Mayara Simões Bispo1, Mário Lúcio Gomes de Queiroz Pierre Júnior2, Antônio Lopes Apolinário3, Jean Nunes Dos Santos4, Braulio Carneiro Junior5, Frederico Sampaio Neves6, Iêda Crusoé-Rebello6. 1. Postgraduate Program in Dentistry and Health, Federal University of Bahia, Salvador, Brazil. 2. Computer Science Department, Federal Institute of Education, Science and Technology of Bahia, Senhor do Bonfim, Bahia, Brazil. 3. Computer Science Department, Federal University of Bahia, Salvador, Brazil. 4. Division of Oral Pathology, Federal University of Bahia, Salvador, Brazil. 5. Division of Oral and Maxillofacial Surgery, Southwest Bahia State University, Vitória da Conquista, Brazil. 6. Division of Oral and Maxillofacial Radiology, Federal University of Bahia, Salvador, Brazil.
Abstract
OBJECTIVE: To analyse the automatic classification performance of a convolutional neural network (CNN), Google Inception v3, using tomographic images of odontogenic keratocysts (OKCs) and ameloblastomas (AMs). METHODS: For construction of the database, we selected axial multidetector CT images from patients with confirmed AM (n = 22) and OKC (n = 18) based on a conclusive histopathological report. The images (n = 350) were segmented manually and data augmentation algorithms were applied, totalling 2500 images. The k-fold × five cross-validation method (k = 2) was used to estimate the accuracy of the CNN model. RESULTS: The accuracy and standard deviation (%) of cross-validation for the five iterations performed were 90.16 ± 0.95, 91.37 ± 0.57, 91.62 ± 0.19, 92.48 ± 0.16 and 91.21 ± 0.87, respectively. A higher error rate was observed for the classification of AM images. CONCLUSION: This study demonstrated a high classification accuracy of Google Inception v3 for tomographic images of OKCs and AMs. However, AMs images presented the higher error rate.
OBJECTIVE: To analyse the automatic classification performance of a convolutional neural network (CNN), Google Inception v3, using tomographic images of odontogenic keratocysts (OKCs) and ameloblastomas (AMs). METHODS: For construction of the database, we selected axial multidetector CT images from patients with confirmed AM (n = 22) and OKC (n = 18) based on a conclusive histopathological report. The images (n = 350) were segmented manually and data augmentation algorithms were applied, totalling 2500 images. The k-fold × five cross-validation method (k = 2) was used to estimate the accuracy of the CNN model. RESULTS: The accuracy and standard deviation (%) of cross-validation for the five iterations performed were 90.16 ± 0.95, 91.37 ± 0.57, 91.62 ± 0.19, 92.48 ± 0.16 and 91.21 ± 0.87, respectively. A higher error rate was observed for the classification of AM images. CONCLUSION: This study demonstrated a high classification accuracy of Google Inception v3 for tomographic images of OKCs and AMs. However, AMs images presented the higher error rate.
Authors: Marc Aubreville; Christian Knipfer; Nicolai Oetter; Christian Jaremenko; Erik Rodner; Joachim Denzler; Christopher Bohr; Helmut Neumann; Florian Stelzle; Andreas Maier Journal: Sci Rep Date: 2017-09-20 Impact factor: 4.379
Authors: Titus Josef Brinker; Achim Hekler; Jochen Sven Utikal; Niels Grabe; Dirk Schadendorf; Joachim Klode; Carola Berking; Theresa Steeb; Alexander H Enk; Christof von Kalle Journal: J Med Internet Res Date: 2018-10-17 Impact factor: 5.428