Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading
Introduction: Oral epithelial dysplasia (OED) is associated with high interobserver and intraobserver disagreement. With the exponential increase in the applicability of artificial intelligence tools such as deep learning (DL) in pathology, it would now be possible to achieve high accuracy and objec...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wolters Kluwer Medknow Publications
2023-01-01
|
| Series: | Advances in Human Biology |
| Subjects: | |
| Online Access: | http://www.aihbonline.com/article.asp?issn=2321-8568;year=2023;volume=13;issue=1;spage=57;epage=60;aulast=Nandini |
| _version_ | 1797904992437272576 |
|---|---|
| author | C Nandini Shaik Basha Aarchi Agarawal R Parikh Neelampari Krishna P Miyapuram R Jadeja Nileshwariba |
| author_facet | C Nandini Shaik Basha Aarchi Agarawal R Parikh Neelampari Krishna P Miyapuram R Jadeja Nileshwariba |
| author_sort | C Nandini |
| collection | DOAJ |
| description | Introduction: Oral epithelial dysplasia (OED) is associated with high interobserver and intraobserver disagreement. With the exponential increase in the applicability of artificial intelligence tools such as deep learning (DL) in pathology, it would now be possible to achieve high accuracy and objectivity in grading of OED. In this research work, we have proposed a DL approach to epithelial dysplasia grading by creating a convolutional neural network (CNN) model from scratch. Materials and Methods: The dataset includes 445 high-resolution ×400 photomicrographs captured from histopathologically diagnosed cases of high-risk dysplasia (HR) and normal buccal mucosa (NBM) that were used to train, validate and test the two-dimensional CNN (2DCNN) model. Results: The whole dataset was divided into 60% training set, 20% validation set and 20% test set. The model achieved training accuracy of 97.21%, validation accuracy of 90% and test accuracy of 91.30%. Conclusion: The DL model was able to distinguish between normal epithelium and HR epithelial dysplasia with high grades of accuracy. These results are encouraging for researchers to formulate DL models to grade and classify OED using various grading systems. |
| first_indexed | 2024-04-10T09:57:49Z |
| format | Article |
| id | doaj.art-944c3220b0c9472d8e89babd25138325 |
| institution | Directory Open Access Journal |
| issn | 2321-8568 2348-4691 |
| language | English |
| last_indexed | 2024-04-10T09:57:49Z |
| publishDate | 2023-01-01 |
| publisher | Wolters Kluwer Medknow Publications |
| record_format | Article |
| series | Advances in Human Biology |
| spelling | doaj.art-944c3220b0c9472d8e89babd251383252023-02-16T11:59:51ZengWolters Kluwer Medknow PublicationsAdvances in Human Biology2321-85682348-46912023-01-01131576010.4103/aihb.aihb_30_22Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia gradingC NandiniShaik BashaAarchi AgarawalR Parikh NeelampariKrishna P MiyapuramR Jadeja NileshwaribaIntroduction: Oral epithelial dysplasia (OED) is associated with high interobserver and intraobserver disagreement. With the exponential increase in the applicability of artificial intelligence tools such as deep learning (DL) in pathology, it would now be possible to achieve high accuracy and objectivity in grading of OED. In this research work, we have proposed a DL approach to epithelial dysplasia grading by creating a convolutional neural network (CNN) model from scratch. Materials and Methods: The dataset includes 445 high-resolution ×400 photomicrographs captured from histopathologically diagnosed cases of high-risk dysplasia (HR) and normal buccal mucosa (NBM) that were used to train, validate and test the two-dimensional CNN (2DCNN) model. Results: The whole dataset was divided into 60% training set, 20% validation set and 20% test set. The model achieved training accuracy of 97.21%, validation accuracy of 90% and test accuracy of 91.30%. Conclusion: The DL model was able to distinguish between normal epithelium and HR epithelial dysplasia with high grades of accuracy. These results are encouraging for researchers to formulate DL models to grade and classify OED using various grading systems.http://www.aihbonline.com/article.asp?issn=2321-8568;year=2023;volume=13;issue=1;spage=57;epage=60;aulast=Nandiniconvolutional neural networkdeep learningepithelial dysplasiaoral cancer |
| spellingShingle | C Nandini Shaik Basha Aarchi Agarawal R Parikh Neelampari Krishna P Miyapuram R Jadeja Nileshwariba Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading Advances in Human Biology convolutional neural network deep learning epithelial dysplasia oral cancer |
| title | Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading |
| title_full | Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading |
| title_fullStr | Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading |
| title_full_unstemmed | Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading |
| title_short | Deep learning approach to detect high-risk oral epithelial dysplasia: A step towards computer-assisted dysplasia grading |
| title_sort | deep learning approach to detect high risk oral epithelial dysplasia a step towards computer assisted dysplasia grading |
| topic | convolutional neural network deep learning epithelial dysplasia oral cancer |
| url | http://www.aihbonline.com/article.asp?issn=2321-8568;year=2023;volume=13;issue=1;spage=57;epage=60;aulast=Nandini |
| work_keys_str_mv | AT cnandini deeplearningapproachtodetecthighriskoralepithelialdysplasiaasteptowardscomputerassisteddysplasiagrading AT shaikbasha deeplearningapproachtodetecthighriskoralepithelialdysplasiaasteptowardscomputerassisteddysplasiagrading AT aarchiagarawal deeplearningapproachtodetecthighriskoralepithelialdysplasiaasteptowardscomputerassisteddysplasiagrading AT rparikhneelampari deeplearningapproachtodetecthighriskoralepithelialdysplasiaasteptowardscomputerassisteddysplasiagrading AT krishnapmiyapuram deeplearningapproachtodetecthighriskoralepithelialdysplasiaasteptowardscomputerassisteddysplasiagrading AT rjadejanileshwariba deeplearningapproachtodetecthighriskoralepithelialdysplasiaasteptowardscomputerassisteddysplasiagrading |