Automated Detection of Oral Pre-Cancerous Tongue Lesions Using Deep Learning for Early Diagnosis of Oral Cavity Cancer

Abstract

Discovering oral cavity cancer (OCC) at an early stage is an effective way to increase patient survival rate. However, current initial screening process is done manually and is expensive for the average individual, especially in developing countries worldwide. This problem is further compounded due to the lack of specialists in such areas. Automating the initial screening process using artificial intelligence (AI) to detect pre-cancerous lesions can prove to be an effective and inexpensive technique that would allow patients to be triaged accordingly to receive appropriate clinical management. In this study, we have applied and evaluated the efficacy of six deep convolutional neural network (DCNN) models using transfer learning, for identifying pre-cancerous tongue lesions directly using a small dataset of clinically annotated photographic images to diagnose early signs of OCC. DCNN models were able to differentiate between benign and pre-cancerous tongue lesions and were also able to distinguish between five types of tongue lesions, i.e. hairy tongue, fissured tongue, geographic tongue, strawberry tongue and oral hairy leukoplakia with high classification performances. Preliminary results using an (AI + Physician) ensemble model demonstrate that an automated pre-screening process of oral tongue lesions using DCNNs can achieve ‘near-human’ level classification performance for diagnosing early signs of OCC in patients.

The use of artificial intelligence, machine learning and deep learning in oncologic histopathology

“…Although initial efforts of applying AI‐based approaches have focused on OSCC, the use of these approaches on oral potentially malignant disorders (OPMDs) has significant potential to improve early detection of OSCC. Shamim et al 37 2019 used an annotated dataset of clinical images of benign and potentially malignant tongue lesions to train deep CNNs for the purposes of image classification. In this study, their model achieved a mean classification accuracy of 0.98 in distinguishing between benign and potentially malignant tongue lesions 37 .…”

“…Shamim et al 37 2019 used an annotated dataset of clinical images of benign and potentially malignant tongue lesions to train deep CNNs for the purposes of image classification. In this study, their model achieved a mean classification accuracy of 0.98 in distinguishing between benign and potentially malignant tongue lesions 37 . Studies on WSIs of OPMDs are lacking but importantly, parallels may be drawn with AI‐based approaches on clinicopathologic parameters of premalignant disorders in different sites such as the esophagus (Barrett's esophagus) and the cervix (cervical intraepithelial neoplasia) 38,39 .…”

Machine learning in dental, oral and craniofacial imaging: a review of recent progress

“…In recent years, conventional and modern ML methods, especially neural networks and SVM, have illustrated the capability of processing oral cavity tumor-related image data. This includes oral cancer detection and tissue cell classification in the stage of cancer diagnosis ( Al-Ma’aitah & AlZubi, 2018 ; Aubreville et al, 2017 ; Das, Hussain & Mahanta, 2020 ; Jeyaraj & Samuel Nadar, 2019 ; Shamim et al, 2019 ), tumor margin assessment and tumor subtype classification in the process of clinical cancer treatment ( Fei et al, 2017 ; Marsden et al, 2020 ; van Rooij et al, 2019 ) and assessment of complications after treatment ( Ariji et al, 2019 ; Dong et al, 2018 ; Men et al, 2019 ). Major tumors like OSCC are able to be detected and evaluated with high accuracy using a timesaving algorithm ( Aubreville et al, 2017 ; Das, Hussain & Mahanta, 2020 ).…”

“…Some other studies have focused on the diagnosis of single oral cancers ( Aubreville et al, 2017 ; Das, Hussain & Mahanta, 2020 ; Rahman et al, 2020 ; Shamim et al, 2019 ). Squamous cell carcinoma is responsible for approximately 90% of total oral cancers and has become the sixth most common cancer worldwide ( D’Souza & Addepalli, 2018 ; Kar et al, 2020 ).…”

“…Trials on automated screening of other oral cavity cancers have been implemented clinically ( Shamim et al, 2019 ) or on experimental animals ( Lu et al, 2018 ). Six deep CNNs have been applied to distinguish tongue lesions before tongue cancer fully takes hold ( Shamim et al, 2019 ).…”