A personalized computational model predicts cancer risk level of oral potentially malignant disorders and its web application for promotion of non‐invasive screening

Abstract

BackgroundDespite their high accuracy to recognize oral potentially malignant disorders (OPMDs) with cancer risk, non‐invasive oral assays are poor in discerning whether the risk is high or low. However, it is critical to identify the risk levels, since high‐risk patients need active intervention, while low‐risk ones simply need to be follow‐up. This study aimed at developing a personalized computational model to predict cancer risk level of OPMDs and explore its potential web application in OPMDs screening.MethodsEach enrolled patient was subjected to the following procedure: personal information collection, non‐invasive oral examination, oral tissue biopsy and histopathological analysis, treatment, and follow‐up. Patients were randomly divided into a training set (N = 159) and a test set (N = 107). Random forest was used to establish classification models. A baseline model (model‐B) and a personalized model (model‐P) were created. The former used the non‐invasive scores only, while the latter was incremented with appropriate personal features.ResultsWe compared the respective performance of cancer risk level prediction by model‐B, model‐P, and clinical experts. Our data suggested that all three have a similar level of specificity around 90%. In contrast, the sensitivity of model‐P is beyond 80% and superior to the other two. The improvement of sensitivity by model‐P reduced the misclassification of high‐risk patients as low‐risk ones. We deployed model‐P in web.opmd-risk.com, which can be freely and conveniently accessed.ConclusionWe have proposed a novel machine‐learning model for precise and cost‐effective OPMDs screening, which integrates clinical examinations, machine learning, and information technology.

Comparison of machine learning algorithms for the prediction of five‐year survival in oral squamous cell carcinoma

“…Firstly, the outcome being assessed varies from diagnostic prediction, risk assessment (which may include malignant transformation in pre‐cancerous lesions), prediction of recurrence and prediction of survival. Kourou et al did not cite any risk assessment studies for oral cancer; however, this is an area that has recently been explored with promising results 11 . Cancer machine learning studies may also be divided by the type of input data used, which varies considerably.…”

Role of Artificial Intelligence in the Early Diagnosis of Oral Cancer. A Scoping Review

“…(5) the diagnosis of OPMDs such as solar cheilosis [53], oral lichen planus [41,46], leukoplakia [29,40,58], with a prediction of their course [43]; (6) aids to differential diagnosis by classifying the lesions as benign or precancerous [50]; normal mucosa or oral cancer [27,32,35,36,39,42,62]; or as different benign, premalignant and malignant lesions [31,33,39,46,59]; (7) classification of oral cancer [56]; and (8) development of oral cancer risk predictive models [42,47,48,51,55,60].…”

Current Insights into Oral Cancer Diagnostics

“…AI technology has also been used in developing prediction models [ 109 , 110 , 111 ]. A recent study developed a personalized prediction model, web.opmd-risk.com (accessed on 15 May 2021), using an ML algorithm generated from 266 OPMDs patients in order to predict cancer risk.…”

“…This model may distinguish high-risk and low-risk lesions with high sensitivity and specificity. In addition, it was able to predict the risk of future oral cancer [ 111 ]. A 15-year cohort study was conducted to validate the ML algorithm for prediction of oral cancer survival risk stratification.…”