Machine learning application for prediction of locoregional recurrences in early oral tongue cancer: a Web-based prognostic tool

Abstract

Estimation of risk of recurrence in early-stage oral tongue squamous cell carcinoma (OTSCC) remains a challenge in the field of head and neck oncology. We examined the use of artificial neural networks (ANNs) to predict recurrences in early-stage OTSCC. A Web-based tool available for public use was also developed. A feedforward neural network was trained for prediction of locoregional recurrences in early OTSCC. The trained network was used to evaluate several prognostic parameters (age, gender, T stage, WHO histologic grade, depth of invasion, tumor budding, worst pattern of invasion, perineural invasion, and lymphocytic host response). Our neural network model identified tumor budding and depth of invasion as the most important prognosticators to predict locoregional recurrence. The accuracy of the neural network was 92.7%, which was higher than that of the logistic regression model (86.5%). Our online tool provided 88.2% accuracy, 71.2% sensitivity, and 98.9% specificity. In conclusion, ANN seems to offer a unique decision-making support predicting recurrences and thus adding value for the management of early OTSCC. To the best of our knowledge, this is the first study that applied ANN for prediction of recurrence in early OTSCC and provided a Web-based tool.Electronic supplementary materialThe online version of this article (10.1007/s00428-019-02642-5) contains supplementary material, which is available to authorized users.

Machine learning and treatment outcome prediction for oral cancer

“…There are few comparable data in the literature, although a recent systematic review reported SVM accuracy between 56.7 and 99.4% 26 . In a study of 311 early‐stage tongue SCCs, an artificial neural network (ANN) was used to characterize invasive histopathology and achieved 88% accuracy and 71% sensitivity for loco‐regional recurrence prediction, 16 whilst a decision forest algorithm to predict occult nodal metastasis in 71 T1/T2 OSCC patients reported an AUC of 0.84, with 91.7% sensitivity and 57.6% specificity 17 . Predictive ability of these models may have been improved by the measurement of specific disease outcomes in better defined patient cohorts with same stage disease.…”

“…Their popularity is based upon a presumed ability to sequentially detect patterns, garner information and undergo automated training based on data input, especially complex non‐homogenous data, ultimately making clinical predictions with minimal human intervention 13,14 . Whilst a degree of predictive accuracy for algorithms has been reported, in particular the use of support vector machines, boosted decision trees, decision forest and artificial neural networks, there is a need to validate the predictive power of machine learning by analysing disease progression within well‐defined OSCC patient cohorts prior to widespread translation to clinical practice 15‐19 …”

Predictive models for stage and risk classification in head and neck squamous cell carcinoma (HNSCC)

“…However, the study was done without considering stages/grade. Recently, a machine learning study used neural networks to predict recurrences in tongue cancer (Alabi et al, 2019). Kim et al (2019) developed predictive models for survival prediction in oral cancer patients.…”

Use of Data Analysis Methods in Dental Publications: Is There Evidence of a Methodological Change?

“…In addition, various analysis methods of agreement and reliability have been implemented in dentistry to measure the extent of agreement among raters, or to validate the questionnaires or diagnostic tests [24,25]. The broad introduction of new data analysis methods that are compatible with the rapid expansion in computing capability has reached also medical publications [26][27][28]. Bayesian methods, artificial neural networks (ANN), and machine learning (ML) are some examples of these highly computational approaches [29].…”