State‐of‐the‐Art Machine Learning Techniques Aiming to Improve Patient Outcomes Pertaining to the Cardiovascular System

Figures

• 

  Figure 1. This figure illustrates the relationships between AI, ML, and DL. DL is a subfield of ML, while ML is a subfield of AI. AI indicates artificial intelligence; DL, deep learning; ML, machine learning.

• 

  Figure 2. This figure illustrates the training on an RF classifier. RF is an ensemble machine learning algorithm. Let n be the number of trees in the random forest classifier; n different training sets are then generated using the bootstrapping technique, and for each training set, 1 decision tree is generated. The ovals in the trees represent the splits, while the rectangles represent the classes. While generating each tree, the most effective feature out of a random subset of features would be selected to create the splits. Gini's diversity index is a commonly used split criterion. During the phase of testing, features of new samples would be passed along all the trees. Each tree would vote for a decision, and the majority of the votes would represent the final decision. RF indicates random forest.

• 

  Figure 3. This figure illustrates the SVM binary classification algorithm, which has been trained over a sample data. Let class 1 refer to the samples belonging to the first class (on the left‐hand side) and class 2 refer to the samples belonging to the other class. The data points (both class 1 and class 2) which are encircled/starred, are the support vectors. The support vectors are those data points that the algorithm identifies to be hardest in getting correctly classified. The SVM algorithm picks an optimal hyperplane that maximizes the margins between itself and the support vectors. SVM indicates support vector machine.

• 

  Figure 4. Clustering is a form of unsupervised learning. Clustering is a task of grouping unannotated data into distinct groups, such that samples of the same group are more similar to each other than those from the other groups. In this figure, unannotated data (data) on the left‐hand side are provided as input to the k‐means algorithm with k=3, and the algorithm groups the raw data into 3 distinct clusters, namely cluster 1, cluster 2, and cluster 3, as shown on the right‐hand side.

• 

  Figure 5. The figure shows a simple CNN meant for classifying images (in this case, images of digits). Most CNN architectures include (1) convolutional layers, (2) pooling layers, and (3) dense (fully connected) layers. A convolutional layer typically has multiple filters (similar to the image filters), wherein the filter weights are allowed to change and learn from the data. Each of these filters is moved across the length and breadth of the entire image as it is convolved with the image pixel values. It should be noted that these filters act like feature extractors, and the output (feature maps) obtained after performing the convolution operation is used as input to the next layer. The pooling layer provides an approach to down sample the feature maps while summarizing the presence of features, either locally or globally. Also, the pooling layer acts like a feature detector that helps identify important features and to a certain degree helps in providing rotational and translational invariance. The dense layer is a fully connected network wherein each neuron receives input from each neuron of the previous layer. Typically, the dense layer contributes to the greatest number of learnable…

• 

  Figure 6. Illustration of an RNN. x i and y i are the input and output at the ith time step, respectively. In RNN, the output is dependent on (1) the current input, (2) the output from the previous time step, and (3) the network weights and biases. In other words, the RNN's output is dependent on the current and previous inputs together. This makes RNN suitable for analyzing sequential data. NN indicates neural network; RNN, recurrent neural network.