The AI technology can automatically detect foods from low-quality, wearable camera-acquired real-world egocentric images with reasonable accuracy, reducing both the burden of data processing and privacy concerns.

Deep learning methods have made some achievements in the automatic skin lesion recognition, but there are still some problems such as limited training samples, too complicated network structure, and expensive computational costs. Considering the inherent power-efficiency, biological plausibility and good image recognition performance of spiking neural networks (SNNs), in this paper we make malignant melanoma and benign melanocytic nevi skin lesions classification using convolutional SNNs with unsupervised spike-timing-dependent plasticity (STDP) learning rule. Efficient temporal coding, event driven learning rule and winner-take-all (WTA) mechanism together ensure sparse spike coding and efficient learning of our networks which achieve an average accuracy of 83.8%. We further propose to use feature selection to select more diagnostic features to improve the classification performance of our networks. Our SNNs with feature selection reach an average accuracy of 87.7%. Experimental results show that comparing to CNNs that need to be trained from scratch, our SNNs (with and without feature selection) not only achieve much better classification accuracies but also have much better runtime efficiency. Moreover, although the pretrained CNNs models can achieve similar running time, our proposed SNNs are more stable and easier to use than the pretrained CNNs because we do not need to try many pretrained models any more, and our SNNs also have much better classification accuracies than the pretrained CNNs. In addition, our networks have only three convolutional layers, and the complexity of the model and the parameters that need to be trained in the networks are greatly reduced. Our works show that STDP-based SNNs are very beneficial for the implementation of automated skin lesion classifiers on small portable devices. INDEX TERMS Melanoma recognition, convolutional spiking neural networks, STDP, deep learning.

Ultrasound scanning has become a highly recommended examination in prenatal diagnosis in many countries. The accurate identification of fetal brain ultrasound scans is crucial to accurate head measurement and brain lesion detection, such as the measurement of the biparietal diameter and the detection of hydrocephalus. In recent years, deep learning has made great progress in the field of image processing. However, there are two difficulties in the identification of fetal brain ultrasound standard planes (FBSPs). First, since the fetal brain tissue is not mature, the fetal brain tissue features are not easy to be detected. Second, because of the expensive collection costs, the amount of labeled image data is limited, which can cause over-fitting and decrease the identification precision. In this study, we proposed a differential convolutional neural network (differential-CNN) to automatically identify six fetal brain standard planes (FBSPs) from the non-standard planes. In this differential-CNN framework, the additional differential feature maps were derived from the feature maps in the original CNN using differential operators. The derivation process did not increase the number of convolution layers and parameters. Moreover, the differential convolution maps have the large advantage of analyzing the directional pattern of pixels and their neighborhoods using additional variation calculations. Therefore, the differential convolution maps would result in good identification performance and cost no extra computational burden. To test the performance of these algorithms, we constructed a dataset consisting of 30,000 2D ultrasound images from 155 fetal subjects ranging from 16 to 34 weeks. The experimental results showed that this method achieved an accuracy of 92.93%. Our work shows that the differential-CNN can be used to facilitate the implementation of the automated identification of FBSPs. INDEX TERMS Convolutional neural network, ultrasound scan image, medical image processing, convolution techniques, differential operator.

Two-dimensional ultrasound scanning (US) has become a highly recommended examination in prenatal diagnosis in many countries. Accurate detection of abnormalities and correct fetal brain standard planes is the most necessary precondition for successful diagnosis and measurement. In the past few years, support vector machine (SVM) and other machine learning methods have been devoted to automatic recognition of 2D ultrasonic images, but the performance of recognition is not satisfactory due to the wide diversity of fetal postures, shortage of data, similarities between standard planes and other reasons. Especially in the recognition of fetal brain images, the features of fetal brain images such as shape, texture, color and others are very similar, which presents great challenges to the recognition work. In this study, we proposed two main methods based on deep convolutional neural networks to automatically recognize six standard planes of fetal brains. One is a deep convolutional neural network (CNN), and the other one is CNN-based domain transfer learning. To examine the performance of these algorithms, we constructed two datasets. Dataset 1 consists of 30,000 2D ultrasound images from 155 subjects between 16 and 34 weeks. Dataset 2, containing 1,200 images, was acquired from a research participant throughout 40 weeks, which is the entire pregnancy. Experimental results show that the proposed solutions achieve promising results and that the frameworks based on deep convolutional neural networks generally outperform the ones using other classical deep learning methods, thus demonstrating the great potential of convolutional neural networks in this area. INDEX TERMS Medical image processing, CNN, transfer learning. CHUNXIA DING received the Bachelor of Medicine degree from the Zhangjiakou Medical College, in 2003. She has been engaged in ultrasound diagnosis for 16 years and worked on prenatal screening for more than ten years. She is skilled in gynecology and obstetrics, heart, blood vessels, abdomen, superficial small organs, and neonatal brain, and other conventional diagnosis, prenatal screening, and fetal malformation diagnosis.

Invasive techniques are becoming increasingly important in the presurgical evaluation of epilepsy. Adopting the electrophysiological source imaging (ESI) of interictal scalp electroencephalography (EEG) to localize the epileptogenic zone remains a challenge. The accuracy of the preoperative localization of the epileptogenic zone is key to curing epilepsy. The T1 MRI and the boundary element method were used to build the realistic head model. To solve the inverse problem, the distributed inverse solution and equivalent current dipole (ECD) methods were employed to locate the epileptogenic zone. Furthermore, a combination of inverse solution algorithms and Granger causality connectivity measures was evaluated. The ECD method exhibited excellent focalization in lateralization and localization, achieving a coincidence rate of 99.02% (p<0.05) with the stereo electroencephalogram. The combination of ECD and the directed transfer function led to excellent matching between the information flow obtained from intracranial and scalp EEG recordings. The ECD inverse solution method showed the highest performance and could extract the discharge information at the cortex level from noninvasive low-density EEG data. Thus, the accurate preoperative localization of the epileptogenic zone could reduce the number of intracranial electrode implantations required.