Objects Characterization-based Approach to Enhance Detection of Degree of Malignancy in Breast Cancer Histopathology

摘要

Histologic grading from images has become widely accepted as a powerful indicator of prognosis in breast cancer.Automated grading can assist the doctor diagnosing the medical condition. But algorithms still lag behind human expertsin this task, as human experts excel in identifying parts, detecting characteristics and relating concepts and semantics.This can be improved by making algorithms distinguish and characterize the most relevant types of objects in the imageand characterizing images from that. We propose a three-stage automated approach named OBI (Object-basedIdentification) with steps: 1. Object-based identification, which identifies the “type of object” of each region andcharacterizes it; 2. Learn about image, which characterizes distribution of characteristics of those types of objects inimage; 3. Determination of degree of malignancy, which assigns a degree of malignancy based on a classifier over objecttype characteristics (the statistical distribution of characteristics of structures) in the image. Our proof-of-conceptprototype uses publicly-available Mytos-Atypia dataset [19] to compare accuracy with alternatives. Results summary:human expert (medical doctor) 84%, classic machine learning 74%, convolution neural networks (CNN), 78%, ourapproach (OBI) 86%. As future work, we expect to generalize our results to other datasets and problems, exploremimicking knowledge of human concepts further, merge the object-based approach with CNN techniques and adapt it toother medical imaging contexts.