Application of Two-Dimensional Entropy Measures to Detect the Radiographic Signs of Tooth Resorption and Hypercementosis in an Equine Model

Abstract

Dental disorders are a serious health problem in equine medicine, their early recognition benefits the long-term general health of the horse. Most of the initial signs of Equine Odontoclastic Tooth Resorption and Hypercementosis (EOTRH) syndrome concern the alveolar aspect of the teeth, thus, the need for early recognition radiographic imaging. This study is aimed to evaluate the applicability of entropy measures to quantify the radiological signs of tooth resorption and hypercementosis as well as to enhance radiographic image quality in order to facilitate the identification of the signs of EOTRH syndrome. A detailed examination of the oral cavity was performed in eighty horses. Each evaluated incisor tooth was assigned to one of four grade–related EOTRH groups (0–3). Radiographs of the incisor teeth were taken and digitally processed. For each radiograph, two–dimensional sample (SampEn2D), fuzzy (FuzzEn2D), permutation (PermEn2D), dispersion (DispEn2D), and distribution (DistEn2D) entropies were measured after image filtering was performed using Normalize, Median, and LaplacianSharpening filters. Moreover, the similarities between entropy measures and selected Gray–Level Co–occurrence Matrix (GLCM) texture features were investigated. Among the 15 returned measures, DistEn2D was EOTRH grade–related. Moreover, DistEn2D extracted after Normalize filtering was the most informative. The EOTRH grade–related similarity between DistEn2D and Difference Entropy (GLCM) confirms the higher irregularity and complexity of incisor teeth radiographs in advanced EOTRH syndrome, demonstrating the greatest sensitivity (0.50) and specificity (0.95) of EOTRH 3 group detection. An application of DistEn2D to Normalize filtered incisor teeth radiographs enables the identification of the radiological signs of advanced EOTRH with higher accuracy than the previously used entropy–related GLCM texture features.

An application of the density standard and scaled–pixel–counting protocol to assess the radiodensity of equine incisor teeth affected by resorption and hypercementosis: preliminary advancement in dental radiography

“…In the current study, most of the teeth in EOTRH group 3 showed a smooth outline despite the obvious internal changes, whereas, on recent raw [ 32 , 37 , 56 , 60 ] and digitally processed [ 38 , 52 ] radiographs, the signs of obviously irregular or rough incisor tooth surfaces were predominately. In recently published research, the accuracy of EOTRH group 0 and 3 differentiation ranged from 0.25 Sp and 0.99 Se for matrix–based texture features to 0.50 Sp and 1.00 Se for two–dimensional entropy measures [ 52 ]. In the current study, the accuracy of EOTRH 0 and 3 differentiation ranged from 0.25 Sp and 1.00 Se for S1 to 0.38 Sp and 0.82 Se for S6.…”

“…The techniques to enhance early detection of radiographic signs of dental diseases in horses is an important direction of recent research in equine dentistry [ 2 , 38 , 52 – 54 ]. Most recently, filtering algorithms and texture analysis of incisor teeth radiographs, based on the first– and second–order statistics [ 38 ] and two–dimensional entropy measures [ 52 ], have been successfully used in EOTRH 0 and 3 differentiation. Visually, it is less difficult to identify radiographic signs of EOTRH 3, representing severe tooth hypercementosis and resorption, and EOTRH 0, representing healthy teeth [ 32 ].…”

“…Therefore, one may conclude the use of digital processing techniques in these cases will enhance the quality of the maxillary incisor radiographs so that the radiologist can more easily identify the radiographic signs. However, no clear differentiation of radiographic signs of EOTRH 1, representing mild tooth hypercementosis and resorption [ 32 ], and EOTRH 0, were evidenced [ 38 , 52 ]. In the case of visual inspection, early signs of EOTRH might be missed by merely personal evaluation of radiographs.…”

“…These observations justify the use of density standard for indirect quantification of the brightness of the radiograph, and thus the use of the proposed scaled–pixel–counting protocol for indirect quantification of the radiographic signs of tooth resorption and hypercementosis on the EOTRH syndrome model. Recent research has discussed the use of histogram–based and matrix–based texture features [ 38 ] and two–dimensional entropy measures [ 52 ] in the digital processing of EOTRH radiographs. These studies were focused only on maxillary, not maxillary and mandibular, incisor teeth.…”

“…The current study was based on the raw radiographs collected directly from the X–ray scanner, which were not digitally processed, therefore the use of radiograph filtering may enhance the quality of the radiographs and thus, increase the Sp and Se of detecting early radiographic signs of EOTRH. Filtering, which enhances brightness, contrast, and/or edges of radiographs, is routinely used in human dentistry [ 61 , 62 ] and has been introduced into equine dentistry [ 38 , 52 ], and this direction of further research development seems to be promising.…”