Advances in Neural Information Processing Systems 27

Diabetic Foot Ulcer (DFU) is the main complication of Diabetes, which, if not properly treated, may lead to amputation. One of the approaches of DFU treatment depends on the attentiveness of clinicians and patients. This treatment approach has drawbacks such as the high cost of the diagnosis as well as the length of treatment. Although this approach gives powerful results, the need for a remote, cost-effective, and convenient DFU diagnosis approach is urgent. In this paper, we introduce a new dataset of 754-ft images which contain healthy skin and skin with a diabetic ulcer from different patients. A novel Deep Convolutional Neural Network, DFU_QUTNet, is also proposed for the automatic classification of normal skin (healthy skin) class versus abnormal skin (DFU) class. Stacking more layers to a traditional Convolutional Neural Network to make it very deep does not lead to better performance, instead leading to worse performance due to the gradient. Therefore, our proposed DFU_QUTNet network is designed based on the idea of increasing the width of the network while keeping the depth compared to the state-of-the-art networks. Our network has been proven to be very beneficial for gradient propagation, as the error can be back-propagated through multiple paths. It also helps to combine different levels of features at each step of the network. Features extracted by DFU_QUTNet network are used to train Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) classifiers. For the sake of comparison, we have fine-tuned then re-trained and tested three pre-trained deep learning networks (GoogleNet, VGG16, and AlexNet) for the same task. The proposed DFU_QUTNet network outperformed the state-of-the-art CNN networks by achieving the F1-score of 94.5%.

/pdf/dfu-qutnet-diabetic-foot-ulcer-classification-using-novel-1u7dye2el1.pdf

DFU_QUTNet: diabetic foot ulcer classification using novel deep convolutional neural network

Accurate prediction of the early stage of Alzheimer's disease (AD) is important but very challenging. The goal of this study was to utilize predictors for diagnosis conversion to AD based on integrating resting-state functional MRI (rs-fMRI) connectivity analysis and structural MRI (sMRI). We included 177 subjects in this study and aimed at identifying patients with mild cognitive impairment (MCI) who progress to AD, MCI converter (MCI-C), patients with MCI who do not progress to AD, MCI non-converter (MCI-NC), patients with AD, and healthy controls (HC). The graph theory was used to characterize different aspects of the rs-fMRI brain network by calculating measures of integration and segregation. The cortical and subcortical measurements, e.g., cortical thickness, were extracted from sMRI data. The rs-fMRI graph measures were combined with the sMRI measures to construct input features of a support vector machine (SVM) and classify different groups of subjects. Two feature selection algorithms [i.e., the discriminant correlation analysis (DCA) and sequential feature collection (SFC)] were used for feature reduction and selecting a subset of optimal features. Maximum accuracy of 67 and 56% for three-group ("AD, MCI-C, and MCI-NC" or "MCI-C, MCI-NC, and HC") and four-group ("AD, MCI-C, MCI-NC, and HC") classification, respectively, were obtained with the SFC feature selection algorithm. We also identified hub nodes in the rs-fMRI brain network which were associated with the early stage of AD. Our results demonstrated the potential of the proposed method based on integration of the functional and structural MRI for identification of the early stage of AD.

/pdf/identification-of-the-early-stage-of-alzheimer-s-disease-2cxt5cx2on.pdf

Identification of the Early Stage of Alzheimer's Disease Using Structural MRI and Resting-State fMRI.

Faulted line detection is a key step of intelligent fault diagnosis of distribution systems, laying the foundation for the further fault location and service restoration. A novel single line-to-ground (SLG) faulted line detection method based on the feature fusion framework is proposed. In the proposed framework, one-dimensional convolutional neural network is employed as a powerful tool to extract more effective features. In addition, there is an imbalance phenomenon between data of the faulted line and healthy lines when a data-driven model is used in the faulted line detection. The proposed framework offers an avenue for overcoming it and improves the accuracy of detection. Considering the limited data of SLG faults in actual power systems, prior knowledge of SLG fault detection is integrated into the data-driven model, which proves useful in reducing dependence on the training data quantity. The experiments verified the superior performance of the proposed feature fusion framework-based method.

Single Line-to-Ground Faulted Line Detection of Distribution Systems With Resonant Grounding Based on Feature Fusion Framework

Humans have an excellent ability to recognize freehand sketch drawings despite their abstract and sparse structures. Understanding freehand sketches with automated methods is a challenging task due to the diversity and abstract structures of these sketches. In this paper, we propose an efficient freehand sketch recognition scheme, which is based on the feature-level fusion of Convolutional Neural Networks (CNNs) in the transfer learning context. Specifically, we analyse different layer performances of distinct ImageNet pretrained CNNs and combine best performing layer features within the CNN-SVM pipeline for recognition. We also employ Principal Component Analysis (PCA) to reduce the fused deep feature dimensions to ensure the efficiency of the recognition application on the limited-capacity devices. We perform evaluations on two real sketch benchmark datasets, namely the Sketchy and the TU-Berlin to show the effectiveness of the proposed scheme. Our experimental results show that, the feature-level fusion scheme with the PCA achieves a recognition accuracy of 97.91% and 72.5% on the Sketchy and TU-Berlin datasets, respectively. This result is promising when compared with the human recognition accuracy of 73.1% on the TU-Berlin dataset. We also develop a sketch recognition application for smart devices to demonstrate the proposed scheme.

Sketch recognition using transfer learning

Understanding hand-free sketches with automated methods is a challenging task due to the diversity and abstract structures of the sketches. In this study, we propose a robust fusion scheme, namely feature-level fusion that use deep convolutional neural networks (CNNs) for recognizing hand-free sketches and develop a sketch recognition application for smartphones based on client-server application architecture. We employ inter-layer CNN features to capture different levels of abstractions of sketches along with fusion operator. Our results on TU-Berlin hand-free sketch benchmark dataset show that, our proposed feature-level fusion scheme achieves a recognition accuracy of 69.175%. This result is promising when compared with the human recognition accuracy of 73.1% on the same dataset.

Feature-level fusion of deep convolutional neural networks for sketch recognition on smartphones

Today, video data, as a powerful multimedia component, is accompanied by some problems with increasing usage in communication, health, education, and social media in particular. Classification and detection of concepts in video data by automatic methods are some of these challenging problems. In this study, we propose a video classification system, which incorporates deep convolutional neural networks (CNNs) by leveraging feature selection and data fusion techniques to improve the accuracy of the classification. Principal Component Analysis (PCA) as a feature selection method and Discriminant Correlation Analysis (DCA) technique, which incorporates class associations into the correlation analysis of feature sets for data fusion, are applied to the problem at the feature level. Support Vector Machines (SVMs) have been trained with new feature vectors obtained from different deep convolutional neural networks by feature selection and data fusion methods. The proposed method is tested for 38 concepts on TRECVID 2013 SIN video task dataset and the results are evaluated. Our results show that the classification accuracy is improved by 4% with an accuracy of 50.27% when the proposed data fusion and feature selection techniques are used.

Emel Boyaci

Papers

Sketch recognition using transfer learning

Feature-level fusion of deep convolutional neural networks for sketch recognition on smartphones

Video classification based on ConvNet collaboration and feature selection