Automated abnormality detection of craniomaxillofacial based on Statistical Deformable Models

In this paper, we propose an algorithm of automated detection for malformed craniomaxillofacial regions based on Statistical Deformable Model. Firstly, craniomaxillofacial is segmented into different regions based on salient feature point identification and K-means clustering. Then, each region is treated as a missing part. Instead, the recovery region is calculated from a pre-trained statistical deformable model. Afterward, the abnormality of the given region is defined by the difference of the original region and the recovered region. The experimental results conducted in 300 samples demonstrate that the proposed detection algorithm can achieve precise detection and quantification of the malformed craniomaxillofacial region.