Deep learning-based categorization of mental foramen location using digital panoramic imaging
Background: The mandible's mental foramen (MF) is an anatomical landmark that is clinically significant in a variety of dental, maxillofacial, plastic, and reconstructive surgeries. Locating the mental foramen is critical to avoid complications from different procedures around the mental forame...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wolters Kluwer Medknow Publications
2023-01-01
|
| Series: | Journal of Indian Academy of Oral Medicine and Radiology |
| Subjects: | |
| Online Access: | http://www.jiaomr.in/article.asp?issn=0972-1363;year=2023;volume=35;issue=4;spage=567;epage=571;aulast=Murugan |
| _version_ | 1797232594661670912 |
|---|---|
| author | Arul Jothi Murugan G Anuradha Krithika C Lakshmi K V Swathi |
| author_facet | Arul Jothi Murugan G Anuradha Krithika C Lakshmi K V Swathi |
| author_sort | Arul Jothi Murugan |
| collection | DOAJ |
| description | Background: The mandible's mental foramen (MF) is an anatomical landmark that is clinically significant in a variety of dental, maxillofacial, plastic, and reconstructive surgeries. Locating the mental foramen is critical to avoid complications from different procedures around the mental foramen and nerve. Objectives: To assess the prevalence of the most typical MF site and to construct a deep-learning model for MF location categorization. Materials and Methods: A total of 468 digital panoramic images of the patients who reported for diagnosis of various diseases were gathered retrospectively. According to Telford classification, the position of the mental foramen is highlighted in the photographs, and data were collected to determine its overall prevalence. TensorFlow/Keras software was used to construct a convolutional neural network (CNN) model for the classification of MF. Statistical analysis was done using SPSS software. Results: According to the findings, type 4 MF more frequently occurs at the long axis of the second premolar. The final epoch the model was able to achieve a dissimilarity coefficient of 0.991 on the validation set. Conclusion: Developing the CNN model for categorization of MF is very helpful to decrease the workload over the radiologists and at the same time is highly advantageous for the dentists in treatment planning. |
| first_indexed | 2024-04-24T16:02:46Z |
| format | Article |
| id | doaj.art-72e77042de28484190c5561555e8932e |
| institution | Directory Open Access Journal |
| issn | 0972-1363 |
| language | English |
| last_indexed | 2024-04-24T16:02:46Z |
| publishDate | 2023-01-01 |
| publisher | Wolters Kluwer Medknow Publications |
| record_format | Article |
| series | Journal of Indian Academy of Oral Medicine and Radiology |
| spelling | doaj.art-72e77042de28484190c5561555e8932e2024-04-01T06:03:03ZengWolters Kluwer Medknow PublicationsJournal of Indian Academy of Oral Medicine and Radiology0972-13632023-01-0135456757110.4103/jiaomr.jiaomr_74_23Deep learning-based categorization of mental foramen location using digital panoramic imagingArul Jothi MuruganG AnuradhaKrithika C LakshmiK V SwathiBackground: The mandible's mental foramen (MF) is an anatomical landmark that is clinically significant in a variety of dental, maxillofacial, plastic, and reconstructive surgeries. Locating the mental foramen is critical to avoid complications from different procedures around the mental foramen and nerve. Objectives: To assess the prevalence of the most typical MF site and to construct a deep-learning model for MF location categorization. Materials and Methods: A total of 468 digital panoramic images of the patients who reported for diagnosis of various diseases were gathered retrospectively. According to Telford classification, the position of the mental foramen is highlighted in the photographs, and data were collected to determine its overall prevalence. TensorFlow/Keras software was used to construct a convolutional neural network (CNN) model for the classification of MF. Statistical analysis was done using SPSS software. Results: According to the findings, type 4 MF more frequently occurs at the long axis of the second premolar. The final epoch the model was able to achieve a dissimilarity coefficient of 0.991 on the validation set. Conclusion: Developing the CNN model for categorization of MF is very helpful to decrease the workload over the radiologists and at the same time is highly advantageous for the dentists in treatment planning.http://www.jiaomr.in/article.asp?issn=0972-1363;year=2023;volume=35;issue=4;spage=567;epage=571;aulast=Murugancomplicationsdeep learningmental foramen |
| spellingShingle | Arul Jothi Murugan G Anuradha Krithika C Lakshmi K V Swathi Deep learning-based categorization of mental foramen location using digital panoramic imaging Journal of Indian Academy of Oral Medicine and Radiology complications deep learning mental foramen |
| title | Deep learning-based categorization of mental foramen location using digital panoramic imaging |
| title_full | Deep learning-based categorization of mental foramen location using digital panoramic imaging |
| title_fullStr | Deep learning-based categorization of mental foramen location using digital panoramic imaging |
| title_full_unstemmed | Deep learning-based categorization of mental foramen location using digital panoramic imaging |
| title_short | Deep learning-based categorization of mental foramen location using digital panoramic imaging |
| title_sort | deep learning based categorization of mental foramen location using digital panoramic imaging |
| topic | complications deep learning mental foramen |
| url | http://www.jiaomr.in/article.asp?issn=0972-1363;year=2023;volume=35;issue=4;spage=567;epage=571;aulast=Murugan |
| work_keys_str_mv | AT aruljothimurugan deeplearningbasedcategorizationofmentalforamenlocationusingdigitalpanoramicimaging AT ganuradha deeplearningbasedcategorizationofmentalforamenlocationusingdigitalpanoramicimaging AT krithikaclakshmi deeplearningbasedcategorizationofmentalforamenlocationusingdigitalpanoramicimaging AT kvswathi deeplearningbasedcategorizationofmentalforamenlocationusingdigitalpanoramicimaging |