Surface Wettability Prediction using ML/AI

Surface wettability is an important factor for determining the biological response to an implanted material. Surface energy is a dominant factor that affects surface wettability. The most common way of determining the surface energy is by measuring the contact angle of the liquid on the surface. Depending mostly on the contact angle of the droplet, the surface is further characterized as hydrophobic or hydrophilic. A hydrophilic surface attracts water strongly, whilst a hydrophobic surface repels water strongly. The material surface is considered to be extremely hydrophilic once the contact angle of liquid is less than 10°, hydrophilic when it has less than 90°, hydrophobic when it is larger than 90°, and extremely hydrophobic when it is more than 150°. Learning about surface wettability is crucial as the surface of any instrumentation is what comes in contact with the environment and is the most vulnerable to environmental effects. The outcome of this review will lead to a concise comparison and description of new methodologies and choosing the most appropriate image classification model for predicting surface wettability of surfaces in an economical and hassle-free manner. Neural networks are a powerful tool for the advancement of material characterization methods. This paper reviewed related studies to find the machine learning algorithms used for predicting wettability and the techniques employed. This is done to make the process of indicating the hydrophobic/hydrophilic nature of the surface simpler. Neural networks to predict the nature of a material based on its interaction with the given liquid have also been reviewed, under this, a comparative study between three major image classification algorithms namely convolution neural network(CNN), Recurrent neural network(RNN), and Artificial neural network(ANN) is done to find the most suited algorithm for the specific use case.