Dento maxillo facial radiology · 2025
Automatic identification of dental implant brands with deep learning algorithms.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Yuce Hasret, Acıkgoz Gozde, Al-Jumaili Saif, Ucan Osman Nuri, Ozlu Ucan Gulfem, Dere Kadriye Ayca
Аннотация
OBJECTIVES: To reduce the problems arising from the inability to identify dental implant brands, this study aims to classify various dental implant brands using deep learning algorithms on panoramic radiographs.
METHODS: Images of 4 different dental implant systems (NucleOSS, Medentika, Nobel, and Implance) were used from a total of 5375 cropped panoramic radiographs. To enhance image clarity and reduce blurriness, the contrast limited adaptive histogram equalization filter was applied. GoogleNet, ResNet-18, VGG16, and ShuffleNet deep learning algorithms were utilized to classify the 4 different dental implant systems. To evaluate the classification performance of the algorithms, Receiver Operating Characteristic (ROC) curves and confusion matrices were generated. Based on these confusion matrices, accuracy, precision, sensitivity, and F1 score were calculated. The Z-test was used to compare the performance metrics across different algorithms.
RESULTS: The accuracy rates of the deep learning algorithms were obtained as 96.00% for GoogleNet, 84.40% for ResNet-18, 98.90% for VGG16, and 84.80% for ShuffleNet. A statistically significant difference was found between the accuracy rate of the VGG16 algorithm and those of GoogleNet, ShuffleNet, and ResNet-18 (P < .001, P < .001, and P < .001, respectively).
CONCLUSIONS: With the achievement of high accuracy rates, deep learning algorithms are considered a valuable and powerful method for identifying dental implant brands.
METHODS: Images of 4 different dental implant systems (NucleOSS, Medentika, Nobel, and Implance) were used from a total of 5375 cropped panoramic radiographs. To enhance image clarity and reduce blurriness, the contrast limited adaptive histogram equalization filter was applied. GoogleNet, ResNet-18, VGG16, and ShuffleNet deep learning algorithms were utilized to classify the 4 different dental implant systems. To evaluate the classification performance of the algorithms, Receiver Operating Characteristic (ROC) curves and confusion matrices were generated. Based on these confusion matrices, accuracy, precision, sensitivity, and F1 score were calculated. The Z-test was used to compare the performance metrics across different algorithms.
RESULTS: The accuracy rates of the deep learning algorithms were obtained as 96.00% for GoogleNet, 84.40% for ResNet-18, 98.90% for VGG16, and 84.80% for ShuffleNet. A statistically significant difference was found between the accuracy rate of the VGG16 algorithm and those of GoogleNet, ShuffleNet, and ResNet-18 (P < .001, P < .001, and P < .001, respectively).
CONCLUSIONS: With the achievement of high accuracy rates, deep learning algorithms are considered a valuable and powerful method for identifying dental implant brands.