Multi-factor COX Model: Early Gastric Cancer Risk After Endoscopic Surgery

290 patient records from Fujian Provincial Hospital detail outcomes after non-radical endoscopic submucosal dissection for early gastric cancer. The dataset includes clinicopathological characteristics like lesion location, size, and invasion depth, used to compare the predictive accuracy of the w-eCura and eCura risk systems. Yangyang Chen published this retrospective case-control study, which monitored patients for a median of 32 months.

Use Cases

• Validate the w-eCura risk scoring system based on clinicopathological features like lymphatic invasion and margin status.
• Train survival prediction models using Kaplan-Meier analysis based on risk grade and follow-up duration.
• Compare the performance of clinical risk models using ROC curve analysis on the area under the curve (AUC) metric.
• Identify independent risk factors for positive cancer residue status based on Cox proportional hazards model outputs.

Strengths

• Includes 290 patient records with a median follow-up duration of 32 months.
• Compares two risk stratification models (w-eCura and eCura) with quantified performance metrics, including an AUC of 0.82.
• Documents multiple clinicopathological characteristics such as lesion location, histological type, and depth of invasion.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data reflects a single-center study at Fujian Provincial Hospital, which may introduce geographic bias.

Provenance

Source

figshare

Collection Method

Retrospective case-control study conducted at Fujian Provincial Hospital.

Time Range

Patients treated between January 2011 and July 2024.

Freshness

Last updated 2026-05-19 17:25:32; freshness should be verified.

Geography

Fujian Provincial Hospital, China.