Image anomaly detection in preventive fluorographic
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
В данной работе рассматривается возможность применения генеративного подхода машинного обучения в области поиска аномалий на рентгенограммах. Применение данного подхода оценивалось по двум метрикам: AUC-ROC и AUC-PR. Обе метрики показали хорошие результаты: площадь под ROC-кривой составила 0.92, под PR-кривой 0.94.
In this paper, we consider the possibility of applying a generative machine learning approach in the search for anomalies on radiographs. Applying this approach was evaluated using two metrics: AUC-ROC and AUC-PR. Both metrics showed good results: the area under the ROC curve was 0.92, under the PR curve of 0.94.
In this paper, we consider the possibility of applying a generative machine learning approach in the search for anomalies on radiographs. Applying this approach was evaluated using two metrics: AUC-ROC and AUC-PR. Both metrics showed good results: the area under the ROC curve was 0.92, under the PR curve of 0.94.