Dissociable Neural Representations of Adversarially Perturbed Images in Convolutional Neural Networks and the Human Brain.

Despite the remarkable similarities between convolutional neural networks (CNN) and the human brain, CNNs still fall behind humans in many visual tasks, indicating that there still exist considerable differences between the two systems. Here, we leverage adversarial noise (AN) and adversarial interference (AI) images to quantify the consistency between neural representations and perceptual outcomes in the two systems. Humans can successfully recognize AI images as the same categories as their corresponding regular images but perceive AN images as meaningless noise. In contrast, CNNs can recognize AN images similar as corresponding regular images but classify AI images into wrong categories with surprisingly high confidence. We use functional magnetic resonance imaging to measure brain activity evoked by regular and adversarial images in the human brain, and compare it to the activity of artificial neurons in a prototypical CNN-AlexNet. In the human brain, we find that the representational similarity between regular and adversarial images largely echoes their perceptual similarity in all early visual areas. In AlexNet, however, the neural representations of adversarial images are inconsistent with network outputs in all intermediate processing layers, providing no neural foundations for the similarities at the perceptual level. Furthermore, we show that voxel-encoding models trained on regular images can successfully generalize to the neural responses to AI images but not AN images. These remarkable differences between the human brain and AlexNet in representation-perception association suggest that future CNNs should emulate both behavior and the internal neural presentations of the human brain.

Long-pulse Nd:YAG 1064-nm laser treatment for onychomycosis.

BACKGROUND: Recent research shows that lasers can inhibit fungal growth and that Nd:YAG 1064-nm lasers can penetrate as deep as the lower nail plate. The aim of this study was to observe the effect of a long-pulse Nd:YAG 1064-nm laser on 154 nails of 33 patients with clinically and mycologically proven onychomycosis. METHODS: Thirty-three patients with 154 nails affected by onychomycosis were randomly assigned to two groups, with the 154 nails divided into three sub-groups (II degree, III degree, and IV degree) according to the Scoring Clinical Index of Onychomycosis. The 15 patients (78 nails) in group 1 were given eight sessions with a one-week interval, and the 18 patients (76 nails) in group 2 were given four sessions with a one-week interval. RESULTS: In group 1, the effective rates at 8 weeks, 16 weeks, and 24 weeks were 63%, 62%, and 51%, respectively, and the effective rates in group 2 were 68%, 67%, and 53% respectively. The treatment effect was not significantly different between any sub-group pair (P > 0.05). CONCLUSIONS: Long pulse Nd:YAG 1064-nm laser was effective for onychomycosis. It is a simple and effective method without significant complications or side effects and is expected to become an alternative or replacement therapy for onychomycosis.

A case of systemic amyloidosis beginning with purpura.

Primary systemic amyloidosis is a relatively rare disease, caused when abnormal extracellular deposition of fibrillary protein builds up in a variety of target organs, such as heart, kidneys, lungs liver, and so forth. The symptoms of the disease are usually vague, while many kinds of auxiliary or laboratory examinations especially pathologic biopsy can provide a clue for the diagnosis. Here we described a case who had purpura-like lesions in the initial stage, followed by progressive malfunctions in the kidneys, the heart, the lungs, as well as the liver. The final diagnosis was primary systemic amyloidosis determined by skin pathologic biopsy. And the disease led to a fatal outcome within three months after the diagnosis.