Usefulness of flexible spectral imaging color enhancement for the detection and diagnosis of small intestinal lesions found by capsule endoscopy

Automatic detection of small bowel lesions with different bleeding risks based on deep learning models

BACKGROUND

Deep learning provides an efficient automatic image recognition method for small bowel (SB) capsule endoscopy (CE) that can assist physicians in diagnosis. However, the existing deep learning models present some unresolved challenges.

AIM

To propose a novel and effective classification and detection model to automatically identify various SB lesions and their bleeding risks, and label the lesions accurately so as to enhance the diagnostic efficiency of physicians and the ability to identify high-risk bleeding groups.

METHODS

The proposed model represents a two-stage method that combined image classification with object detection. First, we utilized the improved ResNet-50 classification model to classify endoscopic images into SB lesion images, normal SB mucosa images, and invalid images. Then, the improved YOLO-V5 detection model was utilized to detect the type of lesion and its risk of bleeding, and the location of the lesion was marked. We constructed training and testing sets and compared model-assisted reading with physician reading.

RESULTS

The accuracy of the model constructed in this study reached 98.96%, which was higher than the accuracy of other systems using only a single module. The sensitivity, specificity, and accuracy of the model-assisted reading detection of all images were 99.17%, 99.92%, and 99.86%, which were significantly higher than those of the endoscopists' diagnoses. The image processing time of the model was 48 ms/image, and the image processing time of the physicians was 0.40 ± 0.24 s/image (P < 0.001).

CONCLUSION

The deep learning model of image classification combined with object detection exhibits a satisfactory diagnostic effect on a variety of SB lesions and their bleeding risks in CE images, which enhances the diagnostic efficiency of physicians and improves the ability of physicians to identify high-risk bleeding groups.

Virtual Chromoendoscopy in Capsule Endoscopy: A Narrative Review

The usefulness of virtual chromoendoscopy (VC) in capsule endoscopy (CE) isa controversial issue, with conflicting studies regarding its efficacy. FICE and a blue filter were embedded in the PillCamTM software, with the aim to assist readers in identifying the source of obscure gastrointestinal (GI) bleeding (OGIB), coeliac disease mucosal changes and other small and large bowel lesions, including polyps and tumors. This review aims to summarize the existing evidence on the value of VC in the visualization and identification of different types of pathology. Overall, VC in CE with FICE 1 and 2 can be a useful adjunctive tool and may increase the visibility of pigmented lesions, such as angiectasias and ulcers. However, it does not appear to improve the detection of polyps or tumors. On the other hand, the role of FICE 3 and the blue filter appears to be limited. FICE may also be helpful in differentiating hyperplastic and adenomatous colonic polyps during colon capsule endoscopy, although more evidence is needed.

Image-Enhanced Capsule Endoscopy Improves the Identification of Small Intestinal Lesions

Image-enhanced endoscopy is useful for diagnosing and identifying lesions in the gastrointestinal tract. Recently, image-enhanced endoscopy has become a breakthrough technology that has attracted significant attention. This image enhancing technology is available for capsule endoscopy, which is an effective tool for small intestinal lesions and has been applied in flexible spectral color enhancement technology and in contrast capsule like narrow-band imaging. In this field, most researchers focus on improving the visibility and detection of small intestinal lesions. This review summarizes previous studies on image-enhanced capsule endoscopy and aims to evaluate the efficacy of this technology.

A comprehensive review on the utility of capsule endoscopy in coeliac disease: From computational analysis to the bedside

Small bowel capsule endoscopy (SBCE) can identify macroscopic changes of coeliac disease and assess the extent of disease in the small bowel beyond the duodenum. SBCE has a good sensitivity for the detection of coeliac disease in comparison to histology owing to several ideal features such as a high magnification. It also plays a useful role in detecting complications in patients with refractory coeliac disease. Several studies have been carried out on transforming images obtained from small bowel capsule endoscopy to enable the automated detection of features related to coeliac disease. This review discusses the current roles played by small bowel capsule endoscopy in coeliac disease. It identifies future potential roles of this technique and describes in great detail the role of computational analysis in the detection of coeliac disease and how it can be adapted to current available technology.

Improved Visibility of Barrett's Esophagus with Linked Color Imaging: Inter- and Intra-Rater Reliability and Quantitative Analysis

BACKGROUND/AIMS

To evaluate the usefulness of linked color imaging (LCI) and blue LASER imaging (BLI) in Barrett's esophagus (BE) compared with white light imaging (WLI).

METHODS

Five expert and trainee endoscopists compared WLI, LCI, and BLI images obtained from 63 patients with short-segment BE. Physicians assessed visibility as follows: 5 (improved), 4 (somewhat improved), 3 (equivalent), 2 (somewhat decreased), and one (decreased). Scores were evaluated to assess visibility. The inter- and intra-rater reliability (intra-class correlation coefficient) of image assessments were also evaluated. Images were objectively evaluated based on L* a* b* color values and color differences (ΔE*) in a CIELAB color space system.

RESULTS

Improved visibility compared with WLI was achieved for LCI: 44.4%, BLI: 0% for all endoscopists; LCI: 55.6%, BLI: 1.6% for trainees; and LCI: 47.6%, BLI: 0% for experts. The visibility score of trainees compared with experts was significantly higher for LCI (p = 0.02). Intra- and inter-rater reliability ratings for LCI compared with WLI were "moderate" for trainees, and "moderate-substantial" for experts. The ΔE* revealed statistically significant differences between WLI and LCI.

CONCLUSION

LCI improved the visibility of short-segment BE compared with WLI, especially for trainees, when evaluated both subjectively and objectively.

White light-emitting contrast image capsule endoscopy for visualization of small intestine lesions: a pilot study

BACKGROUND AND STUDY AIMS

 Capsule endoscopy (CE) has become a routine means of diagnosing obscure gastrointestinal bleeding (OGIB) in the small intestine. Capsules using novel blue-enhanced white light-emitting diodes are expected to acquire clearer contrast images (CIs) of the small bowel vasculature. We conducted a pilot study to examine whether CIs facilitate visualization of small bowel erosions, ulcers, and areas of angioectasia compared with standard white light images (WLIs).

PATIENTS AND METHODS

 A total of 24 patients with OGIB were recruited in this study. The main outcome measure was visibility of lesions on CIs compared with WLIs. We also examined the color difference between lesions and normal mucosa (ΔE) with each imaging modality.

RESULTS

 Three experienced physicians retrospectively evaluated 138 images of small bowel lesions (107 erosions, or ulcers, and 31 areas of angioectasia) obtained from 24 CE examinations. The endoscopists judged that compared with WLIs, CIs afforded easier identification of erosions or ulcers in 29 of 107 cases (27.1 %), were non-inferior in 68 of 107 cases (63.6 %), and were inferior in 10 of 107 cases (9.3 %). Identification of angioectasia was judged to be easier with CIs in 15 of 31 cases (48.4 %), non-inferior in 13 of 31 cases (41.9 %), and inferior in 3 of 31 cases (9.7 %). ΔE was significantly higher for CIs than WLIs, especially for angioectasia, potentially explaining why lesions were easier to visualize.

CONCLUSIONS

 CIs obtained by CE appear to facilitate identification of small bowel erosions, ulcers, and areas of angioectasia compared with WLIs.

Usefulness of virtual chromoendoscopy in the evaluation of subtle small bowel ulcerative lesions by endoscopists with no experience in videocapsule

BACKGROUND AND STUDY AIMS

In videocapsule endoscopy examination (VCE), subtle variations in mucosal hue or pattern such as those seen in ulcerations can be difficult to detect, depending on the experience of the reader. Our aim was to test whether virtual chromoendoscopy (VC) techniques, designed to enhance the contrast between the lesion and the normal mucosa, could improve the characterization of ulcerative mucosal lesions.

PATIENTS AND METHODS

Fifteen trainees or young gastroenterologists with no experience in VCE were randomly assigned to evaluate 250 true ulcerative and 100 false ulcerative, difficult-to-interpret small bowel lesions, initially as white light images (WLI) and then, in a second round, with the addition of one VC setting or again as WLI, labeling them as real lesions or artifacts.

RESULTS

On the overall image evaluation, an improvement in lesion characterization was observed by adding any chromoendoscopy setting, especially Blue mode and FICE 1, with increases in accuracy of 13 % [95 %CI 0.8, 25.3] and 7.1 % [95 %CI - 17.0, 31.3], respectively. However, when only false ulcerative images were considered, with the same presets (Blue mode and FICE 1), there was a loss in accuracy of 10.7 % [95 %CI - 10.9, 32.3] and 7.3 % [95 %CI - 1.3, 16.0], respectively. The interobserver agreement was poor for both readings.

CONCLUSIONS

VC helps beginner VCE readers correctly categorize difficult-to-interpret small bowel mucosal ulcerative lesions. However, false lesions tend to be misinterpreted as true ulcerative with the same presets. Therefore care is advised in using VC especially under poor bowel preparation.

Virtual chromoendoscopy improves the diagnostic yield of small bowel capsule endoscopy in obscure gastrointestinal bleeding

BACKGROUND

Small bowel capsule endoscopy represents the initial investigation for obscure gastrointestinal bleeding. Flexible spectral imaging colour enhancement (FICE) is a virtual chromoendoscopy technique designed to enhance mucosal lesions, available in different settings according to light wavelength-- FICE1, 2 and 3.

AIMS

To compare the diagnostic yield of FICE1 and white light during capsule endoscopy in patients with obscure gastrointestinal bleeding.

METHODS

Retrospective single-centre study including 60 consecutive patients referred for small bowel capsule endoscopy for obscure gastrointestinal bleeding. Endoscopies were independently reviewed in FICE1 and white light; findings were then reviewed by another researcher, establishing a gold standard. Diagnostic yield was defined as the presence of lesions with high bleeding potential (P2) angioectasias, ulcers or tumours.

RESULTS

Diagnostic yield using FICE1 was significantly higher than white light (55% vs. 42%, p=0.021). A superior number of P2 lesions was detected with FICE1 (74 vs. 44, p=0.003), particularly angioectasias (54 vs. 26, p=0.002), but not ulcers or tumours.

CONCLUSIONS

FICE1 was significantly superior to white light, resulting in a 13% improvement in diagnostic yield, and potentially bleeding lesions particularly angioectasias were more often observed. Our results support the use of FICE1 while reviewing small bowel capsule endoscopy for obscure gastrointestinal bleeding.

Efficacy of repeat review with flexible spectral imaging color enhancement in patients with no findings by capsule endoscopy

BACKGROUND/AIM

The efficacy of flexible spectral imaging color enhancement (FICE) ch. 1 (F1) for the detection of ulcerative lesions and angioectasias in the small intestine with capsule endoscopy (CE) has been reported. In the present study, we evaluated whether F1 could detect incremental findings in patients with no findings in a standard review mode.

PATIENTS AND METHODS

In total, 52 patients (age: 60.1 ± 15.3 years; 30 males) with obscure gastrointestinal bleeding (OGIB) who underwent CE and in whom no lesion was detected in the small intestine in the standard mode (first review) were enrolled. Two experienced endoscopists independently reviewed CE videos again by F1 (second review). The following findings were defined to be significant: Ulcers, erosions, aphthas, angioectasias, tumors, and bleeding. Incremental findings at the second review were checked at F1 and in standard mode by the two reviewers (third review). Finally, the findings were confirmed by the agreement of the two reviewers at the third review.

RESULTS

F1 detected five significant lesions in three patients with overt OGIB; three erosions, one aphtha, and one angioectasia. For nonsignificant lesions, F1 detected 12 red mucosas and 16 red spots. Moreover, 29 patients with 71 findings were considered false positives.

CONCLUSION

F1 detected incremental significant findings in a small percentage of patients with no findings in the standard review mode. In addition, F1 showed many false-positive findings. The incremental effect of a repeated review by F1 in patients with no findings in the first review is limited.

Is virtual chromoendoscopy useful in the evaluation of subtle ulcerative small-bowel lesions detected by video capsule endoscopy?

BACKGROUND

The identification of subtle small-bowel mucosal lesions by video capsule endoscopy (VCE) can be challenging. Virtual chromoendoscopy techniques, based on narrowing the bandwidth of conventional white light endoscopic imaging (WLI), were developed to improve the analysis of mucosal patterns. However, data on the already-implemented Flexible spectral Imaging (or Fujinon Intelligent) Color Enhancement (FICE) software application in VCE are limited.

MATERIALS AND METHODS

An evaluation of 250 difficult-to-interpret small-bowel ulcerative and 50 artifact lesions selected from 64 VCE recordings was conducted by four experienced VCE readers in two steps: initially as WLI, then with the addition of all available virtual chromoendoscopy pre-sets (FICE 1, 2, and 3 and Blue mode). The readers labeled them as real or false ulcerative lesions and rated the usefulness of each of the pre-sets.

RESULTS

Between the first (WLI-only) and second (virtual chromoendoscopy-aided) readings, in terms of accuracy there was a global 16.5 % (95 % confidence interval [95 %CI] 13.6 - 19.4 %) improvement (P < 0.001), derived from a 22 % [95 %CI 18.9 - 25.1 %] improvement in the evaluation of true ulcerative images (P < 0.001) and an 11 % (95 %CI 4.1 - 17.7 %) decrease in the evaluation of false ulcerative ones (P = 0.003). The FICE 1 and 2 pre-sets were rated as most useful.

CONCLUSION

The application of virtual chromoendoscopy for VCE is useful to better categorize difficult-to-interpret small-bowel mucosal ulcerative lesions. However, care must be taken, and individual images should be evaluated only as part of a sequence in a recording because the technology can also mistakenly guide to the incorrect interpretation of artifacts as ulcerative lesions.

Image-enhanced capsule endoscopy for characterization of small bowel lesions

Video capsule endoscopy has revolutionized direct endoscopic imaging of the gut. Small-bowel video capsule endoscopy (SBVCE) is now the first-line procedure for exploring the small bowel in case of obscure digestive bleeding and has also some room in case of Crohn's disease, coeliac disease and polyposis syndrome. In case of obscure digestive bleeding the main lesions are angioectasias, erosions/ulcerations and tumors. As for conventional optical endoscopy search was done for improving the detection and characterization of small-bowel lesions. The Fujinon Intelligent Chromoendoscopy (FICE) has been adapted on the software of the SBVCE (Given Imaging(®)/Medtronics). Although there are some conflicting results on the efficacy of FICE for detecting more lesions than with conventional light, it is now recognized that FICE - particularly the setting 1 - may enhance the delineation or characterization of lesions. The use of three-dimensional representation technique is now feasible but still needs further research.

Wireless capsule endoscopy of the small intestine: a review with future directions

PURPOSE OF REVIEW

Here, we review the clinical applications of small bowel capsule endoscopy. Moreover, we provide an outlook on the exceptional future developments of small bowel capsule endoscopy. We discuss clinical algorithms for diagnosis of small bowel diseases. Multiple studies have shown the potential of capsule endoscopy for identification of the bleeding source located in the small bowel and the increased diagnostic yield over radiographic studies. Capsule endoscopy could detect villous atrophy and severe complications in patients with nonresponsive celiac disease. In addition, small bowel capsule endoscopy was proven as a valid tool to diagnose polyps and tumors and Crohn's disease.

SUMMARY

Major current clinical indications of capsule endoscopy in the small bowel include evaluation of obscure gastrointestinal bleeding, diagnosis and surveillance of small bowel polyps and tumors, celiac disease and Crohn's disease. Recent developments have also passed the way for small bowel capsule endoscopy to become a therapeutic instrument.