Cheng Lu1, Mengyao Ji2, Zhen Ma3, Mrinal Mandal4. 1. College of Computer Science, Shaanxi Normal University, Xi'an, Shaanxi Province, China. 2. Department of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan, Hubei, China. 3. College of Food Engineering and Nutritional Science, Shaanxi Normal University, Xi'an, Shaanxi Province, China. 4. Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Alberta, Canada.
Abstract
AIMS: We developed a computer-aided technique to study nuclear atypia classification in high-power field haematoxylin and eosin stained images. METHODS AND RESULTS: An automated technique for nuclear atypia score (NAS) calculation is proposed. The proposed technique uses sophisticated digital image analysis and machine-learning methods to measure the NAS for haematoxylin and eosin stained images. The proposed technique first segments all nuclei regions. A set of morphology and texture features is extracted from presegmented nuclei regions. The histogram of each feature is then calculated to characterize the statistical information of the nuclei. Finally, a support vector machine classifier is applied to classify a high-power field image into different nuclear atypia classes. A set of 1188 digital images was analysed in the experiment. We successfully differentiated the high-power field image with NAS1 versus non-NAS1, NAS2 versus non-NAS2 and NAS3 versus non-NAS3, with area under receiver-operating characteristic curve of 0.90, 0.86 and 0.87, respectively. In three classes evaluation, the average classification accuracy was 78.79%. We found that texture-based feature provides best performance for the classification. CONCLUSION: The automated technique is able to quantify statistical features that may be difficult to be measured by human and demonstrates the future potentials of automated image analysis technique in histopathology analysis.
AIMS: We developed a computer-aided technique to study nuclear atypia classification in high-power field haematoxylin and eosin stained images. METHODS AND RESULTS: An automated technique for nuclear atypia score (NAS) calculation is proposed. The proposed technique uses sophisticated digital image analysis and machine-learning methods to measure the NAS for haematoxylin and eosin stained images. The proposed technique first segments all nuclei regions. A set of morphology and texture features is extracted from presegmented nuclei regions. The histogram of each feature is then calculated to characterize the statistical information of the nuclei. Finally, a support vector machine classifier is applied to classify a high-power field image into different nuclear atypia classes. A set of 1188 digital images was analysed in the experiment. We successfully differentiated the high-power field image with NAS1 versus non-NAS1, NAS2 versus non-NAS2 and NAS3 versus non-NAS3, with area under receiver-operating characteristic curve of 0.90, 0.86 and 0.87, respectively. In three classes evaluation, the average classification accuracy was 78.79%. We found that texture-based feature provides best performance for the classification. CONCLUSION: The automated technique is able to quantify statistical features that may be difficult to be measured by human and demonstrates the future potentials of automated image analysis technique in histopathology analysis.
Authors: Suzanne C Wetstein; Vincent M T de Jong; Nikolas Stathonikos; Mark Opdam; Gwen M H E Dackus; Josien P W Pluim; Paul J van Diest; Mitko Veta Journal: Sci Rep Date: 2022-09-06 Impact factor: 4.996