Non-invasive differential diagnosis of dental periapical lesions in cone-beam CT

This paper proposes a novel application of computer-aided diagnosis to a clinically significant dental problem: non-invasive differential diagnosis of periapical lesions using cone-beam computed tomography (CBCT). The proposed semi-automatic solution combines graph-theoretic random walks segmentation and machine learning-based LDA and AdaBoost classifiers. Our quantitative experiments show the effectiveness of the proposed method by demonstrating 94.1% correct classification rate. Furthermore, we compare classification performances with two independent ground-truth sets from the biopsy and CBCT diagnoses. ROC analysis reveals our method improves accuracy for both cases and behaves more in agreement with the CBCT diagnosis, supporting a hypothesis presented in a recent clinical report.