[논문]후두내시경 영상에서의 라디오믹스에 의한 병변 분류 연구

박준하; 김영재; 우주현; 김광기

doi:10.9718/jber.2022.43.5.353

후두내시경 영상에서의 라디오믹스에 의한 병변 분류 연구
Research on the Lesion Classification by Radiomics in Laryngoscopy Image 원문보기

Journal of biomedical engineering research : the official journal of the Korean Society of Medical & Biological Engineering, v.43 no.5, 2022년, pp.353 - 360

박준하 (가천대학교 보건과학대학 의용생체공학과) , 김영재 (가천대학교 보건과학대학 의용생체공학과) , 우주현 (가천대길병원 이비인후과) , 김광기 (가천대학교 보건과학대학 의용생체공학과)

Abstract ▼ AI-Helper

Laryngeal disease harms quality of life, and laryngoscopy is critical in identifying causative lesions. This study extracts and analyzes using radiomics quantitative features from the lesion in laryngoscopy images and will fit and validate a classifier for finding meaningful features. Searching the region of interest for lesions not classified by the YOLOv5 model, features are extracted with radionics. Selected the extracted features are through a combination of three feature selectors, and three estimator models. Through the selected features, trained and verified two classification models, Random Forest and Gradient Boosting, and found meaningful features. The combination of SFS, LASSO, and RF shows the highest performance with an accuracy of 0.90 and AUROC 0.96. Model using features to select by SFM, or RIDGE was low lower performance than other things. Classification of larynx lesions through radiomics looks effective. But it should use various feature selection methods and minimize data loss as losing color data.

주제어

표/그림 (8)

그림 그림 1. 후두내시경 영상 속 라디오믹스를 통한 병변 분류 연구 흐름도 Fig. 1. Study of lesion classification by radiomics in laryngoscopy image flowchart
표 표 1. 데이터 세트의 구성, a: 병변 탐지 모델 학습에 사용된 데이터 세트 구성 b: 병변 분류 모델 학습에 사용된 데이터 세트 구성 Table 1. Data set composition, a: Composition of data set used to train detect lesions model, b: Composition of data set used to train lesions classification model
그림 그림 2. YOLOv5 병변 검출 결과 예시 이미지, (a) TP, (b) FN, (c) TN, 노란색 경계 상자: 정답 ROI, 청록색 경계 상자: 학습시킨 YOLOv5 가 찾은 ROI 경계 Fig. 2. YOLOv5 Examples of lesion detection results, (a) TP, (b) FN, (c) TN, Yellow bounding box: ground truth ROI, Cyan bounding box: ROI found by YOLOv5
그림 그림 3. 각 요소를 포함하는 모든 모델의 정확도, AUROC 점수의 분포를 나타냈다. 파란색: 정확도, 초록색: AUROC, x: 평균 Fig. 3. The accuracy and AUROC scores distribution of all models including each element are shown. Blue: Accuracy, Green: AUROC, x: mean
표 표 2. YOLOv5의 병변 검출 정밀도, 재현율, F1-점수, 정확도 결과 Table 2. Precision, recall, F1-score, accuracy of lesion detection by YOLOv5
표 표 3. 후두 병변의 분류에 유용한 라디오믹스 특징을 찾기 위해, 18가지의 방법으로 특징을 선택, 분류한 결과이다. 제시된 지표는 정확도와 AUROC이다. 각지표의 최댓값을 굵은 글꼴 처리하였다. Table 3. The result of selecting and classifying features using 18 methods to find radiomics features useful for lesion classification. accuracy and AUROC presented. The maximum value was bold. RF: Random Forest, GB: Gradiant Boosting, SFM: Select From Model, RFE: Recursive Feature Elimination, SFS: SequentialFeatureSelector
표 표 4. SFS와 LASSO, RF 조합의 양성 및 악성의 예측 확률 t 검정 결과 N: 관측 수, M: 평균, SD: 표준편차 Table 4. t-test result of predict probability of benign and malignant in SFS, LASSO, and RF N: The sample size, M: mean, SD: Standard deviation
그림 그림 4. 후두 병변 분류에 유의한 특징을 찾기 위해 SFS와 LASSO, RF 조합의 특징 가중치를 분석한 결과. 유의한 가중치를 가지는 상위 4개의 특징을 보여준다 Fig. 4. The result of feature weights in SFS, LASSO, and RF to find Significant features in the classification of laryngeal lesions. Its shows the top 4 features having significant weight. GLRLM: Gray Level Run Length Metrix, Run Length Non- Uniformity - RLN, Robust Mean Absolute Deviation - rMAD

참고문헌 (39)

Cohen SM, Dupont WD, Courey MS. Quality-of-life impact of non-neoplastic voice disorders: a meta-analysis. Annals of Otology, Rhinology & Laryngology. 2006;115(2):128-134.

상세보기
Kim JO, Lim SE, Park SY, Choi SH, Choi JN, Choi HS. Validity and reliability of Korean-version of voice handicap index and voice-related quality of life. Speech Sciences. 2007;14(3):111-125.
Ni XG, He S, Xu ZG, Gao L, Lu N, Yuan Z Lai S-Q, Zhang T-M, Yi J-L, Wang X-L. Endoscopic diagnosis of laryngeal cancer and precancerous lesions by narrow band imaging. The Journal of Laryngology & Otology. 2010;125(3):288-296.
Lee SH. Diagnostic Role of Stroboscopy. Journal of the Korean Society of Laryngology. Phoniatrics and Logopedics. 2010;21(1):13-16.
Chu EA, Kim YJ. "Laryngeal cancer: diagnosis and preoperative work-up". Otolaryngologic Clinics of North America. 2008;41(4):673-695.

상세보기
Ren J, Jing X, Wang J, Ren X, Xu Y, Yang Q, Ma L, Sun Y, Xu W, Yang N, Zou J, Zheng Y, Chen M, Gan W, Xiang T, An J, Liu R, Lv C, Lin K, Zheng X, Lou F, Rao Y, Yang Hui, Liu Kai, Liu G, Lu T, Zheng X, Zhao Y. Automatic recognition of laryngoscopic images using a deep-learning technique. The Laryngoscope. 2020;130(11):E686-E693.

상세보기
Cho WK, Choi SH. Comparison of convolutional neural network models for determination of vocal fold normality in laryngoscopic images. Journal of Voice. 2020;36(5):590-598.
Xiong H, Lin P, Yu JG, Ye J, Xiao L, Tao Y, Jiang Z, Lin W, Liu M, Xu J, Hu W, Lu Y, Liu H, Li Y, Zheng Y, Yang H. Computeraided diagnosis of laryngeal cancer via deep learning based on laryngoscopic images. EBioMedicine. 2019;48:92-99.

상세보기
Azam MA, Sampieri C, Loppi A, Africano S, Vallin A, Mocellin D, Fragale M, Guastini L, Moccia S, Piazza C, Mattos L, Peretti G. Deep Learning Applied to White Light and Narrow Band Imaging Videolaryngoscopy: Toward Real-Time Laryngeal Cancer Detection. The Laryngoscope. 2021;132(9):1798-1806.
Ye G, Du C, Lin T, Yan Y, Jiang J. Deep Learning for Laryngopharyngeal Reflux Diagnosis. Applied Sciences. 2021;11(11):4753.
Onji K, Yoshida S, Tanaka S, Kawase R, Takemura Y, Oka S, Tamaki T, Raytchev B, Kaneda K, Yoshihara M, Chayama K. Quantitative analysis of colorectal lesions observed on magnified endoscopy images. J Gastroenterol. 2011;46:1382-1390.
Sainju S, Bui FM, Wahid K. Bleeding detection in wireless capsule endoscopy based on color features from histogram probability. 2013 26th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE) IEEE. 2013;1-4
Vu DH, Nguyen LT, Nguyen VT, Tran TH, Dao VH, Vu H. Boundary delineation of reflux esophagitis lesions from endoscopic images using color and texture. 2021 International Conference on Multimedia Analysis and Pattern. 2021;1-6.
Kumar V, Gu Y, Basu S, Berglund A, Eschrich SA, Schabath MB, Foster K, Aerts HJWL, Dekker A, Fenstermacher D, Goldgof DB, Hall LO, Lambin P, Balagurunathan Y, Gatenby RA, Gillies RJ. Radiomics: the process and the challenges. Magnetic Resonance Imaging. 2012;30(9):1234-1248.

상세보기
Griethuysen JJMV, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, Beets-Tan RGH, Fillion-Robin JC, Pieper S, Aerts HJWL. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer research. 2017;77(21):e104-e107.

상세보기
Machicado JD, Koay EJ, Krishna SG. Radiomics for the Diagnosis and Differentiation of Pancreatic Cystic Lesions. Diagnostics. 2020;10(7):505.

상세보기
Yip SSF, Aerts HJWL. Applications and limitations of radiomics. Physics in Medicine & Biology. 2016;61(13):R150.

상세보기
Liang C, Huang Y, He L, Chen X, Ma Z, Dong D, Tian J, Liang C, Liu Z. The development and validation of a CT-based radiomics signature for the preoperative discrimination of stage III and stage III-IV colorectal cancer. Oncotarget. 2016;7(21):31401-31412.

상세보기
Zhang Y, Oikonomou A, Wong A, Haider MA, Khalvati F. Radiomics-based prognosis analysis for non-small cell lung cancer. Scientific reports. 2017;7(1):1-8.
Coroller TP, Grossmann P, Hou Y, Velazquez ER, Leijenaar RTH, Hermann G, Lambin P, Haibe-Kains B, Mak RH, Aerts HJWL. CT-based radiomic signature predicts distant metastasis in lung adenocarcinoma. Radiotherapy and Oncology. 2015;114(3)345-350.

상세보기
Kniep HC, Madesta F, Schneider T, Hanning U, Schonfeld MH, Schon G, Fiehler J, Gauer T, Werner R, Gellissen S. Radiomics of brain MRI: utility in prediction of metastatic tumor type. Radiology. 2019;290(2):479-487.

상세보기
Kandemirli SG, Chopra S, Pyiya S, Ward C, Locke T, Soni N, Srivastava S, Jones K, Bathla G. Presurgical detection of brain invasion status in meningiomas based on first-order histogram-based texture analysis of contrast enhanced imaging. Clinical neurology and neurosurgery. 2020;198:106205.

상세보기
Mohanaiah P, Sathyanarayana P, GuruKumar I. Image texture feature extraction using GLCM approach. International journal of scientific and research publications. 2013;3(5):1-5.
Ahmadi N, Akbarizadeh G. Iris tissue recognition based on GLDM feature extraction and hybrid MLPNN-ICA classifier. Neural Computing and Applications. 2020;32(7):2267-2281.

상세보기
Sohail ASM, Bhattacharya P, Mudur SP, Krishnamurthy S. Local relative GLRLM-based texture feature extraction for classifying ultrasound medical images. 2011 24th Canadian Conference on Electrical and Computer Engineering (CCECE). 2011;001092-001095.
Thibault G, Angulo J, Meyer F. Advanced statistical matrices for texture characterization: Application to DNA chromatin and microtubule network classification. 2011 18th IEEE International Conference on Image Processing. 2011;53-56.
Amadasun M, King R. Textural features corresponding to textural properties. in IEEE Transactions on Systems, Man, and Cybernetics. 1989;19(5):1264-1274.

상세보기
Buitinck L, Louppe G, Blondel M, Pedregosa F, Mueller A, Grisel O, Niculae V, Prettenhofer P, Gramfort A, Grobler J, Layton R, Vanderplas J, Joly A, Varoquaux G. API design for machine learning software: experiences from the scikit-learn project. arXiv preprint arXiv. 2013;1309.0238.
Chandrashekar G, Sahin F. A survey on feature selection methods. Computers & Electrical Engineering. 40.1 (2014):16-28.

상세보기
https://scikit-learn.org/stable/modules/feature_selection.html, Accessed on 29 Jun 2022.
https://scikit-learn.org/stable/modules/linear_model.html. Accessed on 1 Aug 2022.
McDonald GC. Ridge regression. Wiley Interdisciplinary Reviews: Computational Statistics. 2009;1(1):93-100.
https://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html sklearn.svm.LinearSVC. Accessed on 30 Jun 2022.
Biau G, Scornet E. A random forest guided tour. TEST. 2016;25:197-227.
Natekin A, and Knoll A. Gradient boosting machines, a tutorial. Frontiers in neurorobotics. 2013;7:21.

상세보기
Antoine AA. The student's t-test: a brief description. Research & Reviews: Journal of Hospital and Clinical Pharmacy. 2019;5(1):1.
https://pyradiomics.readthedocs.io/en/latest/features.html. Accessed on 4 Jul 2022.
Mazor T, et al. Intratumoral heterogeneity of the epigenome. Cancer cell. 2016;29(4):440-451.

상세보기
Aerts HJWL, Pankov A, Song JS, Costello JF. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nature communications. 2014;5(1):1-9.

저자의 다른 논문 :

표제어: PCR

동의어: Packet Collision Rate

용어 설명 출처 목록 (6)

용어 설명: PCR은 세균 특이성이 있는 primer를 이용하여 적은 수의 세균이 있을지라도 쉽게 검출할 수 있는 유용한 방법이며, 이를 이용하여 구강 내 치면세균막이나 타액에서 직접 세균을 검출할 수 있게 되었다[8].

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증