Safety and clinical efficacy of immune checkpoint inhibitors in advanced gastric cancer in the real world.

BACKGROUND: To evaluate the clinical efficacy and safety of immune checkpoint inhibitors in patients with advanced gastric cancer in the real world. METHODS: The retrospective analysis was conducted on the clinical records of 402 patients with advanced gastric cancer who were admitted to the Nanjing Drum Tower Hospital between December 2017 and April 2022 and who had received immunotherapy. Observation target: drug use, treatment, adverse reaction type and grade, objective response rate (ORR), disease control rate (DCR), progression free survival (PFS), and overall survival (OS). RESULTS: By retrospectively analyzing the data of patients with advanced gastric cancer treated with ICIs previously admitted to our medical center, we found some clinical characteristic factors associated with the occurrence of irAEs as well as the efficacy and prognosis: the presence or absence of hypertension, whether or not to receive targeted therapies can predict the occurrence of immune-related adverse events (irAEs), and the more the presence of irAEs, the better the prognosis. These can help clinicians in clinical drug selection. CONCLUSIONS: The results of this paper show that the occurrence of irAEs is associated with patients' OS. irAEs occurrence can prolong patients' OS. irAEs occurrence may serve as a surrogate marker for ICIs.

A multi-task fusion model based on a residual-Multi-layer perceptron network for mammographic breast cancer screening.

BACKGROUND AND OBJECTIVE: Deep learning approaches are being increasingly applied for medical computer-aided diagnosis (CAD). However, these methods generally target only specific image-processing tasks, such as lesion segmentation or benign state prediction. For the breast cancer screening task, single feature extraction models are generally used, which directly extract only those potential features from the input mammogram that are relevant to the target task. This can lead to the neglect of other important morphological features of the lesion as well as other auxiliary information from the internal breast tissue. To obtain more comprehensive and objective diagnostic results, in this study, we developed a multi-task fusion model that combines multiple specific tasks for CAD of mammograms. METHODS: We first trained a set of separate, task-specific models, including a density classification model, a mass segmentation model, and a lesion benignity-malignancy classification model, and then developed a multi-task fusion model that incorporates all of the mammographic features from these different tasks to yield comprehensive and refined prediction results for breast cancer diagnosis. RESULTS: The experimental results showed that our proposed multi-task fusion model outperformed other related state-of-the-art models in both breast cancer screening tasks in the publicly available datasets CBIS-DDSM and INbreast, achieving a competitive screening performance with area-under-the-curve scores of 0.92 and 0.95, respectively. CONCLUSIONS: Our model not only allows an overall assessment of lesion types in mammography but also provides intermediate results related to radiological features and potential cancer risk factors, indicating its potential to offer comprehensive workflow support to radiologists.

Triplet-branch network with contrastive prior-knowledge embedding for disease grading.

Since different disease grades require different treatments from physicians, i.e., the low-grade patients may recover with follow-up observations whereas the high-grade may need immediate surgery, the accuracy of disease grading is pivotal in clinical practice. In this paper, we propose a Triplet-Branch Network with ContRastive priOr-knoWledge embeddiNg (TBN-CROWN) for the accurate disease grading, which enables physicians to accordingly take appropriate treatments. Specifically, our TBN-CROWN has three branches, which are implemented for representation learning, classifier learning and grade-related prior-knowledge learning, respectively. The former two branches deal with the issue of class-imbalanced training samples, while the latter one embeds the grade-related prior-knowledge via a novel auxiliary module, termed contrastive embedding module. The proposed auxiliary module takes the features embedded by different branches as input, and accordingly constructs positive and negative embeddings for the model to deploy grade-related prior-knowledge via contrastive learning. Extensive experiments on our private and two publicly available disease grading datasets show that our TBN-CROWN can effectively tackle the class-imbalance problem and yield a satisfactory grading accuracy for various diseases, such as fatigue fracture, ulcerative colitis, and diabetic retinopathy.

Case report: A case of advanced gastric cancer with multiple skin metastases, with significant relief from immunotherapy.

Gastric cancer is the fifth leading cause of cancer-related mortality worldwide, with a low 5-year survival rate in advanced stages. Cutaneous metastasis is rare in gastric cancer, with only 0.8-1% incidence. We reported a rare case of female gastric cancer. The patient had undergone subtotal gastrectomy and chemotherapy 13 years ago, followed by a subsequent surgery of residual stomach, partial jejunum, and partial colon resection 11 years later. The pathological examination revealed poorly differentiated stomach adenocarcinoma, Lauren classification: diffuse type. The patient received 2 cycles of SOX chemotherapy. Two years later, cauliflower-like skin nodules, which were surgically excised, appeared on the back. The histopathological examination showed a spindle cell tumor; no specific anti-tumor treatment was administered. Six months later, the skin lesions increased in size and number, spreading to the neck, chest, and abdomen, presenting as erythematous patches with some cauliflower-like elevations. A skin biopsy of a 1cm0.5cm0.3cm lesion on the left abdomen was performed, and based on the immunohistochemistry, clinical history, and the possibility of metastatic or infiltrating adenocarcinoma, the gastrointestinal origin was highly suspected. Genetic testing was performed on the gastric recurrence and skin lesions, revealing 103 shared genetic variations, further suggesting the skin metastasis originated from gastric cancer. Subsequently, the patient received 10 cycles of immunotherapy combined with intravenous chemotherapy (200mg Tislelizumab and 100mg albumin-bound paclitaxel). The treatment response was evaluated as partial remission, with significant improvement in the skin lesions compared to before. This case highlights the possibility of tumor metastasis in patients with extensive skin lesions in advanced gastric cancer. Early examination, diagnosis, skin biopsy, immunohistochemistry, and genetic sequencing are recommended.

Quantitative Modeling of Stemness in Single-Cell RNA Sequencing Data: A Nonlinear One-Class Support Vector Machine Method.

Intratumoral heterogeneity and the presence of cancer stem cells are challenging issues in cancer therapy. An appropriate quantification of the stemness of individual cells for assessing the potential for self-renewal and differentiation from the cell of origin can define a measurement for quantifying different cell states, which is important in understanding the dynamics of cancer evolution, and might further provide possible targeted therapies aimed at tumor stem cells. Nevertheless, it is usually difficult to quantify the stemness of a cell based on molecular information associated with the cell. In this study, we proposed a stemness definition method with one-class Hadamard kernel support vector machine (OCHSVM) based on single-cell RNA sequencing (scRNA-seq) data. Applications of the proposed OCHSVM stemness are assessed by various data sets, including preimplantation embryo cells, induced pluripotent stem cells, or tumor cells. We further compared the OCHSVM model with state-of-the-art methods CytoTRACE, one-class logistic regression, or one-class SVM methods with different kernels. The computational results demonstrate that the OCHSVM method is more suitable for stemness identification using scRNA-seq data.

Epigenetic regulation of programmed cell death in hypoxia-induced pulmonary arterial hypertension.

Pulmonary arterial hypertension (PAH) is a severe progressive disease that may cause early right ventricular failure and eventual cardiac failure. The pathogenesis of PAH involves endothelial dysfunction, aberrant proliferation of pulmonary artery smooth muscle cells (PASMCs), and vascular fibrosis. Hypoxia has been shown to induce elevated secretion of vascular endothelial growth factor (VEGF), leading to the development of hypoxic PAH. However, the molecular mechanisms underlying hypoxic PAH remain incompletely understood. Programmed cell death (PCD) is a natural cell death and regulated by certain genes. Emerging evidence suggests that apoptotic resistance contributes to the development of PAH. Moreover, several novel types of PCD, such as autophagy, pyroptosis, and ferroptosis, have been reported to be involved in the development of PAH. Additionally, multiple diverse epigenetic mechanisms including RNA methylation, DNA methylation, histone modification, and the non-coding RNA molecule-mediated processes have been strongly linked to the development of PAH. These epigenetic modifications affect the expression of genes, which produce important changes in cellular biological processes, including PCD. Consequently, a better understanding of the PCD processes and epigenetic modification involved in PAH will provide novel, specific therapeutic strategies for diagnosis and treatment. In this review, we aim to discuss recent advances in epigenetic mechanisms and elucidate the role of epigenetic modifications in regulating PCD in hypoxia-induced PAH.

The adenosinergic machinery in cancer: In-tandem insights from basic mechanisms to therapy.

Extracellular adenosine (eADO) signaling has emerged as an increasingly important regulator of immune responses, including tumor immunity. eADO is mainly produced from extracellular ATP (eATP) hydrolysis. eATP is rapidly accumulated in the extracellular space following cell death or cellular stress triggered by hypoxia, nutrient starvation, or inflammation. eATP plays a pro-inflammatory role by binding and activating the P2 purinergic receptors (P2X and P2Y), while eADO has been reported in many studies to mediate immunosuppression by activating the P1 purinergic receptors (A1, A2A, A2B, and A3) in diverse immune cells. Consequently, the hydrolysis of eATP to eADO alters the immunosurveillance in the tumor microenvironment (TME) not only by reducing eATP levels but also by enhancing adenosine receptor signaling. The effects of both P1 and P2 purinergic receptors are not restricted to immune cells. Here we review the most up-to-date understanding of the tumor adenosinergic system in all cell types, including immune cells, tumor cells, and stromal cells in TME. The potential novel directions of future adenosinergic therapies in immuno-oncology will be discussed.

Efficacy of laser therapy for temporomandibular disorders: A systematic review and meta-analysis.

OBJECTIVE: The aim of this systematic review was to evaluate the efficacy of laser therapy in temporomandibular disorders (TMD). METHODS: Randomized controlled trials (RCTs) in regard to this issue were searched in electronic databases. Three investigators independently screened the eligible studies, and the quality of the included studies was assessed according to the risk of bias tool recommended by the Cochrane handbook. The primary outcome measure was the degree of pain, reported on a visual analog scale (VAS), and the secondary outcome measures were TMJ function, including maximum active vertical opening (MAVO), maximum passive vertical opening (MPVO), left and right lateral movement (LLE, RLE). Pooled effect sizes were calculated using random effects models and 95% confidence interval (95% CI). RESULTS: A total of 28 randomized controlled trials were included. Laser therapy had a more significant effect in terms of VAS (SMD=ï¹£1.88; 95% CI=ï¹£2.46 toï¹£1.30; P < 0.00001; I2 =93%), MAVO (MD = 4.90; 95% CI= 3.29-6.50; P < 0.00001; I2 =72%), MPVO (MD=5.82; 95% CI= 4.62-7.01; P < 0.00001; I2 =40%) and RLE (MD = 0.73; 95% CI= 0.23-1.22; P = 0.004; I2 = 0%) as compared to placebo group. However, there was no significant difference in LLE between two groups (MD= 0.35; 95% CI=ï¹£0.31-1.01; P = 0.30; I2 =0%). CONCLUSIONS: Laser therapy can effectively reduce pain but have small effect on improving mandibular movement of TMD patients. More well-designed RCTs with large sample sizes are needed for further validation. And these studies should report detailed laser parameters and provide complete outcome measure data.

Lipophagy: Molecular Mechanisms and Implications in Hepatic Lipid Metabolism.

The liver is the most significant metabolic organ in the body and plays an important role in lipid metabolism. Liver lipid metabolism disorders cause hepatic diseases such as hepatitis, hepatic cirrhosis, and hepatoma. Autophagy is a process of generating energy and building blocks by degrading redundant or damaged proteins and organelles. Thus, it helps in the maintenance of cellular homeostasis. Recent discoveries revealed that lipophagy plays a vital role in hepatic cellular homeostasis and lipid metabolism. Its imbalance is always associated with the perturbation of lipid metabolism in the liver. This article reviewed the molecular mechanisms involved in lipophagy and the interaction between lipophagy and hepatic lipid metabolism. Increasing evidence suggests that lipophagy is an effective method to resolve liver diseases.

Mitosis domain generalization in histopathology images - The MIDOG challenge.

The density of mitotic figures (MF) within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. Recognition of MF by pathologists is subject to a strong inter-rater bias, limiting its prognostic value. State-of-the-art deep learning methods can support experts but have been observed to strongly deteriorate when applied in a different clinical environment. The variability caused by using different whole slide scanners has been identified as one decisive component in the underlying domain shift. The goal of the MICCAI MIDOG 2021 challenge was the creation of scanner-agnostic MF detection algorithms. The challenge used a training set of 200 cases, split across four scanning systems. As test set, an additional 100 cases split across four scanning systems, including two previously unseen scanners, were provided. In this paper, we evaluate and compare the approaches that were submitted to the challenge and identify methodological factors contributing to better performance. The winning algorithm yielded an F1 score of 0.748 (CI95: 0.704-0.781), exceeding the performance of six experts on the same task.

DigestPath: A benchmark dataset with challenge review for the pathological detection and segmentation of digestive-system.

Examination of pathological images is the golden standard for diagnosing and screening many kinds of cancers. Multiple datasets, benchmarks, and challenges have been released in recent years, resulting in significant improvements in computer-aided diagnosis (CAD) of related diseases. However, few existing works focus on the digestive system. We released two well-annotated benchmark datasets and organized challenges for the digestive-system pathological cell detection and tissue segmentation, in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI). This paper first introduces the two released datasets, i.e., signet ring cell detection and colonoscopy tissue segmentation, with the descriptions of data collection, annotation, and potential uses. We also report the set-up, evaluation metrics, and top-performing methods and results of two challenge tasks for cell detection and tissue segmentation. In particular, the challenge received 234 effective submissions from 32 participating teams, where top-performing teams developed advancing approaches and tools for the CAD of digestive pathology. To the best of our knowledge, these are the first released publicly available datasets with corresponding challenges for the digestive-system pathological detection and segmentation. The related datasets and results provide new opportunities for the research and application of digestive pathology.

Estrogen and BRCA1 deficiency synergistically induce breast cancer mutation-related DNA damage.

Estrogen (E2) is crucial for the development of breast cancer caused by BRCA1 mutation, and can increase the DNA damage in BRCA1-deficient cells. However, the mechanisms through which BRCA1 deficiency and E2 synergistically induce DNA damage remains unclear. In this study, we analyzed the distribution of DNA damage in E2-treated BRCA1-deficient cells. We detected DNA lesions in the vicinity of genes that are transcriptionally activated by estrogen receptor-α (ER). Loss of BRCA1 altered chromatin binding by ER, which significantly affected the distribution of DNA damage. Moreover, these changes were associated with the established mutations in BRCA1-mutant breast cancer. Taken together, our findings reveal a new mechanism underlying the DNA damage in breast cancer cells that is synergistically induced by BRCA1 deficiency and E2.

Mechanistic insights into catalysis of in-situ iron on pyrolysis of waste printed circuit boards: Comparative study of kinetics, products, and reaction mechanism.

Pyrolysis is a potential recovery method for waste printed circuit boards (WPCBs), but the organic brominated products are toxic and hazardous. Iron has been used as a catalyst for debromination from pyrolysis oil, but the effects of in-situ iron on WPCBs pyrolysis is still not well understood. Herein, the pyrolysis mechanism for laminates of PCBs in the absence and presence of iron was studied via analyzing pyrolysis characteristics, kinetics, and products. According to the thermogravimetry experiments, pyrolysis of all samples could be divided into four stages, and iron accelerated the pyrolysis reaction by decreasing the activation energy as calculated using the Starink method. Volatiles released during the heating process were continuously determined by TG-FTIR-MS, and products generated at different pyrolysis temperatures were collected and characterized. The obtained results exhibited that iron promoted the generation of gaseous products and facilitated the conversion of organic bromides to inorganic bromides. Therefore, retaining iron was beneficial to energy saving and environmental protection for WPCBs pyrolysis. In addition, the decomposition mechanisms of brominated epoxy resin with and without iron were proposed. This work would contribute to the improvement and application of WPCBs pyrolysis technology.

L-Cysteine functionalized graphene oxide nanoarchitectonics: A metal-free Hg2+ nanosensor with peroxidase-like activity boosted by competitive adsorption.

Developing non-noble metal, even metal-free chemical sensors for the detection of toxic heavy metal ions is significantly desirable for economically and environmentally sustainable application but has heretofore remained elusive. Herein, a L-cysteine functionalized graphene oxide nanosheet (CGO) nanoarchitectonics, greenly synthesized by a very simple method at room temperature, was utilized to realize the simultaneous enrichment and colorimetric detection of trace mercury ions (Hg2+). It was discovered that CGO, as a nanozyme mimic exhibited greatly enhanced peroxidase-like catalytic activity than the pristine graphene oxide. By exploring the interactions of CGO nanozyme with colorimetric substrate, 3,3',5,5'-tetramethylbenzidine (TMB) and target Hg2+ ions, we found that the sensing principle was based mainly on the competitive adsorption between Hg2+ ions and TMB over CGO. The pre-capture of Hg2+ ions hindered the TMB binding on CGO, resulting in the promoted oxidation of TMB by H2O2 to produce more colored oxidation products, from which the colorimetric sensing of Hg2+ was realized with a good detection effect on 5 µg L-1 solution. As an enrichment-sensing integration platform, this metal-free sensor is cost-effective and sensitive, and presents considerable anti-interference ability over other metal ions. Overall, this work not only expands the application of graphene-based materials in colorimetric detection but also provides a general sensing principle to construct highly sensitive sensors.

Honokiol attenuates lipotoxicity in hepatocytes via activating SIRT3-AMPK mediated lipophagy.

BACKGROUND: Non-alcoholic fatty liver disease (NAFLD) is characterized by ectopic accumulation of triglycerides in the liver. Emerging evidence has demonstrated that lipophagy regulates lipid mobilization and energy homeostasis in the liver. Sirtuin 3 (SIRT3), a mitochondrial NAD+-dependent deacetylase, modulates the activities of several substrates involving in autophagy and energy metabolism. Honokiol (HK) is a natural lignan from the plants of Magnolia genus that exhibits potent liver protective property. METHODS: AML12 was challenged with 500 µM palmitic acid and 250 µM oleic acid mixture solution to induce lipotoxicity. C57BL/6J mice were fed with a choline-deficient high fat diet (CDHFD) to generate liver steatosis. The expression of autophagy-related and AMP-activated protein kinase (AMPK) pathway proteins was evaluated by Western blotting and immunofluorescence staining. Intracellular lipid accumulation was validated by Nile red staining. Molecular docking analysis was performed on AutoDock 4.2. RESULTS: HK (5 and 10 µM) was found to attenuate lipid accumulation through promoting SIRT3-AMPK-mediated autophagy, mainly on lipid droplets. HK had hydrophobic interaction with amino acid residues (PHE294, GLU323 and VAL324) and NAD+. Moreover, HK improved mitochondrial function to enhance lipolysis, through decreasing the acetylated long-chain acyl-CoA dehydrogenase level. In CDHFD-fed mice, HK (2.5 and 10 mg/Kg) treatment obviously prevented lipid accumulation in the liver. And co-treatment of the AMPK inhibitor, Compound C, almost abolished the above changes. CONCLUSIONS: These results suggest that HK could ameliorate lipotoxicity in hepatocytes by activating SIRT3-AMPK-lipophagy axis, which might be a potential therapeutic agent against NAFLD.

Automated tumor proportion score analysis for PD-L1 (22C3) expression in lung squamous cell carcinoma.

Programmed cell death ligend-1 (PD-L1) expression by immunohistochemistry (IHC) assays is a predictive marker of anti-PD-1/PD-L1 therapy response. With the popularity of anti-PD-1/PD-L1 inhibitor drugs, quantitative assessment of PD-L1 expression becomes a new labor for pathologists. Manually counting the PD-L1 positive stained tumor cells is an obviously subjective and time-consuming process. In this paper, we developed a new computer aided Automated Tumor Proportion Scoring System (ATPSS) to determine the comparability of image analysis with pathologist scores. A three-stage process was performed using both image processing and deep learning techniques to mimic the actual diagnostic flow of the pathologists. We conducted a multi-reader multi-case study to evaluate the agreement between pathologists and ATPSS. Fifty-one surgically resected lung squamous cell carcinoma were prepared and stained using the Dako PD-L1 (22C3) assay, and six pathologists with different experience levels were involved in this study. The TPS predicted by the proposed model had high and statistically significant correlation with sub-specialty pathologists' scores with Mean Absolute Error (MAE) of 8.65 (95% confidence interval (CI): 6.42-10.90) and Pearson Correlation Coefficient (PCC) of 0.9436 ([Formula: see text]), and the performance on PD-L1 positive cases achieved by our method surpassed that of non-subspecialty and trainee pathologists. Those experimental results indicate that the proposed automated system can be a powerful tool to improve the PD-L1 TPS assessment of pathologists.

[Preparation and Evaluation of Intranasal in situ Gel of Bupleuri Radix Volatile Oil and Baicalin].

OBJECTIVE: To prepare and evaluate a new formulation of thermosensitive and ion-sensitive in situ gel for nasal administration, using the volatile oil of Bupleuri radix and baicalin, the effective component extracted from Scutellariae radix . METHODS: Formulation of in situ nasal gel of Bupleuri radix volatile oil and baicalin was prepared by using poloxamer 407 and deacetylated gellan gum as the gel base, 10% pharmasolve and 2% polysorbate 80 as the solubilizer, and 0.8% triethanolamine as the pH regulator. The physical appearance, phase transition temperature, and baicalin release performance of the prepared gel were examined. The pharmacodynamic evaluation was done with the rat fever model developed with dry yeast and the mouse auricle swelling inflammation model. RESULTS: The phase transition temperature of the gel was optimized to be 36 ℃. The release of baicalin from the gel showed obvious features of sustained release, which accorded well the zero-order kinetics equation. The results of experiments with the rat dry yeast fever model and the mouse xylene auricle swelling inflammation model showed that the gel had significant antipyretic and anti-inflammatory effects that were significantly better than those of the groups treated with the blank gel base and the Bupleuri radix and Scutellariae radix granule. Results from the cilia toxicity test showed that the gel did not have obvious toxic effect on toad palate mucosal cilia. CONCLUSION: The in situ nasal gel of Bupleuri radix volatile oil and baicalin prepared in the study had a rapid onset time, high efficiency, and prolonged release of active ingredients, thus showing promises for further applicational development.

Antitumor effects of helenalin in doxorubicin-resistant leukemia cells are mediated via mitochondrial mediated apoptosis, loss of mitochondrial membrane potential, inhibition of cell migration and invasion and downregulation of PI3-kinase/AKT/m-TOR signalling pathway.

Retraction of: 'Antitumor effects of helenalin in doxorubicin-resistant leukemia cells are mediated via mitochondrial mediated apoptosis, loss of mitochondrial membrane potential, inhibition of cell migration and invasion and downregulation of PI3-kinase/AKT/m-TOR signalling pathway', by Jingxin Liu, Yanan Zhao, Zhangzhen Shi, Yuansong Bai, JBUON 2019;24(5):2068-2074; PMID:31786877. Following the publication of the above article, readers drew to our attention that part of the data was unreliable. The authors were requested to provide the raw data to prove the originality, but were unable to do so. After an investigation, the Editors of JBUON decided to retract this article. We thank the readers for bringing this matter to our attention. We apologize for any inconvenience it may cause.

Core transcription regulatory circuitry orchestrates corneal epithelial homeostasis.

Adult stem cell identity, plasticity, and homeostasis are precisely orchestrated by lineage-restricted epigenetic and transcriptional regulatory networks. Here, by integrating super-enhancer and chromatin accessibility landscapes, we delineate core transcription regulatory circuitries (CRCs) of limbal stem/progenitor cells (LSCs) and find that RUNX1 and SMAD3 are required for maintenance of corneal epithelial identity and homeostasis. RUNX1 or SMAD3 depletion inhibits PAX6 and induces LSCs to differentiate into epidermal-like epithelial cells. RUNX1, PAX6, and SMAD3 (RPS) interact with each other and synergistically establish a CRC to govern the lineage-specific cis-regulatory atlas. Moreover, RUNX1 shapes LSC chromatin architecture via modulating H3K27ac deposition. Disturbance of RPS cooperation results in cell identity switching and dysfunction of the corneal epithelium, which is strongly linked to various human corneal diseases. Our work highlights CRC TF cooperativity for establishment of stem cell identity and lineage commitment, and provides comprehensive regulatory principles for human stratified epithelial homeostasis and pathogenesis.

GCSBA-Net: Gabor-Based and Cascade Squeeze Bi-Attention Network for Gland Segmentation.

Colorectal cancer is the second and the third most common cancer in women and men, respectively. Pathological diagnosis is the "gold standard" for tumor diagnosis. Accurate segmentation of glands from tissue images is a crucial step in assisting pathologists in their diagnosis. The typical methods for gland segmentation form a dense image representation, ignoring its texture and multi-scale attention information. Therefore, we utilize a Gabor-based module to extract texture information at different scales and directions in histopathology images. This paper also designs a Cascade Squeeze Bi-Attention (CSBA) module. Specifically, we add Atrous Cascade Spatial Pyramid (ACSP), Squeeze Position Attention (SPA) module and Squeeze Channel Attention module (SCA) to model semantic correlation and maintain the multi-level aggregation on the spatial pyramid with different dilations. Besides, to solve the imbalance of data distribution and boundary blur, we propose a hybrid loss function to response the object boudary better. The experimental results show that the proposed method achieves state-of-the-art performance on the GlaS challenge dataset and CRAG colorectal adenocarcinoma dataset, respectively.