Haitham Askar1, Joachim Krois1, Csaba Rohrer1, Sarah Mertens2, Karim Elhennawy3, Livia Ottolenghi4, Marta Mazur4, Sebastian Paris1, Falk Schwendicke5. 1. Department of Oral Diagnostics, Digital Health and Health Services Research, Charité - Universitätsmedizin Berlin, Germany. 2. Department of Oral Diagnostics, Digital Health and Health Services Research, Charité - Universitätsmedizin Berlin, Germany; Department of Operative and Preventive Dentistry, Charité - Universitätsmedizin Berlin, Germany. 3. Department of Orthodontics, Dentofacial Orthopedics and Pedodontics, Charité - Universitätsmedizin Berlin, Germany. 4. Department of Oral and MaxilloFacial Sciences, Sapienza University of Rome, Italy. 5. Department of Oral Diagnostics, Digital Health and Health Services Research, Charité - Universitätsmedizin Berlin, Germany. Electronic address: falk.schwendicke@charite.de.
Abstract
OBJECTIVES: We aimed to apply deep learning to detect white spot lesions in dental photographs. METHODS: Using 434 photographic images of 51 patients, a dataset of 2781 cropped tooth segments was generated. Pixelwise annotations of sound enamel as well as fluorotic, carious or other types of hypomineralized lesions were generated by experts and assessed by an independent second reviewer. The union of the reviewed annotations were used to segment the hard tissues (region-of-interest, ROI) of each image. SqueezeNet was employed for modelling. We trained models to detect (1) any white spot lesions, (2) fluorotic lesions and (3) other-than-fluorotic lesions. Modeling was performed on both the cropped and the ROI images and using ten-times repeated five-fold cross-validation. Feature visualization was applied to visualize salient areas. RESULTS: Lesion prevalence was 37 %; the majority of lesions (24 %) were fluorotic. None of the metrics differed significantly between the models trained on cropped and ROI imagery (p > 0.05/t-test). Mean accuracies ranged between 0.81-0.84, without significant differences between models trained to detect any, fluorotic or other-than-fluorotic lesions (p > 0.05). Specificities were 0.85-0.86; sensitivities were lower (0.58-0.66). Models to detect any lesions showed positive/negative predictive values (PPV/NPV) between 0.77-0.80, those to detect fluorotic lesions 0.67 (PPV) to 0.86 (NPV), and those to detect other-than-fluorotic lesions 0.46 (PPV) to 0.93 (NPV). Light reflections were the main reason for false positive detections. CONCLUSIONS: Deep learning showed satisfying accuracy to detect white spot lesions, particularly fluorosis. Some models showed limited stability given the small sample available. CLINICAL SIGNIFICANCE: Deep learning is suitable for automated classification of retro- or prospectively collected imagery and may assist practitioners in discriminating white spot lesions. Future studies should expand the scope into more granular multi-class detections on a larger and more generalizable dataset.
OBJECTIVES: We aimed to apply deep learning to detect white spot lesions in dental photographs. METHODS: Using 434 photographic images of 51 patients, a dataset of 2781 cropped tooth segments was generated. Pixelwise annotations of sound enamel as well as fluorotic, carious or other types of hypomineralized lesions were generated by experts and assessed by an independent second reviewer. The union of the reviewed annotations were used to segment the hard tissues (region-of-interest, ROI) of each image. SqueezeNet was employed for modelling. We trained models to detect (1) any white spot lesions, (2) fluorotic lesions and (3) other-than-fluorotic lesions. Modeling was performed on both the cropped and the ROI images and using ten-times repeated five-fold cross-validation. Feature visualization was applied to visualize salient areas. RESULTS: Lesion prevalence was 37 %; the majority of lesions (24 %) were fluorotic. None of the metrics differed significantly between the models trained on cropped and ROI imagery (p > 0.05/t-test). Mean accuracies ranged between 0.81-0.84, without significant differences between models trained to detect any, fluorotic or other-than-fluorotic lesions (p > 0.05). Specificities were 0.85-0.86; sensitivities were lower (0.58-0.66). Models to detect any lesions showed positive/negative predictive values (PPV/NPV) between 0.77-0.80, those to detect fluorotic lesions 0.67 (PPV) to 0.86 (NPV), and those to detect other-than-fluorotic lesions 0.46 (PPV) to 0.93 (NPV). Light reflections were the main reason for false positive detections. CONCLUSIONS: Deep learning showed satisfying accuracy to detect white spot lesions, particularly fluorosis. Some models showed limited stability given the small sample available. CLINICAL SIGNIFICANCE: Deep learning is suitable for automated classification of retro- or prospectively collected imagery and may assist practitioners in discriminating white spot lesions. Future studies should expand the scope into more granular multi-class detections on a larger and more generalizable dataset.
Authors: Baichen Ding; Zhuo Zhang; Yiran Liang; Weiwei Wang; Siwei Hao; Ze Meng; Lian Guan; Ying Hu; Bin Guo; Runlian Zhao; Yan Lv Journal: Ann Transl Med Date: 2021-11
Authors: Marta Mazur; Artnora Ndokaj; Divyambika Catakapatri Venugopal; Michela Roberto; Cristina Albu; Maciej Jedliński; Silverio Tomao; Iole Vozza; Grzegorz Trybek; Livia Ottolenghi; Fabrizio Guerra Journal: Int J Environ Res Public Health Date: 2021-11-10 Impact factor: 3.390
Authors: Antonio Ciardo; Sarah K Sonnenschein; Marlinde M Simon; Maurice Ruetters; Marcia Spindler; Philipp Ziegler; Ingvi Reccius; Alexander-Nicolaus Spies; Jana Kykal; Eva-Marie Baumann; Susanne Fackler; Christopher Büsch; Ti-Sun Kim Journal: PLoS One Date: 2022-05-17 Impact factor: 3.240