Identifying pathological slices of gastric cancer via deep learning.

BACKGROUND: The accuracy of histopathology diagnosis largely depends on the pathologist's experience. It usually takes over 10 years to cultivate a senior pathologist, and small numbers of them lead to a high workload for those available. Meanwhile, inconsistent diagnostic results may arise among different pathologists, especially in complex cases, because diagnosis based on morphology is subjective. Computerized analysis based on deep learning has shown potential benefits as a diagnostic strategy. METHODS: This research aims to automatically determine the location of gastric cancer (GC) in the images of GC slices through artificial intelligence. We use image data from a regional teaching hospital in Taiwan for training. We collect images of patients diagnosed with GC from January 1, 2019 to December 31, 2020. In this study, scanned images are used to dissect 13,600 images from 50 different patients with GC sections whose GC sections are stained with hematoxylin and eosin (H&E stained) through a whole slide scanner, the scanned images from 50 different GC slice patients are divided into 80% training combinations, 2200 images of 40 patients are trained. The remaining 20%, totaling 10 people, are validated from a test set of 550 images. RESULTS: The validation results show that 91% of the correct rates are interpreted as GC images through deep learning. The sensitivity, specificity, PPV, and NPV were 84.9%, 94%, 87.7%, and 92.5%, respectively. After creating a 3D model through the grayscale value, the position of the GC is completely marked by the 3D model. The purpose of this research is to use artificial intelligence (AI) to determine the location of the GC in the image of GC slice. CONCLUSION: In patients undergoing pancreatectomy for pancreatic cancer, intraoperative infusion of lidocaine did not improve overall or disease-free survival. Reduced formation of circulating NETs was absent in pancreatic tumour tissue. CONCLUSION: For AI to assist pathologists in daily practice, to help a pathologist making a definite diagnosis is not the prime purpose at present time. The benefits could come from cancer screening and double-check quality control in a heavy workload which could distract the attention of pathologist during the time constraint examination process. We propose a two-steps method to identify cancerous areas in endoscopic gastric biopsy slices via deep learning. Then a 3D model is used to further mark all the positions of GC in the picture, and the model overcomes the problem that deep learning can't catch all GC.

Factors Associated with the Refusal of Direct-Acting Antiviral Agents for the Treatment of Hepatitis C in Taiwan.

Background and Objectives: Direct-acting antiviral agents (DAA) are a safe and highly effective treatment for hepatitis C virus (HCV) infection. However, the uptake of DAA treatment remains a challenge. This study aims to examine the reasons for DAA refusal among HCV patients covered by the Taiwan National Health Insurance system. Materials and Methods: This retrospective observational study covered the period from January 2009 to December 2019 and was conducted at a single hepatitis treatment center in Taiwan. This study involved chart reviews and phone-based surveys to confirm treatment status and refusal causes. To confirm treatment status, subjects with HCV without treatment records were phone-contacted to confirm treatment status. Patients who did not receive treatment were invited back for treatment. If the patient refused, the reason for refusal was discussed. Results: A total of 3566 patients were confirmed with DAA treatment; 418 patients (179 patients who were lost to contact or refused the survey and 239 patients who completed the survey of DAA refusal) were included in the no-DAA-therapy group. Factors associated with receiving DAAs were hemoglobin levels, hepatitis B virus co-infection, and regular gastroenterology visits. Meanwhile, male sex, platelet levels, and primary care physician visits were associated with DAA refusal. The leading causes of treatment refusal were multiple comorbidities, low health literacy, restricted access to hospitals, nursing home residence, and old age. The rate of DAA refusal remains high (10%). Conclusions: The reasons for treatment refusal are multifactorial, and addressing them requires complex interventions.

Analysis of the effects of cyproheptadine on bladder cancer through big data.

This study was designed to compare the efficacy of Cyproheptadine (CY) in patients with bladder cancer (BC) who received different therapeutic modalities. We used the database from a hospital in Taiwan for analysis. We included patients diagnosed as having bladder cancer from January 1, 2008, to December 31, 2017. The patient cohort comprised those who received different treatments, and we compared patients who received CY with those who did not. In total, 627 patients were included, and the mean follow-up duration was 3.26 years. All data were filtered out by 230 million data and 119 patients had used CY. Among them, 32 patients were used over 3 months of CY. The CY treatment curve shown by Kaplan-Meier survival curves for patients treated is higher than that of the non-CY effect. The value of Chi-squared statistic was 4.138 with associated p-value less than 0.05. Two survival curves shown by the result of the log rank test differ significantly. The grouping variable different treatments for non-CY and CY has a significant influence on survival rate. These results suggest that the use of CY may improve the survival rate of patients with BC.