TY - JOUR
T1 - Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers
AU - Ünsal, Gürkan
AU - Chaurasia, Akhilanand
AU - Akkaya, Nurullah
AU - Chen, Nadler
AU - Abdalla-Aslan, Ragda
AU - Koca, Revan Birke
AU - Orhan, Kaan
AU - Roganovic, Jelena
AU - Reddy, Prashanti
AU - Wahjuningrum, Dian Agustin
N1 - Publisher Copyright:
© IMechE 2023.
PY - 2023/6
Y1 - 2023/6
N2 - This study aimed to develop an algorithm to automatically segment the oral potentially malignant diseases (OPMDs) and oral cancers (OCs) of all oral subsites with various deep convolutional neural network applications. A total of 510 intraoral images of OPMDs and OCs were collected over 3 years (2006—2009). All images were confirmed both with patient records and histopathological reports. Following the labeling of the lesions the dataset was arbitrarily split, using random sampling in Python as the study dataset, validation dataset, and test dataset. Pixels were classified as the OPMDs and OCs with the OPMD/OC label and the rest as the background. U-Net architecture was used and the model with the best validation loss was chosen for the testing among the trained 500 epochs. Dice similarity coefficient (DSC) score was noted. The intra-observer ICC was found to be 0.994 while the inter-observer reliability was 0.989. The calculated DSC and validation accuracy across all clinical images were 0.697 and 0.805, respectively. Our algorithm did not maintain an excellent DSC due to multiple reasons for the detection of both OC and OPMDs in oral cavity sites. A better standardization for both 2D and 3D imaging (such as patient positioning) and a bigger dataset are required to improve the quality of such studies. This is the first study which aimed to segment OPMDs and OCs in all subsites of oral cavity which is crucial not only for the early diagnosis but also for higher survival rates.
AB - This study aimed to develop an algorithm to automatically segment the oral potentially malignant diseases (OPMDs) and oral cancers (OCs) of all oral subsites with various deep convolutional neural network applications. A total of 510 intraoral images of OPMDs and OCs were collected over 3 years (2006—2009). All images were confirmed both with patient records and histopathological reports. Following the labeling of the lesions the dataset was arbitrarily split, using random sampling in Python as the study dataset, validation dataset, and test dataset. Pixels were classified as the OPMDs and OCs with the OPMD/OC label and the rest as the background. U-Net architecture was used and the model with the best validation loss was chosen for the testing among the trained 500 epochs. Dice similarity coefficient (DSC) score was noted. The intra-observer ICC was found to be 0.994 while the inter-observer reliability was 0.989. The calculated DSC and validation accuracy across all clinical images were 0.697 and 0.805, respectively. Our algorithm did not maintain an excellent DSC due to multiple reasons for the detection of both OC and OPMDs in oral cavity sites. A better standardization for both 2D and 3D imaging (such as patient positioning) and a bigger dataset are required to improve the quality of such studies. This is the first study which aimed to segment OPMDs and OCs in all subsites of oral cavity which is crucial not only for the early diagnosis but also for higher survival rates.
KW - Oral cancers
KW - convolutional neural networks
KW - deep learning
KW - machine learning
KW - oral cavity
UR - http://www.scopus.com/inward/record.url?scp=85163005133&partnerID=8YFLogxK
U2 - 10.1177/09544119231176116
DO - 10.1177/09544119231176116
M3 - Article
AN - SCOPUS:85163005133
SN - 0954-4119
VL - 237
SP - 719
EP - 726
JO - Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine
JF - Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine
IS - 6
ER -