The Possibility of Deep Learning-Based, Computer-Aided Skin Tumor Classifiers

Abstract

The incidence of skin tumors has steadily increased. Although most are benign and do not affect survival, some of the more malignant skin tumors present a lethal threat if a delay in diagnosis permits them to become advanced. Ideally, an inspection by an expert dermatologist would accurately detect malignant skin tumors in the early stage; however, it is not practical for every single patient to receive intensive screening by dermatologists. To overcome this issue, many studies are ongoing to develop dermatologist-level, computer-aided diagnostics. Whereas, many systems that can classify dermoscopic images at this dermatologist-equivalent level have been reported, a much fewer number of systems that can classify conventional clinical images have been reported thus far. Recently, the introduction of deep-learning technology, a method that automatically extracts a set of representative features for further classification has dramatically improved classification efficacy. This new technology has the potential to improve the computer classification accuracy of conventional clinical images to the level of skilled dermatologists. In this review, this new technology and present development of computer-aided skin tumor classifiers will be summarized.

Keywords: artificial intelligence; clinical image; convolutional neural network; deep learning; dermoscopy; skin tumor classifier.

Figure 1

Supervised or unsupervised training. (A) Supervised training which can predict classification or regression of the input data. (B) Unsupervised training which can cluster the data. (C) Semi-supervised training. First, train the algorithm by small number of labeled data. Then, use trained algorithm to “label” unlabeled data. Next, re-train algorithm using newly labeled data and originally labeled data. Finally, algorithm trained with all data and can predict classification or regression of the input data.

Figure 2

Supervised, semi-supervised, and unsupervised training. Supervised training needs labeled data but can learn the most efficient. Unsupervised training does not need labeled data which sometimes difficult to prepare, but can only cluster the input data. Semi-supervised training can produce labeled data from unlabeled data using small number of labeled data.

Figure 3

Skin tumor classifier by “traditional” machine learning. Digital image data needs pre-processing to remove noise or artifact to improve the efficacy of the next step. Pre-processed images then analyzed to extract features required for classification step. Finally, classifier use extracted features to classify input images.

Figure 4

Artificial neural network (ANN). (A) Single perceptron model which mimics the structure of biological neural networks in the human brain. Each node receives signal from other nodes (X1, X2… Xx). Add the multiplied values of input and weight (W) and when this sum(Σ) cross the threshold, then this node outputs signal. (B) An example of artificial neural network model which has hidden layer between input layer and output layer. All the nodes between the layers are fully connected and each connection has weight. Machine learning is adjusting each weights and thresholds in the network to reach the correct output (back propagation).

Figure 5

Convolutional neural network (CNN). (A) Schematic image of CNN. Between the convolutional layers, each nodes connect to distinct nodes of the previous layer, which is different compared with ANN (as in Figure 4B, all the nodes between the layers are fully connected). By this feature, CNN can successfully capture the spatial and temporal dependencies in an image. Then, the fully connected classifier output the result as a probability distribution. (B) An example of transfer learning in CNN. In this example, replace classifier and use pre-trained CNN layers as feature extractor. Then, train the system to fit the new task.

Figure 6

Accuracy of skin tumor classification by our CNN classifier. (A) Result of 14-class classification by dermatology trainees, board-certified dermatologists, and CNN classifier. Adapted from Fujisawa et al. (39). (B) Result of 2-class classification (benign or malignant). In both classification level, the accuracy of CNN surpassed board-certified dermatologists.