CNN 알고리즘을 이용한 의료 영상 데이터 크기에 따른 정확성 분석
Accuracy analysis according to the size of medical image data using CNN algorithm
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
최근 인공지능에 대한 관심과 활용이 증가하면서 의료 산업의 패러다임도 치료나 병원중심의 이용에서 예방과 이용자 중심으로 변화되고 있다. 이와 관련된 기술발전은 의료 빅데이터 구축과 활용을 통한 서비스 향상 및 발전으로 이어지는 사례들로 많이 구축 활용되고 있다. 이에 본 연구에서는9천여 장의 의료 영상 데이터를 기반으로 입력 영상 크기 및 훈련 회수에 따라 정확성 값을 비교하여높은 결과 값을 갖는 입력 영상 데이터 조건을 찾는다. 이렇게 함으로써 소규모 의료 영상 데이터를이용하여 높은 정확성을 확보함으로써 적은 비용으로도 질병 유무 판단에 활용할 수 있을 것이다. 이를 위해 2차원 영상 데이터 학습과 정확성 분석에 효과적인 CNN 알고리즘을 사용하여 진행한다.
Abstract
Recently, as interest in and use of artificial intelligence increases, the paradigm of the medical industry is also changing from treatment or hospital-centered use to prevention and user-centered use. Technological development related to this is being built and utilized in many cases leading to service improvement and development through the establishment and use of medical big data. Therefore, in this study, based on about 9,000 medical image data, the accuracy value is compared according to the input image size and the number of training times to find the input image data condition with a high result value. In this way, it will be possible to use small-scale medical image data to secure high accuracy and use it to determine the presence or absence of disease at low cost. To this end, we use a CNN algorithm that is effective for 2D image data learning and accuracy analysis.
참고문헌 (15)
- "Big Data Definition"[URL]
- S-H. Lee, "Artificial intelligence technology trend based on medical big data", Journal of Communication Association, 37(9), pp.85~91 (2020)
- J-K. Bae, "A study on the application of artificial intelligence technology in the smart healthcare industry", Journal of Korea Logos Management Association, 19(4), pp.179~196 (2021)
- K. Jung, "Artificial intelligence-based medical image analysis technology", pp.195~208 (2018), IITP
- D. Laney, "3D Data Management: Controlling Data Volume, Velocity and Variety", META Group Research Note, 6 (2001)
- A. Ahmed, "Learning to unlearn: Building immunity to dataset bias in medical imaging studies" (2018), Machine Learning for Health
- "The deep Poincare map: A novel approach for left ventricle segmentation" (2018)[URL]
- G. Litjens, "A Survey on Deep Learning in Medical Image Analysis", Medical image analysis, 42, pp.60~88 (2017)
- R. Yamashita, "Convolutional neural networks: an overview and application in radiology", Insights into Imaging, 9, pp.611~629 (2018)
- Y. LeCun, "Deep learning", Nature, 521, pp.436~444 (2015)
- A. Esteva, "Dermatologist-level classification of skin cancer with deep neural networks", Nature, 542, pp.115~118 (2017)
- E. B. Bejnordi, "Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer", JAMA, 318(22), pp.2199~2210 (2017)
- L. Paras, "Deep learning at chest radiography : automated classification of pulmonary tuberculosis by using convolutional neural networks", Radiology, 284(2), pp.574~582 (2017)
- Y. Koichiro, "Deep learning with convolutional neural network for differentiation of liver masses at dynamic contrast-enhanced CT : a preliminary study", Radiology, 286(3), pp.887~896 (2017)
- C. Kenneth, "The Cancer Imaging Archive(TCIA) : maintaining and operating a public information repository", J Digit Imaging, 26(6), pp.1045~1057 (2013)