SSSNet: Small-Scale-Aware Siamese Network for Gastric Cancer Detection

2019 ◽  
Author(s):  
Chih-Chung Hsu ◽  
Hsin-Ti Ma ◽  
Jun-Yi Lee
Keyword(s):  
Gastric Cancer ◽  
Cancer Detection ◽  
Small Scale ◽  
Siamese Network

scholarly journals Sharing With Patients the Uncertainties Regarding the Management of Dyspepsia

2021 ◽  
Vol 8 ◽  
Author(s):  
Jochanan Benbassat
Keyword(s):  
Gastric Cancer ◽  
Patient Satisfaction ◽  
High Risk ◽  
Life Expectancy ◽  
Early Diagnosis ◽  
Cancer Detection ◽  
Symptom Relief ◽  
Management Strategies ◽  
Detection Rates ◽  
Test And Treat

Background: The management of patients with dyspepsia is uncertain. Some authors advocate endoscopy for all; others restrict endoscopy only to patients at high risk of gastric cancer, namely to those above an age threshold, or with a family history, dysphagia, loss of weight, anemia, or a childhood in Asian countries. Still others recommend various combinations between test-and-treat for Helicobacter pylori, anti-secretory treatment, and/or endoscopy.Objective: To highlight the uncertainties in the choice between the various strategies and argue that these uncertainties should be shared with the patient.Method: An overview of reported life expectancy, patient satisfaction, gastric cancer detection rates, symptom relief, and cost effectiveness of the management strategies for dyspepsia.Main Findings: There are no randomized controlled trials of the effect of screening by endoscopy on mortality of patients with gastric cancer. Lower grades of evidence suggest that early diagnosis reduces this mortality. Analyses, which assume a survival benefit of early diagnosis, indicate that mass screening in countries of high incidence gastric cancer (> 10 cases per 100,000) and targeted screening of high-risk persons in countries of low-intermediate incidence (<10 cases per 100,000) is cost-effective at a willingness to pay of $20,000–50,000 per QALY. Prompt endoscopy appears to be best for patient satisfaction and gastric cancer detection, and test-and-treat for H pylori—for symptom relief and avoiding endoscopies.Conclusions: The gain in life expectancy is the main source of uncertainty in the choice between management strategies. This choice should be shared with the patients after explaining uncertainties and eliciting their preferences.

scholarly journals Familial gastric cancer: detection of a hereditary cause helps to understand its etiology

2012 ◽  
Vol 10 (1) ◽  
Author(s):  
Ingrid P Vogelaar ◽  
Rachel S van der Post ◽  
Tanya M Bisseling ◽  
J Han JM van Krieken ◽  
Marjolijn JL Ligtenberg ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Cancer Detection ◽  
Familial Gastric Cancer

scholarly journals Online Learned Siamese Network with Auto-Encoding Constraints for Robust Multi-Object Tracking

Electronics ◽  
2019 ◽  
Vol 8 (6) ◽  
pp. 595 ◽  
Author(s):  
Peixin Liu ◽  
Xiaofeng Li ◽  
Han Liu ◽  
Zhizhong Fu
Keyword(s):  
Deep Learning ◽  
Object Tracking ◽  
Greedy Algorithm ◽  
Small Scale ◽  
Challenging Problem ◽  
Complex Environments ◽  
Siamese Network ◽  
Sample Data ◽  
New Feature ◽  
Object Trajectories

Multi-object tracking aims to estimate the complete trajectories of objects in a scene. Distinguishing among objects efficiently and correctly in complex environments is a challenging problem. In this paper, a Siamese network with an auto-encoding constraint is proposed to extract discriminative features from detection responses in a tracking-by-detection framework. Different from recent deep learning methods, the simple two layers stacked auto-encoder structure enables the Siamese network to operate efficiently only with small-scale online sample data. The auto-encoding constraint reduces the possibility of overfitting during small-scale sample training. Then, the proposed Siamese network is improved to extract the previous-appearance-next vector from tracklet for better association. The new feature integrates the appearance, previous, and next stage motions of an element in a tracklet. With the new features, an online incremental learned tracking framework is established. It contains reliable tracklet generation, data association to generate complete object trajectories, and tracklet growth to deal with missing detections and to enhance the new feature for tracklet. Benefiting from discriminative features, the final trajectories of objects can be achieved by an efficient iterative greedy algorithm. Feature experiments show that the proposed Siamese network has advantages in terms of both discrimination and correctness. The system experiments show the improved tracking performance of the proposed method.

Early gastric cancer detection in four areas at different gastric cancer death rate

1981 ◽  
Vol 11 (2) ◽  
pp. 123-132 ◽  
Author(s):  
H. Aste ◽  
D. Amadori ◽  
C. Maltoni ◽  
M. Crespi ◽  
V. Pugliese ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Early Gastric Cancer ◽  
Cancer Detection ◽  
Death Rate ◽  
Cancer Death ◽  
Cancer Death Rate

Long-term outcome of a randomized phase III trial exploring the significance of extensive intraoperative peritoneal lavage in addition to standard treatment for ≥ T3 resectable gastric cancer: CCOG 1102.

2018 ◽  
Vol 36 (4_suppl) ◽  
pp. 1-1 ◽  
Author(s):  
Daishi Morimoto ◽  
Kazunari Misawa ◽  
Yoshinari Mochizuki ◽  
Mitsuru Sakai ◽  
Jin Teramoto ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Peritoneal Lavage ◽  
Disease Free Survival ◽  
Phase Iii ◽  
Small Scale ◽  
Clinicopathologic Characteristics ◽  
Term Outcome ◽  
Long Term Outcome ◽  
Phase Iii Trial ◽  
Significant Difference

1 Background: Extensive intraoperative peritoneal lavage (EIPL) for gastric cancer reportedly improved survival by physically removing and reducing intraperitoneal-free cancer cells in a small-scale randomized trial (Ann Surg 2009). To confirm this, we conducted a multicenter randomized Phase III trial. Methods: Eligibility criteria: (i) histologically confirmed primary gastric adenocarcinoma, (ii) clinically T3(SS), T4a(SE) or T4b(SI), (iii) clinically M0, and (iv) scheduled for total or distal gastrectomy. Patients were intraoperatively randomized to either the EIPL group or the non-EIPL group after confirming the ≥T3 status and resectability. In the EIPL group, peritoneal lavage was conducted at least 10 times using 1L of saline before the closure of the abdomen. In the non-EIPL group, the lavage was conducted with ≤3L of saline. The primary end-point was disease-free survival (DFS). To detect the difference of 15% in 3-year DFS with a both-sided alpha of 5% and 80% power, the planned sample size was 300 cases. This study was registered as UMIN000005907. Results: Between July 2011 and January 2014, 314 patients were registered from 15 institutions. After excluding the R1/R2 resection cases, 295 patients (145 in the EIPL group and 150 in the non-EIPL group) were analyzed. There were no significant differences between the groups in clinicopathologic characteristics. The median volume of saline for the peritoneal lavage was 10.0 L (10 - 12) for the EIPL group and 3.0 L (1 - 4) for the non-EIPL group. No difference was observed in the incidence of postoperative complications. The 3-year DFS was 63.9% in the EIPL group and 59.7% in the non-EIPL group (p = 0.25, Hazard ratio 0.81 [95% CI 0.57-1.16]). The overall survival rate of the EIPL group and non-EIPL group were 75.0%, 73.7% (3 years), and 62.5%, 57.1% (5 years), respectively (p = 0.65). In the subset analysis, no subgroup with significant difference in survival was identified. Conclusions: Although EIPL for advanced gastric cancer was safe and suggested some efficacy, the primary endpoint designed based on the previous small-scale trial was not met. Clinical trial information: 000005907.

Sign in / Sign up

Export Citation Format

Share Document