메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색
질문

논문 기본 정보

자료유형
학술대회자료
저자정보
Dawei Luo (Jeonbuk National University) Joonsoo Bae (Jeonbuk National University) Yoonhyuck Woo (Jeonbuk National University) Jang Hyun Baek (Jeonbuk National University)
저널정보
대한산업공학회 대한산업공학회 추계학술대회 논문집 2022년 대한산업공학회 추계학술대회
발행연도
2022.11
수록면
3,399 - 3,410 (12page)

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
🏆
연구결과
AI에게 요청하기
추천
검색
질문

초록· 키워드

오류제보하기
Segmenting lung nodules in medical images and classifying them into benign and malignant categories are vital to the diagnosis and treatment of cancerous patients. Much research effort has been devoted to building state-of-art models which process the input images and output the final result. So how to choose the proper metrics to compare the performance of different models is quite challenging. In the pulmonary nodule classification process, the number of negative samples is much higher than that of positive ones, leading to imbalanced datasets, which impose a negative effect on classifiers and make it trickier for metrics to assess the performance of different classifiers. This paper analyzed the most common-used metrics (e.g., Geometric Mean, F score, Matthew Correlation Coefficients, Area Under ROC Curve) in the context of lung nodule detection and assessed their characteristics, finding weaknesses in these metrics. After that, some modified metrics were proposed according to the analyses above. At last, we used these modified metrics to assess different augmentation methods for nodule classification tasks. Both the result of the theoretical analyses and that of the above experiments proved that modified metrics can make a better assessment of different augmentation methods.

목차

Abstract
1. Introduction
2. Literature review
3. Performance metrics for binary classification
4. Augmentation methods for 3D data
5. Results and discussion
6. Conclusions
7. References

참고문헌 (0)

참고문헌 신청

함께 읽어보면 좋을 논문

논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!

이 논문의 저자 정보

최근 본 자료

전체보기

댓글(0)

0

UCI(KEPA) : I410-ECN-0101-2023-530-000169452