Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Oct 12;20(20):5780.
doi: 10.3390/s20205780.

Healthcare Professional in the Loop (HPIL): Classification of Standard and Oral Cancer-Causing Anomalous Regions of Oral Cavity Using Textural Analysis Technique in Autofluorescence Imaging

Muhammad Awais  1 , Hemant Ghayvat  2 , Anitha Krishnan Pandarathodiyil  3 , Wan Maria Nabillah Ghani  4 , Anand Ramanathan  4   5 , Sharnil Pandya  6 , Nicolas Walter  7 , Mohamad Naufal Saad  7 , Rosnah Binti Zain  4   8 , Ibrahima Faye  9
Affiliations

Healthcare Professional in the Loop (HPIL): Classification of Standard and Oral Cancer-Causing Anomalous Regions of Oral Cavity Using Textural Analysis Technique in Autofluorescence Imaging

Muhammad Awais et al. Sensors (Basel). .

Abstract

Oral mucosal lesions (OML) and oral potentially malignant disorders (OPMDs) have been identified as having the potential to transform into oral squamous cell carcinoma (OSCC). This research focuses on the human-in-the-loop-system named Healthcare Professionals in the Loop (HPIL) to support diagnosis through an advanced machine learning procedure. HPIL is a novel system approach based on the textural pattern of OML and OPMDs (anomalous regions) to differentiate them from standard regions of the oral cavity by using autofluorescence imaging. An innovative method based on pre-processing, e.g., the Deriche-Canny edge detector and circular Hough transform (CHT); a post-processing textural analysis approach using the gray-level co-occurrence matrix (GLCM); and a feature selection algorithm (linear discriminant analysis (LDA)), followed by k-nearest neighbor (KNN) to classify OPMDs and the standard region, is proposed in this paper. The accuracy, sensitivity, and specificity in differentiating between standard and anomalous regions of the oral cavity are 83%, 85%, and 84%, respectively. The performance evaluation was plotted through the receiver operating characteristics of periodontist diagnosis with the HPIL system and without the system. This method of classifying OML and OPMD areas may help the dental specialist to identify anomalous regions for performing their biopsies more efficiently to predict the histological diagnosis of epithelial dysplasia.

Keywords: VELscope®; autofluorescence imaging; oral cavity mucosal lesions; oral mucosal cancer; oral potentially malignant disorders (OPMD); texture analysis.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Autofluorescence devices: (a) 3rd generation VELscope®; (b) 2nd generation VELscope®.
Figure 2
Figure 2
Illustration of the quadtree distribution diagram.
Figure 3
Figure 3
The Healthcare Professional in the loop (HPIL) workflow.
Figure 4
Figure 4
Flow chart of the textural analysis algorithm.
Figure 5
Figure 5
VELscope® images with Canon A620 settings; the field of view (FOV) is quite straightforward and focused, as in RGB frames.
Figure 6
Figure 6
VELscope® images without the Canon A620 settings; the field of view (FOV) is quite blurred and not as focused as in RGB frames.
Figure 7
Figure 7
VELscope® images of oral cavity annotated by the clinicians.
Figure 8
Figure 8
RGB image of lateral tongue.
Figure 9
Figure 9
VELscope image of oral cavity.
Figure 10
Figure 10
Flow chart for removing VELscope® device area.
Figure 11
Figure 11
Grayscale VELscope® image of an oral cavity.
Figure 12
Figure 12
Region edge detected using Deriche–Canny edge detector.
Figure 13
Figure 13
Region of interest detected using CHT.
Figure 14
Figure 14
Oral cavity region extracted without the VELscope® device.
Figure 15
Figure 15
VELscope® image without device area.
Figure 16
Figure 16
Grayscale VELscope® image of the oral mucosal cavity.
Figure 17
Figure 17
Region edge detected using Deriche–Canny edge detector.
Figure 18
Figure 18
VELscope® image with no circular area detected.
Figure 19
Figure 19
Gray-level co-occurrence matrix (GLCM) texture features of all the ten parameters using 64 × 64 quadtree size.
Figure 20
Figure 20
Graphical analysis of each feature selected via LDA using GLCM features.
Figure 20
Figure 20
Graphical analysis of each feature selected via LDA using GLCM features.
Figure 21
Figure 21
Graphical analysis of parameters vs. statistical results.
Figure 22
Figure 22
Comparative ROC between periodontist, HPIL system, and HPIL system with the periodontist.
Figure 23
Figure 23
GUI for OPMDs.
See this image and copyright information in PMC

References

    1. American Cancer Society Global Cancer Facts & Figures, Atlanta American Cancer Society. [(accessed on 11 October 2020)];2007 Available online: https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts....
    1. Williams P.M., Poh C.F.-Y., Hovan A.J., Ng S., Rosin M.P. Evaluation of a suspicious oral mucosal lesion. J. Can. Dent. Assoc. 2008;74:275–280. - PubMed
    1. Carreras-Torras C., Gay-Escoda C. Techniques for early diagnosis of oral squamous cell carcinoma: Systematic review. Med. Oral Patol. Oral Cir. Bucal. 2015;20:e305–e315. doi: 10.4317/medoral.20347. - DOI - PMC - PubMed
    1. El-Aziz A.A., Aboushousha A., Ali S., Zahran F. Prevalence of Potentially Malignant Lesions and Oral Cancer Among Smokers in an Egyptian cohort: A Hospital-based Cross-Sectional Study. Adv. Dent. J. 2020;2:14–21. doi: 10.21608/adjc.2020.25108.1062. - DOI
    1. Nagao T., Warnakulasuriya S. Screening for oral cancer: Future prospects, research and policy development for Asia. Oral Oncol. 2020;105:104632. doi: 10.1016/j.oraloncology.2020.104632. - DOI - PubMed