Tumors evade immune cytotoxicity by altering the surface topology of NK cells.

The highly variable response rates to immunotherapies underscore our limited knowledge about how tumors can manipulate immune cells. Here the membrane topology of natural killer (NK) cells from patients with liver cancer showed that intratumoral NK cells have fewer membrane protrusions compared with liver NK cells outside tumors and with peripheral NK cells. Dysregulation of these protrusions prevented intratumoral NK cells from recognizing tumor cells, from forming lytic immunological synapses and from killing tumor cells. The membranes of intratumoral NK cells have altered sphingomyelin (SM) content and dysregulated serine metabolism in tumors contributed to the decrease in SM levels of intratumoral NK cells. Inhibition of SM biosynthesis in peripheral NK cells phenocopied the disrupted membrane topology and cytotoxicity of the intratumoral NK cells. Targeting sphingomyelinase confers powerful antitumor efficacy, both as a monotherapy and as a combination therapy with checkpoint blockade.

Mitochondrial fragmentation limits NK cell-based tumor immunosurveillance.

Natural killer (NK) cells have crucial roles in tumor surveillance. We found that tumor-infiltrating NK cells in human liver cancers had small, fragmented mitochondria in their cytoplasm, whereas liver NK cells outside tumors, as well as peripheral NK cells, had normal large, tubular mitochondria. This fragmentation was correlated with reduced cytotoxicity and NK cell loss, resulting in tumor evasion of NK cell-mediated surveillance, which predicted poor survival in patients with liver cancer. The hypoxic tumor microenvironment drove the sustained activation of mechanistic target of rapamycin-GTPase dynamin-related protein 1 (mTOR-Drp1) in NK cells, resulting in excessive mitochondrial fission into fragments. Inhibition of mitochondrial fragmentation improved mitochondrial metabolism, survival and the antitumor capacity of NK cells. These data reveal a mechanism of immune escape that might be targetable and could invigorate NK cell-based cancer treatments.

Quantitative imaging for predicting hematoma expansion in intracerebral hemorrhage: A multimodel comparison.

BACKGROUND: Several studies report that radiomics provides additional information for predicting hematoma expansion in intracerebral hemorrhage (ICH). However, the comparison of diagnostic performance of radiomics for predicting revised hematoma expansion (RHE) remains unclear. METHODS: The cohort comprised 312 consecutive patients with ICH. A total of 1106 radiomics features from seven categories were extracted using Python software. Support vector machines achieved the best performance in both the training and validation datasets. Clinical factors models were constructed to predict RHE. Receiver operating characteristic curve analysis was used to assess the abilities of non-contrast computed tomography (NCCT) signs, radiomics features, and combined models to predict RHE. RESULTS: We finally selected the top 21 features for predicting RHE. After univariate analysis, 4 clinical factors and 5 NCCT signs were selected for inclusion in the prediction models. In the training and validation dataset, radiomics features had a higher predictive value for RHE (AUC = 0.83) than a single NCCT sign and expansion-prone hematoma. The combined prediction model including radiomics features, clinical factors, and NCCT signs achieved higher predictive performances for RHE (AUC = 0.88) than other combined models. CONCLUSIONS: NCCT radiomics features have a good degree of discrimination for predicting RHE in ICH patients. Combined prediction models that include quantitative imaging significantly improve the prediction of RHE, which may assist in the risk stratification of ICH patients for anti-expansion treatments.

Interactions between driver genes shape the signaling pathway landscape and direct hepatocellular carcinoma therapy.

Hepatocellular carcinoma (HCC) is one of the most lethal malignancies, whose initiation and development are driven by alterations in driver genes. In this study, we identified four driver genes (TP53, PTEN, CTNNB1, and KRAS) that show a high frequency of somatic mutations or copy number variations (CNVs) in patients with HCC. Four different spontaneous HCC mouse models were constructed to screen for changes in various kinase signaling pathways. The sgTrp53 + sgPten tumor upregulated mTOR and noncanonical nuclear factor-κB signaling, which was shown to be strongly inhibited by rapamycin (an mTOR inhibitor) in vitro and in vivo. The JAK-signal transducer and activator of transcription (STAT) signaling was activated in Ctnnb1mut + sgPten tumor, the proliferation of which was strongly inhibited by napabucasin (a STAT3 inhibitor). Additionally, mTOR, cytoskeleton, and AMPK signaling were upregulated while rapamycin and ezrin inhibitors exerted potent antiproliferative effects in sgPten + KrasG12D tumor. We found that JAK-STAT, MAPK, and cytoskeleton signaling were activated in sgTrp53 + KrasG12D tumor and the combination of sorafenib and napabucasin led to the complete inhibition of tumor growth in vivo. In patients with HCC who had the same molecular classification as our mouse models, the downstream signaling pathway landscapes associated with genomic alterations were identical. Our research provides novel targeted therapeutic options for the clinical treatment of HCC, based on the presence of specific genetic alterations within the tumor.

Macrophage-Derived MMP-9 and MMP-2 are Closely Related to the Rupture of the Fibrous Capsule of Hepatocellular Carcinoma Leading to Tumor Invasion.

BACKGROUND: Hepatocellular carcinoma (HCC) is an aggressive tumor with a poor clinical prognosis. Rupture of the fibrous capsule (FC) is a very important clinical phenomenon in the invasion and metastasis of HCC. FC is mainly composed of type I collagen (COL1A1). However, it is not clear what caused the FC rupture. In this study, we aimed to determine whether the rupture of FC in HCC patients was related to macrophage-derived MMP-9 and MMP-2, and their clinical diagnostic value for FC rupture. RESULTS: By performing immunohistochemical and immunofluorescence staining of ruptured FC and intact FC, the results showed that the ruptured area of FC aggregated a large number of macrophages with MMP-9 and MMP-2. Western blot analysis and Quantitative real-time PCR were used to assess the expression of MMP-9 and MMP-2 in the ruptured and relatively intact area of FC in ruptured FC patients, and the results revealed a significantly different expression of MMP-9 and MMP-2. ELISA experiments show that we could discriminate effectively between ruptured FC and intact FC by MMP-9 and MMP-2. CONCLUSIONS: Taken together, macrophage-derived MMP-9 and MMP-2 were closely related to the rupture of the FC of HCC and subsequently led to the migration and invasion of the tumor cells through the ruptured area of FC to the para cancer. It is suggested that when performing surgical resection, it is necessary to expand the range of tumor resection for patients with ruptured FC and hence reduce the possibility of recurrence and metastasis in HCC patients.

MicroRNA Transcriptomics Analysis Identifies Dysregulated Hedgehog Signaling Pathway in a Mouse Model of Acute Intracerebral Hemorrhage Exposed to Hyperglycemia.

OBJECTIVE: Hyperglycemia is often observed in the patients after acute stroke. This study aims to elucidate the potential effect and mechanism of hyperglycemia by screening microRNAs expression in intracerebral hemorrhage mice. METHODS: We employed the collagenase model of intracerebral hemorrhage. Twenty male C57BL/6 mice were used and randomly divided in normo- and hyperglycemic. The hyperglycemia was induced by intraperitoneally injection of 50% of Dextrose (8 mL/kg) 3 hours after intracerebral hemorrhage. The neurologic impairment was investigated by neurologic deficit scale. To study the specific mechanisms of hyperglycemia, microRNAs expression in perihematomal area was investigated by RNA sequencing. MicroRNAs expression in hyperglycemic intracerebral hemorrhage animals were compared normoglycemic mice. Functional annotation analysis was used to indicate potential pathological pathway, underlying observed effects. Finally, polymerase chain reaction validation was administered. RESULTS: Intraperitoneal injection of dextrose significantly increased blood glucose level. That was associated with aggravation of neurological deficits in hyperglycemic compared to normoglycemic animals. A total of 73 differentially expressed microRNAs were identified via transcriptomics analysis. Bioinformatics analyses showed that these microRNAs were significantly altered in several signaling pathways, of which the hedgehog signaling pathway was regarded as the most potential pathway associated with the effect of hyperglycemia on acute intracerebral hemorrhage. Furthermore, polymerase chain reaction results validated the correlation between microRNAs and hedgehog signaling pathway. CONCLUSIONS: MicroRNA elevated in hyperglycemia group may be involved in worsening the neurological function via inhibiting the hedgehog signaling, which provides a novel molecular physiological mechanism and lays the foundation for treatment of intracerebral hemorrhage.

Quantitative analysis of abnormalities in gynecologic cytopathology with deep learning.

Cervical cancer is one of the most frequent cancers in women worldwide, yet the early detection and treatment of lesions via regular cervical screening have led to a drastic reduction in the mortality rate. However, the routine examination of screening as a regular health checkup of women is characterized as time-consuming and labor-intensive, while there is lack of characteristic phenotypic profile and quantitative analysis. In this research, over the analysis of a privately collected and manually annotated dataset of 130 cytological whole-slide images, the authors proposed a deep-learning diagnostic system to localize, grade, and quantify squamous cell abnormalities. The system can distinguish abnormalities at the morphology level, namely atypical squamous cells of undetermined significance, low-grade squamous intraepithelial lesion, high-grade squamous intraepithelial lesion, and squamous cell carcinoma, as well as differential phenotypes of normal cells. The case study covered 51 positive and 79 negative digital gynecologic cytology slides collected from 2016 to 2018. Our automatic diagnostic system demonstrated its sensitivity of 100% at slide-level abnormality prediction, with the confirmation with three pathologists who performed slide-level diagnosis and training sample annotations. In the cellular-level classification, we yielded an accuracy of 94.5% in the binary classification between normality and abnormality, and the AUC was above 85% for each subtype of epithelial abnormality. Although the final confirmation from pathologists is often a must, empirically, computer-aided methods are capable of the effective extraction, interpretation, and quantification of morphological features, while also making it more objective and reproducible.

Hydrocephalus Growth: Definition, Prevalence, Association with Poor Outcome in Acute Intracerebral Hemorrhage.

BACKGROUND/OBJECTIVES: To propose a novel definition for hydrocephalus growth and to further describe the association between hydrocephalus growth and poor outcome among patients with intracerebral hemorrhage (ICH). METHODS: We analyzed consecutive patients who presented within 6 h after ICH ictus between July 2011 and June 2017. Follow-up CT scans were performed within 36 h after initial CT scans. The degree of hydrocephalus were evaluated by the hydrocephalus score of Diringer et al. The optimal increase of the hydrocephalus scores between initial and follow-up CT scan was estimated to define hydrocephalus growth. Poor long-term outcome was defined as a modified Rankin Scale of 4-6 at 3 months. Multivariate logistic regression analysis was performed to investigate the hydrocephalus growth for predicting 30-day mortality, 90-day mortality, and poor long-term outcome. RESULTS: A total of 321 patients with ICH were included in the study. Of 64 patients with hydrocephalus growth, 34 (53.1%) patients presented with both concurrent hematoma expansion and intraventricular hemorrhage (IVH) growth. After adjusting for potential confounding factors, hydrocephalus growth independently predicted 30-day mortality, 90-day mortality, and 90-day poor long-term outcome in multivariate logistic regression analysis. Hydrocephalus growth showed higher accuracy for predicting 30-day mortality, 90-day mortality, and poor long-term outcome than IVH growth or hematoma expansion, respectively. CONCLUSIONS: Hydrocephalus growth is defined by strongly predictive of short- or long-term mortality and poor outcome at 90 days, and might be a potential indicator for assisting clinicians for clinical decision-making.

Importance of lipid ratios for predicting intracranial atherosclerotic stenosis.

BACKGROUND: This study aims to investigate the association of lipid ratios with intracranial atherosclerotic stenosis (ICAS) in a Chinese population. METHODS: This cross-sectional study included 658 consecutive patients with ischemic stroke. Intracranial and extracranial arteries were evaluated for atherosclerotic stenosis using digital subtraction angiography or computed tomography angiography. Lipid ratios [total cholesterol (TC)/high-density lipoprotein-cholesterol (HDL-C), triglycerides (TG)/HDL-C, low-density lipoprotein-cholesterol (LDL-C)/HDL-C, non-high-density lipoprotein-cholesterol (non-HDL-C)/HDL-C, remnant cholesterol (RC)/HDL-C, apolipoprotein B (apo B)/apolipoprotein A-I (apo A-I), and apo B/HDL-C] were calculated. RESULTS: The TC/HDL-C, LDL-C/HDL-C, RC/HDL-C, non-HDL-C/HDL-C, apo B/HDL-C and apo B/apo A-I ratios (all P < 0.05) were significantly associated with ICAS but not with extracranial atherosclerotic stenosis after adjustment for confounding factors. Receiver operating characteristic (ROC) curves analysis revealed that the apo B/apo A-I ratio had the largest area under the ROC curve (AUC) among lipid levels alone and for lipid ratios (AUC = 0.588). Lipid ratios had higher AUC values than those for lipid levels alone for the identification of ICAS. CONCLUSION: The TC/HDL-C, LDL-C/HDL-C, RC/HDL-C, non-HDL-C/HDL-C apo B/HDL-C, and apo B/apo A-I ratios were significantly related to ICAS risk. Compared with the other variables tested, the apo B/apo A-I ratio appeared to be a better discriminator for identifying ICAS risk in stroke patients.

Association between Extracranial Carotid Artery Tortuosity and Clinical Outcomes in Anterior Circulation Acute Ischemic Stroke without Undergoing Endovascular Treatment.

BACKGROUND: Extracranial carotid artery (ECA) tortuosity may influences successful recanalization rates of mechanical thrombectomy in acute ischemic stroke (AIS), yet the relationship between ECA tortuosity and the prognosis of patients with anterior circulation AIS who cannot undergo endovascular treatment remains uncertain. We hypothesized that increased tortuosity of the ECA leads to unfavorable outcomes in such patients. METHODS: Patients with anterior circulation AIS who underwent computed tomography angiography of the head and neck in our hospital between March 2018 and November 2018 were retrospectively analyzed. The tortuosity of the bilateral ECA was measured, and functional outcomes were evaluated by a modified Rankin Scale (mRS) at 90 days. Multivariate logistic regression models were used to determine the association between ECA tortuosity and outcomes of patients. RESULTS: A total of 203 patients were enrolled in our study, including 140 patients (68.97%) with favorable outcomes (mRS, 0-2) and 63 patients (31.03%) with unfavorable outcomes (mRS, 3-6). After adjusting for age, atrial fibrillation, stroke territory, and posthospital antithrombotics/statins therapy in multivariate logistic regression model I, ECA tortuosity (odds ratio, 1.052; 95% confidence interval, 1.010-1.096; P = .015) was an independent risk of unfavorable outcomes in enrolled patients. In the other 2 models (II and III) which adjusted for age, sex, baseline National Institutes of Health Stroke Scale score, and with or without posthospital medication, ECA tortuosity was also showed independent relationship to unfavorable outcomes. The optimal cutoff was 12.5 to predict the unfavorable outcomes in a receiver operating characteristic curve. CONCLUSIONS: Our study demonstrated that the ECA tortuosity is an independent predictor of unfavorable outcomes for anterior circulation AIS patients who without undergoing endovascular treatment after hospital admission. ECA tortuosity values greater than 12.5 may indicate an unfavorable outcome.

Expansion-Prone Hematoma: Defining a Population at High Risk of Hematoma Growth and Poor Outcome.

BACKGROUND: Noncontrast computed tomography (CT) markers are increasingly used for predicting hematoma expansion. The aim of our study was to investigate the predictive value of expansion-prone hematoma in predicting hematoma expansion and outcome in patients with intracerebral hemorrhage (ICH). METHODS: Between July 2011 and January 2017, ICH patients who underwent baseline CT scan within 6 h of symptoms onset and follow-up CT scan were recruited into the study. Expansion-prone hematoma was defined as the presence of one or more of the following imaging markers: blend sign, black hole sign, or island sign. The diagnostic performance of blend sign, black hole sign, island sign, and expansion-prone hematoma in predicting hematoma expansion was assessed. Predictors of hematoma growth and poor outcome were analyzed using multivariable logistical regression analysis. RESULTS: A total of 282 patients were included in our final analysis. Of 88 patients with early hematoma growth, 69 (78.4%) had expansion-prone hematoma. Expansion-prone hematoma had a higher sensitivity and accuracy for predicting hematoma expansion and poor outcome when compared with any single imaging marker. After adjustment for potential confounders, expansion-prone hematoma independently predicted hematoma expansion (OR 28.33; 95% CI 12.95-61.98) and poor outcome (OR 5.67; 95% CI 2.82-11.40) in multivariable logistic model. CONCLUSION: Expansion-prone hematoma seems to be a better predictor than any single noncontrast CT marker for predicting hematoma expansion and poor outcome. Considering the high risk of hematoma expansion in these patients, expansion-prone hematoma may be a potential therapeutic target for anti-expansion treatment in future clinical studies.

Automatic recognition of white blood cell images with memory efficient superpixel metric GNN: SMGNN.

An automatic recognizing system of white blood cells can assist hematologists in the diagnosis of many diseases, where accuracy and efficiency are paramount for computer-based systems. In this paper, we presented a new image processing system to recognize the five types of white blood cells in peripheral blood with marked improvement in efficiency when juxtaposed against mainstream methods. The prevailing deep learning segmentation solutions often utilize millions of parameters to extract high-level image features and neglect the incorporation of prior domain knowledge, which consequently consumes substantial computational resources and increases the risk of overfitting, especially when limited medical image samples are available for training. To address these challenges, we proposed a novel memory-efficient strategy that exploits graph structures derived from the images. Specifically, we introduced a lightweight superpixel-based graph neural network (GNN) and broke new ground by introducing superpixel metric learning to segment nucleus and cytoplasm. Remarkably, our proposed segmentation model superpixel metric graph neural network (SMGNN) achieved state of the art segmentation performance while utilizing at most 10000$ \times $ less than the parameters compared to existing approaches. The subsequent segmentation-based cell type classification processes showed satisfactory results that such automatic recognizing algorithms are accurate and efficient to execeute in hematological laboratories. Our code is publicly available at https://github.com/jyh6681/SPXL-GNN.

RandStainNA++: Enhance Random Stain Augmentation and Normalization Through Foreground and Background Differentiation.

The wide prevalence of staining variations in digital pathology presents a significant obstacle, often undermining the effectiveness of diagnosis and analysis. The current strategies to counteract this issue primarily revolve around Stain Normalization (SN) and Stain Augmentation (SA). Nonetheless, these methodologies come with inherent limitations. They struggle to adapt to the vast array of staining styles, tend to presuppose linear associations between color spaces, and often lead to unrealistic color transformations. In response to these challenges, we introduce RandStainNA++, a novel method seamlessly integrating SN and SA. This method exploits the versatility of random SN and SA within randomly selected color spaces, effectively managing variations for the foreground and background independently. By refining the transformations of staining styles for the foreground and background within a realistic scope, this strategy promotes the generation of more practical staining transformations during the training phase. Further enhancing our approach, we propose a unique self-distillation method. This technique incorporates prior knowledge of stain variation, substantially augmenting the generalization capability of the network. The striking results yield that, compared to conventional classification models, our method boosts performance by a significant margin of 16-25%. Furthermore, when juxtaposed with baseline segmentation models, the Dice score registers an increase of 0.06.

Scalable Swin Transformer network for brain tumor segmentation from incomplete MRI modalities.

BACKGROUND: Deep learning methods have shown great potential in processing multi-modal Magnetic Resonance Imaging (MRI) data, enabling improved accuracy in brain tumor segmentation. However, the performance of these methods can suffer when dealing with incomplete modalities, which is a common issue in clinical practice. Existing solutions, such as missing modality synthesis, knowledge distillation, and architecture-based methods, suffer from drawbacks such as long training times, high model complexity, and poor scalability. METHOD: This paper proposes IMS2Trans, a novel lightweight scalable Swin Transformer network by utilizing a single encoder to extract latent feature maps from all available modalities. This unified feature extraction process enables efficient information sharing and fusion among the modalities, resulting in efficiency without compromising segmentation performance even in the presence of missing modalities. RESULTS: Two datasets, BraTS 2018 and BraTS 2020, containing incomplete modalities for brain tumor segmentation are evaluated against popular benchmarks. On the BraTS 2018 dataset, our model achieved higher average Dice similarity coefficient (DSC) scores for the whole tumor, tumor core, and enhancing tumor regions (86.57, 75.67, and 58.28, respectively), in comparison with a state-of-the-art model, i.e. mmFormer (86.45, 75.51, and 57.79, respectively). Similarly, on the BraTS 2020 dataset, our model scored higher DSC scores in these three brain tumor regions (87.33, 79.09, and 62.11, respectively) compared to mmFormer (86.17, 78.34, and 60.36, respectively). We also conducted a Wilcoxon test on the experimental results, and the generated p-value confirmed that our model's performance was statistically significant. Moreover, our model exhibits significantly reduced complexity with only 4.47 M parameters, 121.89G FLOPs, and a model size of 77.13 MB, whereas mmFormer comprises 34.96 M parameters, 265.79 G FLOPs, and a model size of 559.74 MB. These indicate our model, being light-weighted with significantly reduced parameters, is still able to achieve better performance than a state-of-the-art model. CONCLUSION: By leveraging a single encoder for processing the available modalities, IMS2Trans offers notable scalability advantages over methods that rely on multiple encoders. This streamlined approach eliminates the need for maintaining separate encoders for each modality, resulting in a lightweight and scalable network architecture. The source code of IMS2Trans and the associated weights are both publicly available at https://github.com/hudscomdz/IMS2Trans.

MoViT: Memorizing Vision Transformers for Medical Image Analysis.

The synergy of long-range dependencies from transformers and local representations of image content from convolutional neural networks (CNNs) has led to advanced architectures and increased performance for various medical image analysis tasks due to their complementary benefits. However, compared with CNNs, transformers require considerably more training data, due to a larger number of parameters and an absence of inductive bias. The need for increasingly large datasets continues to be problematic, particularly in the context of medical imaging, where both annotation efforts and data protection result in limited data availability. In this work, inspired by the human decision-making process of correlating new "evidence" with previously memorized "experience", we propose a Memorizing Vision Transformer (MoViT) to alleviate the need for large-scale datasets to successfully train and deploy transformer-based architectures. MoViT leverages an external memory structure to cache history attention snapshots during the training stage. To prevent overfitting, we incorporate an innovative memory update scheme, attention temporal moving average, to update the stored external memories with the historical moving average. For inference speedup, we design a prototypical attention learning method to distill the external memory into smaller representative subsets. We evaluate our method on a public histology image dataset and an in-house MRI dataset, demonstrating that MoViT applied to varied medical image analysis tasks, can outperform vanilla transformer models across varied data regimes, especially in cases where only a small amount of annotated data is available. More importantly, MoViT can reach a competitive performance of ViT with only 3.0% of the training data. In conclusion, MoViT provides a simple plug-in for transformer architectures which may contribute to reducing the training data needed to achieve acceptable models for a broad range of medical image analysis tasks.

Atrial lead perforation early after device implantation-a case report and literature review.

We report three patients with screw-in lead perforation in the right atrial free wall not long after device implantation. All the patients complained of intermittent stabbing chest pain associated with deep breathing during the implantation. The "dry" epicardial puncture was utilized to avoid hemopericardium during lead extraction in the first case. The atrial electrode was repositioned in all cases and replaced by a new passive fixation lead in two patients with resolution of the pneumothorax or pericardial effusion. A literature review of 50 reported cases of atrial lead perforation was added to the findings in our case report.

SERPINA3: A novel inflammatory biomarker associated with cerebral small vessel disease burden in ischemic stroke.

BACKGROUND AND OBJECTIVE: Inflammation has emerged as a prominent risk factor for cerebral small vessel disease (CSVD). However, the specific association between various inflammatory biomarkers and the development of CSVD remains unclear. Serine proteinase inhibitor A3 (SERPINA3), Matrix metalloproteinase-9 (MMP-9), Tissue inhibitor metalloproteinase-1 (TIMP-1), Monocyte Chemoattractant Protein-1 (MCP-1) are several inflammatory biomarkers that are potentially involved in the development of CSVD. In this present study, we aimed to investigate the relationship between candidate molecules and CSVD features. METHOD: The concentration of each biomarker was measured in 79 acute ischemic stroke patients admitted within 72 h after symptom onset. The associations between blood levels of inflammatory markers and CSVD score were investigated, as well as each CSVD feature, including white matter hyperintensities (WMH), lacunes, and enlarged perivascular spaces (EPVS). RESULTS: The mean age was 69.0 ± 11.8 years, and 65.8% of participants were male. Higher SERPINA3 level (>78.90 ng/mL) was significantly associated with larger WMH volume and higher scores on Fazekas's scale in all three models. Multiple regression analyses revealed the linear association between absolute WMH burden and SERPINA3 level, especially in model 3 (ß = 0.14; 95% confidence interval [CI], 0.04-0.24 ; p = 0.008 ). Restricted cubic spline regression demonstrated a dose-response relationship between SERPINA3 level and larger WMH volume (p nonlineariy = 0.0366 and 0.0378 in model 2 and mode 3, respectively). Using a receiving operating characteristic (ROC) curve, plasma SERPINA3 level of 64.15 ng/mL distinguished WMH >7.8 mL with the highest sensitivity and specificity (75.92% and 60%, respectively, area under curve [AUC] = 0.668, p = 0.0102). No statistically significant relationship has been found between other candidate biomarkers and CSVD features. CONCLUSION: In summary, among four inflammatory biomarkers that we investigated, SERPINA3 level at baseline was associated with WMH severity, which revealed a novel biomarker for CSVD and validated its relationship with inflammation and endothelial dysfunction.

Age-related changes after intracerebral hemorrhage: a comparative proteomics analysis of perihematomal tissue.

The risk factors and causes of intracerebral hemorrhage (ICH) and the degree of functional recovery after ICH are distinct between young and elderly patients. The increasing incidence of ICH in young adults has become a concern; however, research on the molecules and pathways involved ICH in subjects of different ages is lacking. In this study, tandem mass tag (TMT)-based proteomics was utilized to examine the protein expression profiles of perihematomal tissue from young and aged mice 24 h after collagenase-induced ICH. Among the 5,129 quantified proteins, ICH induced 108 and 143 differentially expressed proteins (DEPs) in young and aged mice, respectively; specifically, there were 54 common DEPs, 54 unique DEPs in young mice and 89 unique DEPs in aged mice. In contrast, aging altered the expression of 58 proteins in the brain, resulting in 39 upregulated DEPs and 19 downregulated DEPs. Bioinformatics analysis indicated that ICH activated different proteins in complement pathways, coagulation cascades, the acute phase response, and the iron homeostasis signaling pathway in mice of both age groups. Protein-protein interaction (PPI) analysis and ingenuity pathway analysis (IPA) demonstrated that the unique DEPs in the young and aged mice were related to lipid metabolism and carbohydrate metabolism, respectively. Deeper paired-comparison analysis demonstrated that apolipoprotein M exhibited the most significant change in expression as a result of both aging and ICH. These results help illustrate age-related protein expression changes in the acute phase of ICH.

A Federated Learning System for Histopathology Image Analysis With an Orchestral Stain-Normalization GAN.

Currently, data-driven based machine learning is considered one of the best choices in clinical pathology analysis, and its success is subject to the sufficiency of digitized slides, particularly those with deep annotations. Although centralized training on a large data set may be more reliable and more generalized, the slides to the examination are more often than not collected from many distributed medical institutes. This brings its own challenges, and the most important is the assurance of privacy and security of incoming data samples. In the discipline of histopathology image, the universal stain-variation issue adds to the difficulty of an automatic system as different clinical institutions provide distinct stain styles. To address these two important challenges in AI-based histopathology diagnoses, this work proposes a novel conditional Generative Adversarial Network (GAN) with one orchestration generator and multiple distributed discriminators, to cope with multiple-client based stain-style normalization. Implemented within a Federated Learning (FL) paradigm, this framework well preserves data privacy and security. Additionally, the training consistency and stability of the distributed system are further enhanced by a novel temporal self-distillation regularization scheme. Empirically, on large cohorts of histopathology datasets as a benchmark, the proposed model matches the performance of conventional centralized learning very closely. It also outperforms state-of-the-art stain-style transfer methods on the downstream Federated Learning image classification task, with an accuracy increase of over 20.0% in comparison to the baseline classification model.

Mine local homogeneous representation by interaction information clustering with unsupervised learning in histopathology images.

BACKGROUND AND OBJECTIVE: The success of data-driven deep learning for histopathology images often depends on high-quality training sets and fine-grained annotations. However, as tumors are heterogeneous and annotations are expensive, unsupervised learning approaches are desirable to obtain full automation. METHODS: In this paper, an Interaction Information Clustering (IIC) method is proposed to extract locally homogeneous features in mutually exclusive clusters. Trained in an unsupervised paradigm, the framework learns invariant information from multiple spatially adjacent regions for improved classification. Additionally, an adaptive Conditional Random Field (CRF) model is incorporated to detect spatially adjacent image patches of high morphological homogeneity in an offset-constraint free manner. RESULTS: Empirically, the proposed model achieves an observable improvement of 11.4% on the downstream patch-level classification accuracy, compared with state-of-the-art unsupervised learning approaches. CONCLUSION: Furthermore, evaluated with our clinically collected histopathology whole-slide images, the proposed model shows high consistency in tissue distribution compared with well-trained supervised learning, which is of important diagnostic significance in clinical practice.