Assessment of still and moving images in the diagnosis of gastric lesions using magnifying narrow-band imaging in a prospective multicenter trial

Exploration of an effective training system for diagnosis of superficial esophageal squamous cell carcinoma with magnifying narrow-band imaging: Prospective research

BACKGROUND AND AIMS

The aim was to explore an effective training system for diagnosis of superficial esophageal squamous cell carcinoma (SESCC) and its staging with magnifying narrow-band imaging (M-NBI).

PATIENTS AND METHODS

Fifteen endoscopists with no or less M-NBI experience participated in this training, which consisted of four stages and five teaching methods (M-NBI classification criterion, case analysis, hands-on operation, error correction and SESCC pathological knowledge). M-NBI images were evaluated and diagnostic accuracy was analyzed.

RESULTS

After training, the accuracy of distinguishing neoplastic esophageal from non-neoplastic (0.58 ± 0.16 vs. 0.95 ± 0.05, P = 0.000) and diagnosing SESCC staging (0.25 ± 0.26 vs. 0.89 ± 0.08, P = 0.000) with M-NBI were significantly increased. Participants with no M-NBI experience achieve equivalent diagnostic accuracy with less experienced trainees after the training (0.91 ± 0.08 vs. 0.92 ± 0.04, P = 0.816). Besides, diagnosis of MM (muscularis mucosa)/SM1 (submucosal) staging tumors (Stage I, 0.47 ± 0.15; Stage II-III-IV, 0.76 ± 0.12) with M-NBI was difficult for trainees and should be the focus of this training. Every teaching method could improve the diagnostic accuracy for esophageal lesions, especially for case analysis (from 0.59 ± 0.10 to 0.85 ± 0.08, P = 0.000). In addition, the average operation score for trainees was significantly increased after hands-on teaching (60.40 ± 11.11 vs. 91.80 ± 4.28, P = 0.0001).

CONCLUSIONS

For novices, this training system showed efficient performance for diagnosing SESCC staging with M-NBI. Diagnosing MM/SM1 staging SESCC was difficult for beginners, and should be the focus of training.

Interobserver agreement in detection of "white globe appearance" and the ability of educational lectures to improve the diagnosis of gastric lesions

BACKGROUND

White globe appearance (WGA) refers to a small white lesion of globular shape underneath cancerous gastric epithelium that can be clearly visualized by magnifying endoscopy with narrowband imaging (M-NBI). WGA has been reported to be a novel endoscopic marker that is highly specific in differentiating early gastric cancer (GC) from low-grade adenoma, and has a significantly higher prevalence in early GCs than in noncancerous lesions. However, interobserver agreement in detecting WGA and whether training intervention can improve diagnostic accuracy are unknown.

METHODS

Twenty sets of M-NBI images were examined by 16 endoscopists. The endoscopists attended a lecture about WGA, and examined the images again after the lecture. Interobserver agreement in detecting WGA in the second examination and increases in the proportion of correct diagnoses and the degree of confidence of diagnoses of cancerous lesions were evaluated.

RESULTS

The kappa value for interobserver agreement in detecting WGA in the second examination was 0.735. The proportion of correct diagnoses was significantly higher in the second examination compared with the first examination when WGA was present (95.5% vs 55.4%; P < 0.001), but not when WGA was absent (61.6% vs 52.7%; P = 0.190). The proportion of correct diagnoses with a high degree of confidence was significantly higher in the second examination, both with WGA (91.1% vs 29.5%; P < 0.001) and without WGA (36.6% vs 20.5%; P = 0.031).

CONCLUSIONS

The detection of WGA by endoscopists was highly reproducible. A brief educational lecture about WGA increased the proportion of correct diagnoses and the degree of confidence of diagnoses of GC with WGA.

Hue-texture-embedded region-based model for magnifying endoscopy with narrow-band imaging image segmentation based on visual features

BACKGROUND AND OBJECTIVE

Magnification endoscopy with narrow-band imaging (ME-NBI) has become a feasible tool for detecting diseases within the human gastrointestinal tract, and is more applied by physicians to search for pathological abnormalities with gastric cancer such as precancerous lesions, early gastric cancer and advanced cancer. In order to improve the reliability of diseases detection, there is a need for applying or proposing computer-assisted methodologies to efficiently analyze and process ME-NBI images. However, traditional computer vision methodologies, mainly segmentation, do not express well to the specific visual characteristics of NBI scenario.

METHODS

In this paper, two energy functional items based on specific visual characteristics of ME-NBI images have been integrated in the framework of Chan-Vese model to construct the Hue-texture-embedded model. On the one hand, a global hue energy functional was proposed representing a global color information extracted in H channel (HSI color space). On the other hand, a texture energy was put forward presenting local microvascular textures extracted by the PIF of adaptive threshold in S channel.

RESULTS

The results of our model have been compared with Chan-Vese model and manual annotations marked by physicians using F-measure and false positive rate. The value of average F-measure and FPR was 0.61 and 0.16 achieved through the Hue-texture-embedded region-based model. And the C-V model achieved the average F-measure and FPR value of 0.52 and 0.32, respectively. Experiments showed that the Hue-texture-embedded region-based outperforms Chan-Vese model in terms of efficiency, universality and lesion detection.

CONCLUSIONS

Better segmentation results are acquired by the Hue-texture-embedded region-based model compared with the traditional region-based active contour in these five cases: chronic gastritis, intestinal metaplasia and atrophy, low grade neoplasia, high grade neoplasia and early gastric cancer. In the future, we are planning to expand the universality of our proposed methodology to segment other lesions such as intramucosal cancer etc. As long as these issues are solved, we can proceed with the classification of clinically relevant diseases in ME-NBI images to implement a fully automatic computer-assisted diagnosis system.