TY - JOUR
T1 - Alcoholism Detection by Data Augmentation and Convolutional Neural Network with Stochastic Pooling
AU - Wang, Shui Hua
AU - Lv, Yi Ding
AU - Sui, Yuxiu
AU - Liu, Shuai
AU - Wang, Su Jing
AU - Zhang, Yu Dong
N1 - Publisher Copyright:
© 2017, Springer Science+Business Media, LLC.
PY - 2018/1/1
Y1 - 2018/1/1
N2 - 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.
AB - 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.
KW - Alcohol use disorder
KW - Average pooling
KW - Convolutional neural network
KW - Data augmentation
KW - Graphical processing unit
KW - Max pooling
KW - Stochastic pooling
UR - http://www.scopus.com/inward/record.url?scp=85038077811&partnerID=8YFLogxK
U2 - 10.1007/s10916-017-0845-x
DO - 10.1007/s10916-017-0845-x
M3 - Article
C2 - 29159706
AN - SCOPUS:85038077811
SN - 0148-5598
VL - 42
JO - Journal of Medical Systems
JF - Journal of Medical Systems
IS - 1
M1 - 2
ER -