Deep learning assessment of breast terminal duct lobular unit involution: Towards automated prediction of breast cancer risk

Autor:	Rulla M. Tamimi, Christina Luffman, Suzanne C. Wetstein, Kevin H. Kensler, Marinus Bastiaan Van Leeuwen, Gabrielle M. Baker, Allison M. Onken, Laura C. Collins, Ruud Vlutters, Ying Liu, Stuart J. Schnitt, Josien P. W. Pluim, Michael E. Pyle, Bart Bakker, Mitko Veta, Yujing J. Heng
Přispěvatelé:	Medical Image Analysis, EAISI Health
Jazyk:	angličtina
Rok vydání:	2020
Předmět:	FOS: Computer and information sciences 0301 basic medicine Medical Doctors Health Care Providers Biopsy Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Nurses SDG 3 – Goede gezondheid en welzijn Cohort Studies 0302 clinical medicine Risk Factors Breast Tumors Medicine and Health Sciences Image Processing Computer-Assisted Medical Personnel Breast Multidisciplinary Breast tissue medicine.diagnostic_test Image and Video Processing (eess.IV) Age Factors Qualitative Studies Middle Aged 3. Good health Professions medicine.anatomical_structure Adipose Tissue Oncology Research Design 030220 oncology & carcinogenesis Medicine Benign Breast Conditions Female Radiology Anatomy Research Article Adult Computer and Information Sciences medicine.medical_specialty Neural Networks Science Breast Neoplasms Research and Analysis Methods Risk Assessment 03 medical and health sciences Computational pathology Deep Learning Breast cancer SDG 3 - Good Health and Well-being Breast Cancer FOS: Electrical engineering electronic engineering information engineering medicine Humans Involution (medicine) business.industry Reproductive System Biology and Life Sciences Cancers and Neoplasms Reproducibility of Results Electrical Engineering and Systems Science - Image and Video Processing medicine.disease Large cohort Pathologists Health Care Biological Tissue 030104 developmental biology People and Places Women's Health Population Groupings business Breast Tissue Duct (anatomy) Neuroscience
Zdroj:	PLoS ONE, Vol 15, Iss 4, p e0231653 (2020) PLoS ONE, 15(4):e0231653. Public Library of Science PLoS ONE
ISSN:	1932-6203
Popis:	Terminal duct lobular unit (TDLU) involution is the regression of milk-producing structures in the breast. Women with less TDLU involution are more likely to develop breast cancer. A major bottleneck in studying TDLU involution in large cohort studies is the need for labor-intensive manual assessment of TDLUs. We developed a computational pathology solution to automatically capture TDLU involution measures. Whole slide images (WSIs) of benign breast biopsies were obtained from the Nurses' Health Study. A set of 92 WSIs was annotated for acini, TDLUs and adipose tissue to train deep convolutional neural network (CNN) models for detection of acini, and segmentation of TDLUs and adipose tissue. These networks were integrated into a single computational method to capture TDLU involution measures including number of TDLUs per tissue area, median TDLU span and median number of acini per TDLU. We validated our method on 40 additional WSIs by comparing with manually acquired measures. Our CNN models detected acini with an F1 score of 0.73±0.07, and segmented TDLUs and adipose tissue with Dice scores of 0.84±0.13 and 0.87±0.04, respectively. The inter-observer ICC scores for manual assessments on 40 WSIs of number of TDLUs per tissue area, median TDLU span, and median acini count per TDLU were 0.71, 0.81 and 0.73, respectively. Intra-observer reliability was evaluated on 10/40 WSIs with ICC scores of >0.8. Inter-observer ICC scores between automated results and the mean of the two observers were: 0.80 for number of TDLUs per tissue area, 0.57 for median TDLU span, and 0.80 for median acini count per TDLU. TDLU involution measures evaluated by manual and automated assessment were inversely associated with age and menopausal status. We developed a computational pathology method to measure TDLU involution. This technology eliminates the labor-intensiveness and subjectivity of manual TDLU assessment, and can be applied to future breast cancer risk studies.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5349429d6c0991aebd45bd5cb862a2c0 https://doaj.org/article/9f205b7ac43a42f2bdc816d3256b1561 Zobrazit plný text záznamu Plný text ve formátu PDF Plný text ve formátu HTML
Nepřihlášeným uživatelům se plný text nezobrazuje	K zobrazení výsledku je třeba se přihlásit.