Improved Barrett's Cancer Detection in Volumetric Laser Endomicroscopy scans using Multiple-Frame Voting
This paper explores the feasibility of using multi-frame analysis to increase the classification performance of machine learning methods for cancer detection in Volumetric Laser Endomicroscopy (VLE). VLE is a novel and promising modality for the detection of neoplasia in patients with Barett's Esophagus (BE). It produces hundreds of high-resolution, cross-sectional images of the esophagus and offers considerable advantages compared to current methods. While some recent studies have proposed cancer detection algorithms for single VLE frames, the study described in this paper is the first to make use of VLE volumes for the differentiation between dysplastic and non-dysplastic tissue. We explore the use of various voting schemes for a broad range of features and classification methods. Our results demonstrate that multi-frame analysis leads to superior performance, irrespectively of the chosen feature-classifier combination. By using multi-frame analysis with straightforward voting methods, the Area Under the receiver operating Curve (AUC) is increased by an average of over 12% compared to using single VLE frames. When only considering methods that achieve expert performance or higher (AUC>=0.81), an even larger performance improvement of up to 16.9% is observed. Furthermore, many feature-classifier combinations result in AUC values ranging from 0.90 to 0.98, thereby significantly outperforming medical experts using a recently proposed clinical prediction model, which has demonstrated an AUC of 0.81.