
http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
SwinResNet: Swin Transformer와 ResNet 융합을 통한 Volumetric 의료 영상 분할
최성호,박경범,이재열 한국CDE학회 2023 한국CDE학회 논문집 Vol.28 No.3
Volumetric medical image segmentation is critical in diagnosing diseases and planning subse- quent treatment. The convolutional neural network (CNN)-based U-Net was proposed for con- ducting accurate and robust medical image segmentation since the skip connection of U-Net and deep feature representation significantly improved its performance. However, since CNN-based models mainly focus on local and low-level features, they cannot extract global and high-level features effectively. Meanwhile, the Vision Transformer developed in natural language process- ing is proposed to improve image classification performance by splitting an input image into patches and conducting linear embeddings of the patches, which can extract global features. However, the Vision Transformer has difficulty in handling detailed and low-level features. This study proposes SwinResNet which can effectively conduct volumetric medical image segmenta- tion by fusing the Swin Transformer and CNN models. The combination can take advantage of both models and complement each other. Swin Transformer and ResNet are used as encoders, and the receptive field blocks and aggregation modules are applied to the multi-level features extracted from both encoders. Comprehensive evaluation shows that the proposed approach out- performs well-known previous studies.
최성호,정정훈,정상원 한국질적탐구학회 2016 질적탐구 Vol.2 No.1
This study was aimed to discuss the theoretical aspects of qualitative content analysis to provide deep and detailed understanding about qualitative data analysis, as it is now getting growing attention in Korea. Although discussions for qualitative research have become more detailed and extensive and there have been many previous studies that adopted qualitative content analysis as an analytic method for data analysis, theoretical discussions about qualitative content analysis have not been sufficient in Korea. So this study was designed to review the concept, development process and scholars' methodological procedures of qualitative content analysis. In conclusion, this study assessed and found first, qualitative content analysis can be used as an effective method to analyze messages from mass media, second, a balanced approach is required between qualitative and quantitative aspects, third, the consideration of the procedures in content analysis can ensure researchers to secure the validity of the analytic results, and fourth, in spite of the third finding, a cautious approach is required before taking the procedures as fixed rules. The discussion about qualitative content analysis is expected to provide extensive insights to data analysis of qualitative researchers. 이 연구는 최근 관심이 높아지고 있는 질적 자료 분석에 대해 좀 더 깊고 세분화된 이해를 도모하기 위해 질적 내용 분석에 대해 이론적으로 논의하였다. 최근 질적 연구에 대한 이론적 관심이 세분화 되고 있는 측면을 고려한다면 질적 분석의 한 접근으로서 질적 내용 분석은 우리나라에서 많은 연구에 사용되고 있음에도 불구하고 이에 대한 이론적 논의가 부족하였다. 따라서 이 연구에서는 질적 내용 분석의 개념, 발달과정, 그리고 그 분석 단계에 대한 학자들의 논의를 살펴보았다. 이러한 논의의 결과, 첫째, 질적 내용분석은 대중매체 속의 메시지 분석에 효과적으로 사용될 수 있으며, 둘째, 내용분석에 있어 질적, 양적 측면의 균형이 필요하며, 셋째, 내용분석에 있어서 절차에 대한 고려는 분석 결과의 타당성 확보를 위한 효과적인 도구가 될 수 있다는 점, 넷째, 그럼에도 불구하고 이러한 절차를 정형화된 규칙으로 받아들여서는 안 된다는 점을 논의하였다. 이러한 질적 자료 분석에 대한 이론적 논의는 질적 연구자들의 자료분석에 적절한 시사점을 제공해 줄 수 있을 것이다.