Treatment of lung diseases, which are the third most common cause of death in the world, is of great importance in the medical field. Many studies using lung sounds recorded with stethoscope have been conducted in the literature in order to diagnose the lung diseases with artificial intelligence-compatible devices and to assist the experts in their diagnosis. In this paper, ICBHI 2017 database which includes different sample frequencies, noise and background sounds was used for the classification of lung sounds. The lung sound signals were initially converted to spectrogram images by using time–frequency method. The short time Fourier transform (STFT) method was considered as time–frequency transformation. Two deep learning based approaches were used for lung sound classification. In the first approach, a pre-trained deep convolutional neural networks (CNN) model was used for feature extraction and a support vector machine (SVM) classifier was used in classification of the lung sounds. In the second approach, the pre-trained deep CNN model was fine-tuned (transfer learning) via spectrogram images for lung sound classification. The accuracies of the proposed methods were tested by using the ten-fold cross validation. The accuracies for the first and second proposed methods were 65.5% and 63.09%, respectively. The obtained accuracies were then compared with some of the existing results and it was seen that obtained scores were better than the other results.

Emotion recognition (ER) from Electroencephalogram (EEG) signals is a challenging task due to the non-linearity and non-stationarity nature of EEG signals. Existing feature extraction methods cannot extract the deep concealed characteristics of EEG signals from different layers for efficient classification scheme and also hard to select appropriate and effective feature extraction methods for different types of EEG data. Hence this study intends to develop an efficient deep feature extraction based method to automatically classify emotion status of people. In order to discover reliable deep features, five deep convolutional neural networks (CNN) models are considered: AlexNet, VGG16, ResNet50, SqueezeNet and MobilNetv2. Pre-processing, Wavelet Transform (WT), and Continuous Wavelet Transform (CWT) are employed to convert the EEG signals into EEG rhythm images then five well-known pretrained CNN models are employed for feature extraction. Finally, the proposed method puts the obtained features as input to the support vector machine (SVM) method for classifying them into binary emotion classes: valence and arousal classes. The DEAP dataset was used in experimental works. The experimental results demonstrate that the AlexNet features with Alpha rhythm produces better accuracy scores (91.07% in channel Oz) than the other deep features for the valence discrimination, and the MobilNetv2 features yields the highest accuracy score (98.93% in Delta rhythm (with channel C3) for arousal discrimination.

Changes in and around anatomical structures such as blood vessels, optic disc, fovea, and macula can lead to ophthalmological diseases such as diabetic retinopathy, glaucoma, age-related macular degeneration (AMD), myopia, hypertension, and cataracts. If these diseases are not diagnosed early, they may cause partial or complete loss of vision in patients. Fundus imaging is the primary method used to diagnose ophthalmologic diseases. In this study, a powerful R-CNN+LSTM-based approach is proposed that automatically detects eight different ophthalmologic diseases from fundus images. Deep features were extracted from fundus images with the proposed R-CNN+LSTM structure. Among the deep features extracted, those with high representative power were selected with an approach called NCAR, which is a multilevel feature selection algorithm. In the classification phase, the SVM algorithm, which is a powerful classifier, was used. The proposed approach is evaluated on the eight-class ODIR dataset. The accuracy (main metric), sensitivity, specificity, and precision metrics were used for the performance evaluation of the proposed approach. Besides, the performance of the proposed approach was compared with the existing approaches using the ODIR dataset.