Alcohol use disorder (AUD) is an important brain disease. It alters the brain structure. Recently, scholars tend to use computer vision based techniques to detect AUD. We collected 235 subjects, 114 alcoholic and 121 non-alcoholic. Among the 235 image, 100 images were used as training set, and data augmentation method was used. The rest 135 images were used as test set. Further, we chose the latest powerful technique-convolutional neural network (CNN) based on convolutional layer, rectified linear unit layer, pooling layer, fully connected layer, and softmax layer. We also compared three different pooling techniques: max pooling, average pooling, and stochastic pooling. The results showed that our method achieved a sensitivity of 96.88%, a specificity of 97.18%, and an accuracy of 97.04%. Our method was better than three state-of-the-art approaches. Besides, stochastic pooling performed better than other max pooling and average pooling. We validated CNN with five convolution layers and two fully connected layers performed the best. The GPU yielded a 149× acceleration in training and a 166× acceleration in test, compared to CPU.

No abstract

It is of enormous significance to detect abnormal brains automatically. This paper develops an efficient pathological brain detection system based on the artificial intelligence method. We first extract brain edges by a Canny edge detector. Next, we estimated the fractal dimension using box counting method with grid sizes of 1, 2, 4, 8, and 16, respectively. Afterward, we employed the single-hidden layer feedforward neural network. Finally, we proposed an improved particle swarm optimization based on three-segment particle representation, time-varying acceleration coefficient, and chaos theory. This three-segment particle representation encodes the weights, biases, and number of hidden neuron. The statistical analysis showed the proposed method achieves the detection accuracies of 100%, 98.19%, and 98.08% over three benchmark data sets. Our method costs merely 0.1984 s to predict one image. Our performance is superior to the 11 state-of-the-art approaches.INDEX TERMS Minkowski Bouligand dimension, genetic algorithm, artificial bee colony, logistic map, number of hidden neuron, K-fold cross validation.