Prediction of Radiation-Related Dental Caries Through PyRadiomics Features and Artificial Neural Network on Panoramic Radiography

Autor(es)

Faria, Vanessa De Araujo ^[1] ; Azimbagirad, Mehran ^[2] ; Arruda, Gustavo Viani ^[1] ; Pavoni, Juliana Fernandes ^[3] ; Felipe, Joaquim Cezar ^[4] ; Mendes Ferreira dos Santos, Elza Maria Carneiro ^[5] ; Murta Junior, Luiz Otavio ^[4]

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Fac Med, Dept Radiol, Ribeirao Preto, SP - Brazil

^[2] Univ Western Brittany, Dept Med Sci & Hlth, IBRBS LATIM Lab, 12 Av Foch, F-29200 Brest - France

^[3] Univ Sao Paulo, Fac Philosophy Sci & Languages, Dept Phys, Ribeirao Preto, SP - Brazil

^[4] Univ Sao Paulo, Fac Philosophy Sci & Languages, Dept Comp & Math, Ribeirao Preto, SP - Brazil

^[5] Med & Diagnost Lab, 3S Diagnost Imagem, Leme, SP - Brazil

• Tipo de documento: Artigo Científico
• Fonte: JOURNAL OF DIGITAL IMAGING; v. 34, n. 5 JUL 2021.
• Citações Web of Science: 0

Resumo

The prediction and detection of radiation-related caries (RRC) are crucial to manage the side effects of the head and the neck cancer (HNC) radiotherapy (RT). Despite the demands for the prediction of RRC, no study proposes and evaluates a prediction method. This study introduces a method based on artificial intelligence neural network to predict and detect either regular caries or RRC in HNC patients under RT using features extracted from panoramic radiograph. We selected fifteen HNC patients (13 men and 2 women) to analyze, retrospectively, their panoramic dental images, including 420 teeth. Two dentists manually labeled the teeth to separate healthy and teeth with either type caries. They also labeled the teeth by resistant and vulnerable, as predictive labels telling about RT aftermath caries. We extracted 105 statistical/morphological image features of the teeth using PyRadiomics. Then, we used an artificial neural network classifier (ANN), firstly, to select the best features (using maximum weights) and then label the teeth: in caries and non-caries while detecting RRC, and resistant and vulnerable while predicting RRC. To evaluate the method, we calculated the confusion matrix, receiver operating characteristic (ROC), and area under curve (AUC), as well as a comparison with recent methods. The proposed method showed a sensibility to detect RRC of 98.8% (AUC = 0.9869) and to predict RRC achieved 99.2% (AUC = 0.9886). The proposed method to predict and detect RRC using neural network and PyRadiomics features showed a reliable accuracy able to perform before starting RT to decrease the side effects on susceptible teeth. (AU)

• Processo FAPESP: 17/20598-9 - Métodos computacionais para o planejamento, avaliação e acompanhamento de cirurgias e terapias através de imagens médicas.
• Beneficiário: Luiz Otavio Murta Junior
• Modalidade de apoio: Auxílio à Pesquisa - Regular