The Effectiveness of Data Augmentation for Detection of Gastrointestinal Diseases from Endoscopical Images
This work addresses the lack of large public medical databases due to privacy concerns, which hinders deep learning applications in detecting gastrointestinal diseases, but it is incremental as it applies standard augmentation methods to a specific dataset.
The study tackled the problem of limited medical image data for deep learning by applying data augmentation techniques to the Kvasir dataset of gastrointestinal disease images, resulting in sensible improvements in classification precision and recall compared to previous approaches.
The lack, due to privacy concerns, of large public databases of medical pathologies is a well-known and major problem, substantially hindering the application of deep learning techniques in this field. In this article, we investigate the possibility to supply to the deficiency in the number of data by means of data augmentation techniques, working on the recent Kvasir dataset of endoscopical images of gastrointestinal diseases. The dataset comprises 4,000 colored images labeled and verified by medical endoscopists, covering a few common pathologies at different anatomical landmarks: Z-line, pylorus and cecum. We show how the application of data augmentation techniques allows to achieve sensible improvements of the classification with respect to previous approaches, both in terms of precision and recall.