There has been a rapid increase in volume, variety, and velocity of clinical data available (1). The availability of this “Big Data” represents a potential for better evidence-based medicine allowing the delivery of more efficient healthcare with improved patient outcome (2). However, the absence of techniques for collecting, storing and analysing such large and complex data sets limited the utilisation of this “Big Data” in the past (2, 3). Recently, there has been a growing interest in the application of machine learning algorithms as a method to utilise this “Big Data” in research and healthcare settings (3, 4).
Machine learning describes a subfield of artificial intelligence which utilises statistical algorithms to identify patterns in large datasets. Based on previous learning, inferences or predictions can be made given novel data (3). Machine learning can be classified into two broad categories based on its approach: supervised and unsupervised learning. Supervised learning involves training of an algorithm with a set of existing data made up of inputs that are associated with a known output (i.e. labelled data). Once learned, through optimisation of its algorithms, predicting outcomes from previously unseen data becomes possible (3, 4). With unsupervised learning, there are no outputs to predict, and inputs are unlabelled. The aim is to infer patterns and the structure in the given data set to generate novel associations (4, 5).
Two areas within the medical field have gained particular attention for the application of machine learning: diagnosis and outcome prediction (4). Recently, there has been a drastic increase in the volume of literature describing the application of machine learning across a range of specialities within medicine and surgery (3–6). Otolaryngology-Head and Neck surgery is no exception. For example, studies have reported its use in diagnosing plethora of diseases of the head and neck through evaluation of patients’ presenting complaint (for example, voice analysis) (7, 8), or through evaluation of radiological, histological and/or endoscopic images (9–11). Other studies have reported its use in disease prognostication (for example, predicting hearing outcomes in sudden sensorineural hearing loss) (12), or in terms of predicting post-operative outcomes (for example, predicting post-operative complications in head and neck microvascular free tissue transfer) (13).
Aim
In this review, our aim is to evaluate the clinical application of machine learning in the field of Otolaryngology-Head and Neck surgery. Specifically, this will include the following subspecialties: otology and neurotology/lateral skull base surgery; rhinology and anterior skull base surgery; facial plastics; laryngology; head and neck surgery; paediatric otolaryngology.
Objectives
-
To evaluate the accuracy of machine learning models in diagnosing a clinical condition relevant to the field of Otolaryngology-Head and Neck surgery.
-
To evaluate the accuracy of machine learning models in disease prognosis for a condition relevant to the field of Otolaryngology-Head and Neck surgery.
-
To evaluate the accuracy of machine learning models in predicting the post-operative outcomes of Otolaryngology-Head and Neck surgical procedures.