Table 6 Machine learning models frequently used in the biomedical field

K-nearest neighbors

KNN

Simple model

Each feature value corresponds to a specific coordinate. The classification process consists of identifying the K nearest neighbours of a given data point and assigning it the most prevalent label among these neighbours [39].

K

Support vector machine

SVM

Support vector machine

In SVM, each data item is plot as a point in an n-dimensional space with the value of each feature related to the value of a specific coordinate. It performs classification by finding the hyper-plane that differentiates the two classes [41].

C

Stochastic Gradient Boosting

GBM

Ensemble model

By combining multiple weak learners, GBM creates a powerful ensemble model. In classification, it uses a specific loss function to process the classification results, providing high accuracy and flexibility [56].

n.trees; shrinkage; n.minobsinnode

eXtreme Gradient Boosting

XGBoost

Ensemble model

XGBoost is an optimised and enhanced version of GBM that enhances the gradient boosting algorithm by introducing regularisation, sparsity awareness, parallel learning, missing value handling and tree pruning [40].

max_depth; n_estimators; learning_rate; colsample_bytree…

Multilayer Perceptron

MLP

Feedforward artificial neural network

The MLP incorporates a series of hidden layers positioned between the input and output layers, forming a directly connected mechanism. This allows data to flow forward through the network, resembling a feed-forward network structure [42].

hidden_layer_size