STNet: shape and texture joint learning through two-stream network for knowledge-guided image recognition.

Introduction: The human brain processes shape and texture information separately through different neurons in the visual system. In intelligent computer-aided imaging diagnosis, pre-trained feature extractors are commonly used in various medical image recognition methods, common pre-training datasets such as ImageNet tend to improve the texture representation of the model but make it ignore many shape features. Weak shape feature representation is disadvantageous for some tasks that focus on shape features in medical image analysis. Methods: Inspired by the function of neurons in the human brain, in this paper, we proposed a shape-and-texture-biased two-stream network to enhance the shape feature representation in knowledge-guided medical image analysis. First, the two-stream network shape-biased stream and a texture-biased stream are constructed through classification and segmentation multi-task joint learning. Second, we propose pyramid-grouped convolution to enhance the texture feature representation and introduce deformable convolution to enhance the shape feature extraction. Third, we used a channel-attention-based feature selection module in shape and texture feature fusion to focus on the key features and eliminate information redundancy caused by feature fusion. Finally, aiming at the problem of model optimization difficulty caused by the imbalance in the number of benign and malignant samples in medical images, an asymmetric loss function was introduced to improve the robustness of the model. Results and conclusion: We applied our method to the melanoma recognition task on ISIC-2019 and XJTU-MM datasets, which focus on both the texture and shape of the lesions. The experimental results on dermoscopic image recognition and pathological image recognition datasets show the proposed method outperforms the compared algorithms and prove the effectiveness of our method.

MPMR: Multi-Scale Feature and Probability Map for Melanoma Recognition.

Malignant melanoma (MM) recognition in whole-slide images (WSIs) is challenging due to the huge image size of billions of pixels and complex visual characteristics. We propose a novel automatic melanoma recognition method based on the multi-scale features and probability map, named MPMR. First, we introduce the idea of breaking up the WSI into patches to overcome the difficult-to-calculate problem of WSIs with huge sizes. Second, to obtain and visualize the recognition result of MM tissues in WSIs, a probability mapping method is proposed to generate the mask based on predicted categories, confidence probabilities, and location information of patches. Third, considering that the pathological features related to melanoma are at different scales, such as tissue, cell, and nucleus, and to enhance the representation of multi-scale features is important for melanoma recognition, we construct a multi-scale feature fusion architecture by additional branch paths and shortcut connections, which extracts the enriched lesion features from low-level features containing more detail information and high-level features containing more semantic information. Fourth, to improve the extraction feature of the irregular-shaped lesion and focus on essential features, we reconstructed the residual blocks by a deformable convolution and channel attention mechanism, which further reduces information redundancy and noisy features. The experimental results demonstrate that the proposed method outperforms the compared algorithms, and it has a potential for practical applications in clinical diagnosis.

SIRT1 protects against myocardial ischemia-reperfusion injury via activating eNOS in diabetic rats.

BACKGROUND: Diabetic patients are more sensitive to myocardial ischemic injury than non-diabetic patients. Silent information regulator 1 (SIRT1) is a nicotinamide adenine dinucleotide-dependent histone deacetylase making the heart more resistant to ischemic injury. As SIRT1 expression is considered to be reduced in diabetic heart, we therefore hypothesized that up-regulation of SIRT1 in the diabetic heart may overcome its increased susceptibility to ischemic injury. METHODS: Male Sprague-Dawley rats were fed with high-fat diet and injected with streptozotocin once to induce diabetes. Diabetic rats received injections of adenoviral vectors encoding SIRT1 (Ad-SIRT1) at five myocardial sites. Four days after adenoviral injection, the rats were subjected to myocardial ischemia and reperfusion (MI/R). Outcome measures included left ventricular function, infarct size, cellular death and oxidative stress. RESULTS: Delivery of Ad-SIRT1 into the hearts of diabetic rats markedly increased SIRT1 expression. Up-regulation of SIRT1 in diabetic hearts improved cardiac function and reduced infarct size to the extent as in non-diabetic animals following MI/R, which was associated with reduced serum creatine kinase-MB, lactate dehydrogenase activities and cardiomyocyte apoptosis. Moreover, Ad-SIRT1 reduced the increase in the superoxide generation and malonaldialdehyde content and simultaneously increased the antioxidant capability. Furthermore, Ad-SIRT1 increased eNOS phosphorylation and reduced eNOS acetylation in diabetic hearts. NOS inhibitor L-NAME inhibited SIRT1-enhanced eNOS phosphorylation, and blunted SIRT1-mediated anti-apoptotic and anti-oxidative effects and cardioprotection. CONCLUSIONS: Overexpression of SIRT1 reduces diabetes-exacerbated MI/R injury and oxidative stress via activating eNOS in diabetic rats. The findings suggest SIRT1 may be a promising novel therapeutic target for diabetic cardiac complications.

Allicin protects rat cortical neurons against mechanical trauma injury by regulating nitric oxide synthase pathways.

Allicin, a small molecule that is responsible for the typical smell and most of the functions of garlic, possesses a broad spectrum of pharmacological activities and is considered to have therapeutic potential in many pathologic conditions. In the present study, we investigated the potential protective effect of allicin in an in vitro model of traumatic brain injury (TBI) using primary cultured rat cortical neurons. We found that allicin treatment significantly reduced mechanical trauma-induced lactate dehydrogenase (LDH) release and inhibited apoptotic neuronal death in a dose-dependent manner. These protective effects were observed even if allicin treatment was delayed to 2h after injury. Allicin significantly decreased the expression of inducible nitric oxide synthase (iNOS) and increased the phosphorylation of endothelial NOS (eNOS) but had no effect on neuronal NOS (nNOS) expression. Allicin-induced protection in cortical neurons was augmented by iNOS and nNOS antagonists and was partly reversed by blocking eNOS phosphorylation. In addition, allicin treatment inhibited the TBI-induced activation of ERK and further enhanced the phosphorylation of Akt in TBI-injured neurons. The Akt inhibitor LY294002 attenuated the allicin-induced increase in eNOS expression and phosphorylation, whereas the ERK inhibitor PD98059 had opposite effects on the expression of iNOS and eNOS. Pretreatment with LY294002 or PD98059 partly prevented or further enhanced allicin-induced neuroprotection, respectively. Collectively, these data demonstrate that allicin treatment may be an effective therapeutic strategy for traumatic neuronal injury and that the potential underlying mechanism involves Akt- and ERK-mediated regulation of NOS pathways.