival (LRFFS), distant failure-free survival (DFFS), progression-free survival (PFS), and overall survival (OS) were calculated using the Kaplan-Meier method. Differences in survival based on risk factors were tested using a log-rank test. Results: Median total radiotherapy dose was 60 Gy (range, 52.5 to 63.6 Gy). Four patients received concurrent weekly chemotherapy with cisplatin. Among 10 patients who underwent surgery with neck dissection, 7 received modified radical neck dissection. With a median follow-up time of 38 months (range, 24 to 105 months), 4-year rates were 86% for LRFFS, 51% for DFFS, 46% for PFS, and 93% for OS. Local failure was observed in 2 patients (13%), and distant failure was observed in 7 (47%). The lung was the most common involved site of distant metastasis. Conclusion: Surgery and postoperative radiotherapy in SDC patients resulted in good local control, but high distant metastasis remained a major challenge.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.299-302
• /
• 2001

The statistical analysis of the feature extraction and the neural networks are proposed to recognize a human face. In the preprocessing step, the normalized skin color map with Gaussian functions is employed to extract the region of face candidate. The feature information in the region of the face candidate is used to detect the face region. In the recognition step, as a tested, the 120 images of 10 persons are trained by the backpropagation algorithm. The images of each person are obtained from the various direction, pose, and facial expression. Input variables of the neural networks are the geometrical feature information and the feature information that comes from the eigenface spaces. The simulation results of$.$10 persons show that the proposed method yields high recognition rates.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2009년도 정보 및 제어 심포지움 논문집
• /
• pp.408-410
• /
• 2009

We present an efficient and robust measurement model for visual tracking. This approach builds on and extends work on subspace representations of measurement model. Subspace-based tracking algorithms have been introduced to visual tracking literature for a decade and show considerable tracking performance due to its robustness in matching. However the measures used in their measurement models are often restricted to few approaches. We propose a novel measure of object matching using Angle In Feature Space, which aims to improve the discriminability of matching in subspace. Therefore, our tracking algorithm can distinguish target from similar background clutters which often cause erroneous drift by conventional Distance From Feature Space measure. Experiments demonstrate the effectiveness of the proposed tracking algorithm under severe cluttered background.

• 한국지능시스템학회논문지
• /
• 제11권9호
• /
• pp.822-826
• /
• 2001

The statistical analysis of the feature extraction and the neural networks are proposed to recognize a human face. In the preprocessing step, the normalized skin color map with Gaussian functions is employed to extract the region of face candidate. The feature information in the region of the face candidate is used to detect the face region. In the recognition step, as a tested, the 120 images of 10 persons are trained by the backpropagation algorithm. The images of each person are obtained from the various direction, pose, and facial expression. Input variables of the neural networks are the geometrical feature information and the feature information that comes from the eigenface spaces. The simulation results of 10 persons show that the proposed method yields high recognition rates.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2004년도 춘계종합학술대회
• /
• pp.608-611
• /
• 2004

본 논문에서는 고유얼굴 특성에 기반한 강건한 얼굴 인식 기술을 제안한다. 전형적인 고유얼굴 인식방법은 학습영역에서 고유얼굴을 생성시키고, 모든 학습영상을 이 얼굴공간에 투영시켜 각각의 사람마다 저장된 성분들을 비교하거나 상관시켜 특징들을 추출합니다. 복잡한 배경에 있는 얼굴들을 인식할 때 EFR방법은 얼굴인식에는 강하지만, 얼굴과 배경들 사이의 구분을 실패하게 된다 배경에서 강건한 얼굴인식을 위해서 배경패턴을 학습하며, 배경영역은 배경패턴으로부터 생성되어 얼굴영역과 함께 얼굴 인식을 위하여 사용된다. 본 논문에서 제안한 방법이 EFR방법보다 성능과 복잡한 배경하에서 매우 좋은 결과를 나타냄을 확인할 수 있었다.