MTANet: Multi-Task Attention Network for Automatic Medical Image Segmentation and Classification.

Medical image segmentation and classification are two of the most key steps in computer-aided clinical diagnosis. The region of interest were usually segmented in a proper manner to extract useful features for further disease classification. However, these methods are computationally complex and time-consuming. In this paper, we proposed a one-stage multi-task attention network (MTANet) which efficiently classifies objects in an image while generating a high-quality segmentation mask for each medical object. A reverse addition attention module was designed in the segmentation task to fusion areas in global map and boundary cues in high-resolution features, and an attention bottleneck module was used in the classification task for image feature and clinical feature fusion. We evaluated the performance of MTANet with CNN-based and transformer-based architectures across three imaging modalities for different tasks: CVC-ClinicDB dataset for polyp segmentation, ISIC-2018 dataset for skin lesion segmentation, and our private ultrasound dataset for liver tumor segmentation and classification. Our proposed model outperformed state-of-the-art models on all three datasets and was superior to all 25 radiologists for liver tumor diagnosis.

EPolar-UNet: An edge-attending polar UNet for automatic medical image segmentation with small datasets.

BACKGROUND: Medical image segmentation is one of the most key steps in computer-aided clinical diagnosis, geometric characterization, measurement, image registration, and so forth. Convolutional neural networks especially UNet and its variants have been successfully used in many medical image segmentation tasks. However, the results are limited by the deficiency in extracting high resolution edge information because of the design of the skip connections in UNet and the need for large available datasets. PURPOSE: In this paper, we proposed an edge-attending polar UNet (EPolar-UNet), which was trained on the polar coordinate system instead of classic Cartesian coordinate system with an edge-attending construction in skip connection path. METHODS: EPolar-UNet extracted the location information from an eight-stacked hourglass network as the pole for polar transformation and extracted the boundary cues from an edge-attending UNet, which consisted of a deconvolution layer and a subtraction operation. RESULTS: We evaluated the performance of EPolar-UNet across three imaging modalities for different segmentation tasks: CVC-ClinicDB dataset for polyp, ISIC-2018 dataset for skin lesion, and our private ultrasound dataset for liver tumor segmentation. Our proposed model outperformed state-of-the-art models on all three datasets and needed only 30%-60% of training data compared with the benchmark UNet model to achieve similar performances for medical image segmentation tasks. CONCLUSIONS: We proposed an end-to-end EPolar-UNet for automatic medical image segmentation and showed good performance on small datasets, which was critical in the field of medical image segmentation.

A ROS-responsive loaded desferoxamine (DFO) hydrogel system for traumatic brain injury therapy.

Traumatic brain injury (TBI) produces excess iron, and increased iron accumulation in the brain leads to lipid peroxidation and reactive oxygen species (ROSs), which can exacerbate secondary damage and lead to disability and death. Therefore, inhibition of iron overload and oxidative stress has a significant role in the treatment of TBI. Functionalized hydrogels with iron overload inhibiting ability and of oxidative stress inhibiting ability will greatly contribute to the repair of TBI. Herein, an injectable, post-traumatic microenvironment-responsive, ROS-responsive hydrogel encapsulated with deferrioxamine mesylate (DFO) was developed. The hydrogel is rapidly formed via dynamic covalent bonding between phenylboronic acid grafted hyaluronic acid (HA-PBA) and polyvinyl alcohol (PVA), and phenylboronate bonds are used to respond to and reduce ROS levels in damaged brain tissue to promote neuronal recovery. The release of DFO from HA-PBA/PVA hydrogels in response to ROS further promotes neuronal regeneration and recovery by relieving iron overload and thus eradicating ROS. In the Feeney model of Sprague Dawley rats, HA-PBA/PVA/DFO hydrogel treatment significantly improved the behavior of TBI rats and reduced the area of brain contusion in rats. In addition, HA-PBA/PVA/DFO hydrogel significantly reduced iron overload to reduce ROS and could effectively promote post-traumatic neuronal recovery. Its effects were also explored, and notably, HA-PBA/PVA/DFO hydrogel can reduce iron overload as well as ROS, thus protecting neurons from death. Thus, this injectable, biocompatible and ROS-responsive drug-loaded hydrogel has great potential for the treatment of TBI. This work suggests a novel method for the treatment of secondary brain injury by inhibiting iron overload and the oxidative stress response after TBI.

Human neural stem cell secretome relieves endoplasmic reticulum stress-induced apoptosis and improves neuronal functions after traumatic brain injury in a rat model.

Neural stem cell secretome (NSC-S) plays an important role in neuroprotection and recovery. Studies have shown that endoplasmic reticulum stress (ER stress) is involved in the progression of traumatic brain injury (TBI) and is a crucial cause of secondary damage and neuronal death after brain injury. Whether NSC-S is engaged in ER stress and ER stress-mediated neuronal apoptosis post-TBI has not been investigated. In the study, the Feeney SD male rat model was established. The results showed that NSC-S treatment significantly improved the behavior of rats with TBI. In addition, NSC-S relieved ER stress in TBI rats and was observed by transmission electron microscopy and western blot. The specific mechanism was further elucidated that restoration was achieved by alleviating the PERK-eIF2α pathway and thus protecting neurons from apoptosis. Notably, the discovery of calumenin (CALU) in NSC-S by liquid chromatography-tandem mass spectrometry (LC-MS/MS/MS) may be related to the protective effect of NSC-S on ER stress in neurons. Also, the mechanism by which it functions may be related to ubiquitination. In summary, NSC-S improved prognosis and ER stress in TBI rats and might be a promising treatment for relieving TBI.

Human Neural Stem Cell Secretome Inhibits Neuron Heme Uptake and Ferroptosis in Intracerebral Hemorrhage Through Nrf-2 Signaling Pathway.

Intracerebral hemorrhage (ICH) is a common subtype of stroke with a very high mortality rate, but there is still no effective cure. Increasing evidence suggests that heme accumulation and neuronal ferroptosis play an important role in secondary injury after ICH. Neural stem cells (NSCs), as seed cells of the central nervous system, have received much attention due to their abundant paracrine product components and low immunogenicity. In this study, we focused on the protective mechanism of neural stem cell secretome (NSC-S) against neuronal ferroptosis in an ICH mouse model using hemin-induced in vitro models and collagenase type IV-induced in vivo models. The results showed that NSC-S could ameliorate neurological deficits and reduce neuronal injury in ICH model mice. In addition, NSC-S reduced heme uptake and ferroptosis in hemin-treated N2a cells in vitro. NSC-S induced the activation of Nrf-2 signaling pathway. However, these effects of NSC-S were abolished by the Nrf-2 inhibitor ML385. Notably, HSPE1 in NSC-S may be associated with the protection of NSC-S against hemin-injured neurons via the Nrf-2 signaling pathway. In summary, NSC-S protects against secondary neuronal injury in ICH via the Nrf-2 signaling pathway. Also, this functionality may be implemented by HSPE1.

Recent advances in nanomedicine development for traumatic brain injury.

Traumatic brain injury (TBI) is one of the major causes of morbidity and mortality worldwide, and it is also a risk factor for neurodegeneration. However, there has not been perceptible progress in treating acute TBI over the last few years, mainly due to the inability of therapeutic drugs to cross the blood-brain barrier (BBB), failing to exert significant pharmacological effects on the brain parenchyma. Recently, nanomedicines are emerging as a powerful tool for the treatment of TBI where nanoscale materials (also called nanomaterials) are employed to deliver therapeutic agents. The advantages of using nanomaterials as a drug carrier include their high solubility and stability, high carrier capacity, site-specific, improved pharmacokinetics, and biodistribution. Keeping these points in consideration, this article reviews the pathophysiology, current treatment options, and emerging nanomedicine strategies for the treatment of TBI. The review will help readers to gain insight into the state-of-the-art of nanomedicine as a new tool for the treatment of TBI.

Human Neural Stem Cell Secretome Inhibits Lipopolysaccharide-Induced Neuroinflammation Through Modulating Microglia Polarization by Activating Peroxisome Proliferator-Activated Receptor Gamma.

Neuroinflammation is one of the typical events in multiple neurodegenerative diseases, whereas microglia are the critical participants in the pathogenesis of neuroinflammation. Several studies suggest that neural stem cells (NSCs) present immunomodulatory benefits due to their paracrine products, which contain mounting trophic factors. In the current study, the anti-inflammatory effects of NSC secretome (NSC-S) on lipopolysaccharide (LPS)-induced neuroinflammatory models were evaluated in vivo and the underlying mechanism was further investigated in vitro. It was revealed that NSC-S significantly attenuated the severity of LPS-induced behavior disorders and inflammatory response in mice. In vitro studies found that NSC-S significantly promoted the polarization of microglia from proinflammatory M1 to anti-inflammatory M2 phenotype, and reduced the production of proinflammatory cytokines, whereas elevated anti-inflammatory cytokines in BV2 cells. NSC-S promoted peroxisome proliferator-activated receptor gamma (PPAR-γ) pathway activation. However, these effects of NSC-S were abrogated by PPAR-γ inhibitor GW9662. Notably, the fatty acid-binding protein 5 (FABP5) in NSC-S may mediate PPAR-γ activation and inflammation remission. In summary, NSC-S promotes the regression of LPS-induced microglia-mediated inflammation through the PPAR-γ pathway. This function might be achieved through FABP5.

Automatic volumetric diagnosis of hepatocellular carcinoma based on four-phase CT scans with minimum extra information.

Summary: We built a deep-learning based model for diagnosis of HCC with typical images from four-phase CT and MEI, demonstrating high performance and excellent efficiency. Objectives: The aim of this study was to develop a deep-learning-based model for the diagnosis of hepatocellular carcinoma. Materials and methods: This clinical retrospective study uses CT scans of liver tumors over four phases (non-enhanced phase, arterial phase, portal venous phase, and delayed phase). Tumors were diagnosed as hepatocellular carcinoma (HCC) and non-hepatocellular carcinoma (non-HCC) including cyst, hemangioma (HA), and intrahepatic cholangiocarcinoma (ICC). A total of 601 liver lesions from 479 patients (56 years ± 11 [standard deviation]; 350 men) are evaluated between 2014 and 2017 for a total of 315 HCCs and 286 non-HCCs including 64 cysts, 178 HAs, and 44 ICCs. A total of 481 liver lesions were randomly assigned to the training set, and the remaining 120 liver lesions constituted the validation set. A deep learning model using 3D convolutional neural network (CNN) and multilayer perceptron is trained based on CT scans and minimum extra information (MEI) including text input of patient age and gender as well as automatically extracted lesion location and size from image data. Fivefold cross-validations were performed using randomly split datasets. Diagnosis accuracy and efficiency of the trained model were compared with that of the radiologists using a validation set on which the model showed matched performance to the fivefold average. Student's t-test (T-test) of accuracy between the model and the two radiologists was performed. Results: The accuracy for diagnosing HCCs of the proposed model was 94.17% (113 of 120), significantly higher than those of the radiologists, being 90.83% (109 of 120, p-value = 0.018) and 83.33% (100 of 120, p-value = 0.002). The average time analyzing each lesion by our proposed model on one Graphics Processing Unit was 0.13 s, which was about 250 times faster than that of the two radiologists who needed, on average, 30 s and 37.5 s instead. Conclusion: The proposed model trained on a few hundred samples with MEI demonstrates a diagnostic accuracy significantly higher than the two radiologists with a classification runtime about 250 times faster than that of the two radiologists and therefore could be easily incorporated into the clinical workflow to dramatically reduce the workload of radiologists.

Neural stem cell secretome exerts a protective effect on damaged neuron mitochondria in Parkinson's disease model.

Parkinson's disease (PD) is a common neurodegenerative disease. The main pathological changes are the loss of dopaminergic neurons and the formation of Lewy bodies. There is still no effective cure for PD, and cell replacement therapy has entered a bottleneck period due to tumorigenicity and rejection. Therefore, stem cell secretome has received widespread attention. However, the exploration of the secretome components of neural stem cells (NSCs) is still in its infancy. In this study, 6-hydroxydopamine (6-OHDA) was used to establish a PD rat model in vito and the PC12 cell-damaged model in vitro. The results indicated that the injection of neural stem cell-conditioned medium (NSC-CM) into the striatum and substantia nigra could improve the motor and non-motor deficits of PD rats and rescue the loss of dopaminergic neurons. In addition, NSC-CM alleviated 6-OHDA-induced apoptosis of PC12 cells, reduced the level of oxidative stress, and improved mitochondrial dysfunction in vitro. Parkinson disease protein 7 (Park7) was found in NSC-CM by Liquid chromatography-tandem mass spectrometry (LC-MS/MS), and it may be related to the protective effect of NSC-CM on 6-OHDA-injured neurons through Sirt1 pathway. In conclusion, NSC secretome might provide new ideas for the treatment of PD.

Peroxiredoxin 6 secreted by Schwann-like cells protects neuron against ischemic stroke in rats via PTEN/PI3K/AKT pathway.

Schwann cells can promote the survival of damaged neurons and axon regeneration by secreting or releasing some proteins and factors which may provide effective strategies to the remedy for ischemic stroke. The models of middle cerebral artery occlusion and oxygen-glucose deprivation (OGD) were established. Peroxiredoxin 6 (PRDX6) was found in Schwann-like cell conditioned medium (SCLC-CM) by mass spectrometry. The rehabilitative performance of SCLC-CM on focal cerebral ischemia of rats and on OGD-induced PC12 cells were assessed. SCLC-CM significantly improved neurological recovery, reducing the infarct volume of rats after stroke. PRDX6 could significantly inhibit neuron apoptosis in the OGD injury by mediating oxidative stress and activating the PTEN/PI3K/AKT pathway. In conclusion, PRDX6 secreted by Schwann-like cell protects neuron against focal cerebral ischemia, SCLC-CM might be a new effective early intervention for ischemic stroke.

Microvesicles from Schwann-Like Cells as a New Biomaterial Promote Axonal Growth.

Schwann cells promote axonal regeneration following peripheral nerve injury. However, in terms of clinical treatment, the therapeutic effects of Schwann cells are limited by their source. The transmission of microvesicles from neuroglia cells to axons is a novel communication mechanism in axon regeneration.To evaluate the effect of microvesicles released from Schwann-like cells on axonal regeneration, neural stem cells derived from human embryonic stem cells differentiated into Schwann-like cells, which presented a typical morphology and characteristics similar to those of schwann cells. The glial markers like MBP, P0, P75NTR, PMP-22, GFAP, HNK-1 and S100 were upregulated, whereas the neural stem markers like NESTIN, SOX1 and SOX2 were significantly downregulated in schwann-like cells. Microvesicles enhanced axonal growth in dorsal root ganglia neurons and regulated GAP43 expression in neuron-like cells (N2A and PC12) through the PTEN/PI3 K/Akt signaling pathway. A 5 mm section of sciatic nerve was transected in Sprague-Dawley rats. With microvesicles transplantation, regenerative nerves were evaluated after 6 weeks. Microvesicles increased sciatic function index scores, delayed gastrocnemius muscle atrophy and elevated ßIII-tubulin-labeled axons in vivo. Schwann-like cells serve as a convenient source and promote axonal growth by secreting microvesicles, which may potentially be used as bioengineering materials for nerve tissue repair.