Abstract
Background: A pink color change occasionally found by us under magnifying endoscopy with narrow-band imaging (ME-NBI) may be a special feature of early gastric cancer (EGC), and was designated the "pink pattern". The purposes of this study were to determine the relationship between the pink pattern and the cytopathological changes in gastric cancer cells and whether the pink pattern is useful for the diagnosis of EGC. Methods: The color features of ME-NBI images and pathological images of cancerous gastric mucosal surfaces were extracted and quantified. The cosine similarity was calculated to evaluate the correlation between the pink pattern and the nucleus-to-cytoplasm ratio of cancerous epithelial cells. Two diagnostic tests were performed by 12 endoscopists using stored ME-NBI images of 185 gastric lesions to investigate the diagnostic efficacy of the pink pattern for EGC. Results: The cosine similarity between the color values of ME-NBI images and pathological images of 20 lesions was at least 0.744. The median AUC, accuracy, sensitivity, specificity, PPV, and NPV of test 2 were significantly better than those of test 1 for all endoscopists and for the junior and experienced groups. Conclusions: The pink pattern observed in ME-NBI images correlated strongly with the change in the nucleus-to-cytoplasm ratio of gastric epithelial cells, and could be considered a useful marker for the diagnosis of differentiated EGC.
Introduction
Early detection of gastric cancer is paramount for improving patient outcomes. Currently, endoscopists rely on Magnifying Endoscopy with Narrow-Band Imaging (ME-NBI) and the "vessel-plus-surface" (VS) classification system. While effective, the VS system depends on identifying a clear demarcation line (DL) or irregular microvascular patterns—features that are often indeterminate in early-stage cases. This study introduces and validates the "pink pattern," a novel visual cue that serves as a bridge between endoscopic imaging and histopathological changes.
The "Pink Pattern" as a Biological Bridge
The "pink pattern" refers to a distinct color change observed under ME-NBI in certain early gastric cancer (EGC) lesions. Researchers hypothesized that this visual phenomenon is a macro-scale representation of micro-scale pathological changes, specifically an increased nucleus-to-cytoplasm (N/C) ratio. In cancerous cells, the nucleus enlarges while the cytoplasm shrinks. Because NBI light (415nm and 540nm) has limited penetration, it reflects off these altered cellular structures, creating the characteristic pink hue.
Methodology: Quantifying the Correlation
To validate this link, the study employed a computer vision pipeline using Python and OpenCV. They quantified color features from endoscopic images and compared them to gray values from H&E-stained pathological slides. By calculating the cosine similarity, the team confirmed a strong mathematical correlation (at least 0.744) between the visual "pinkness" and the cellular N/C ratio.
Clinical Impact and Diagnostic Performance
The clinical utility of the pink pattern was tested by 12 endoscopists (6 junior and 6 experienced) evaluating 185 gastric lesions. The study compared standard VS classification (Test 1) against a workflow incorporating the pink pattern (Test 2).
The results demonstrated significant improvements across all metrics:
- Accuracy: Median accuracy rose from 78.7% to 87.0%.
- Sensitivity: Correct identification of cancer jumped from 67.4% to 78.4%.
- Interobserver Agreement: Fleiss’ kappa value improved from 0.51 (moderate) to 0.65 (substantial), indicating higher diagnostic consistency.
Notably, the pink pattern acted as a "great equalizer," providing a significant performance boost for junior endoscopists, suggesting it is a more intuitive feature to master than complex vascular patterns.
Conclusion
The pink pattern in ME-NBI images is a reliable biomarker for early differentiated gastric cancer. By providing a visual link to histopathological changes, it offers a "safety net" for cases where traditional markers like the demarcation line are absent. Future work will focus on multi-center prospective trials and the integration of this feature into AI-assisted diagnostic systems.
Summary updated