Telemetric capsule-based upper gastrointestinal tract - blood detection - first multicentric experience

AI-KODA score application for cleanliness assessment in video capsule endoscopy frames

BACKGROUND

Currently, there is no automated method for assessing cleanliness in video capsule endoscopy (VCE). Our objectives were to automate the process of evaluating and collecting medical scores of VCE frames according to the existing KOrea-CanaDA (KODA) scoring system by developing an easy-to-use mobile application called artificial intelligence-KODA (AI-KODA) score, as well as to determine the inter-rater and intra-rater reliability of the KODA score among three readers for prospective AI applications, and check the efficacy of the application.

METHOD

From the 28 patient capsule videos considered, 1539 sequential frames were selected at five-minute intervals, and 634 random frames were selected at random intervals during small bowel transit. The frames were processed and shifted to AI-KODA. Three readers (gastroenterology fellows), who had been trained in reading VCE, rated 2173 frames in duplicate four weeks apart after completing the training module on AI-KODA. The scores were saved automatically in real time. Reliability was assessed for each video using estimate of intra-class correlation coefficients (ICCs). Then, the AI dataset was developed using the frames and their respective scores, and it was subjected to automatic classification of the scores via the random forest and the k-nearest neighbors classifiers.

RESULTS

For sequential frames, ICCs for inter-rater variability were 'excellent' to 'good' among the three readers, while ICCs for intra-rater variability were 'good' to 'moderate'. For random frames, ICCs for inter-rater and intra-rater variability were 'excellent' among the three readers. The overall accuracy achieved was up to 61% for the random forest classifier and 62.38% for the k-nearest neighbors classifier.

CONCLUSIONS

AI-KODA automates the process of scoring VCE frames based on the existing KODA score. It saves time in cleanliness assessment and is user-friendly for research and clinical use. Comprehensive benchmarking of the AI dataset is in process.

Conference Report: The FutuRE oF MinimalLy InvasivE GI and Capsule DiagnosTics (REFLECT) Nyborg, Denmark, October 2023

The gastrointestinal (GI) tract, particularly the small bowel (SB), can be challenging for novel investigation tools [...].

Preclinical study of a novel ingestible bleeding sensor for upper gastrointestinal bleeding

BACKGROUND/AIMS

Upper gastrointestinal bleeding (UGIB) is a life-threatening condition that necessitates early identification and intervention and is associated with substantial morbidity, mortality, and socioeconomic burden. However, several diagnostic challenges remain regarding risk stratification and the optimal timing of endoscopy. The PillSense System is a noninvasive device developed to detect blood in patients with UGIB in real time. This study aimed to assess the safety and performance characteristics of PillSense using a simulated bleeding model.

METHODS

A preclinical study was performed using an in vivo porcine model (14 animals). Fourteen PillSense capsules were endoscopically placed in the stomach and blood was injected into the stomach to simulate bleeding. The safety and sensitivity of blood detection and pill excretion were also investigated.

RESULTS

All the sensors successfully detected the presence or absence of blood. The minimum threshold was 9% blood concentration, with additional detection of increasing concentrations of up to 22.5% blood. All the sensors passed naturally through the gastrointestinal tract.

CONCLUSION

This study demonstrated the ability of the PillSense System sensor to detect UGIB across a wide range of blood concentrations. This ingestible device detects UGIB in real time and has the potential to be an effective tool to supplement the current standard of care. These favorable results will be further investigated in future clinical studies.

A convolutional neural network for bleeding detection in capsule endoscopy using real clinical data

BACKGROUND

The goal of the present study was to develop a convolutional neural network for the detection of bleedings in capsule endoscopy videos using realistic clinical data from one single-centre.

METHODS

Capsule endoscopy videos from all 133 patients (79 male, 54 female; meanage = 53.73 years, SDage = 26.13) who underwent capsule endoscopy at our institution between January 2014 and August 2018 were screened for pathology. All videos were screened for pathology by two independent capsule experts and confirmed findings were checked again by a third capsule expert. From these videos, 125 pathological findings (individual episodes of bleeding spanning a total of 5696 images) and 103 non-pathological findings (sections of normal mucosal tissue without pathologies spanning a total of 7420 images) were used to develop and validate a neural network (Inception V3) using transfer learning.

RESULTS

The overall accuracy of the model for the detection of bleedings was 90.6% [95%CI: 89.4%-91.7%], with a sensitivity of 89.4% [95%CI: 87.6%-91.2%] and a specificity of 91.7% [95%CI: 90.1%-93.2%].

CONCLUSION

Our results show that neural networks can detect bleedings in capsule endoscopy videos under realistic, clinical conditions with an accuracy of 90.6%, potentially reducing reading time per capsule and helping to improve diagnostic accuracy.

[Gastrointestinal bleeding]

GASTROINTESTINAL BLEEDING

Gastrointestinal bleeding occurs frequently in clinical practice. The presence of gastrointestinal bleeding usually is suspected by typical clinical history and results of clinical examination and laboratory findings. Endoscopic treatment of gastrointestinal bleeding is associated with high success rates over 90%. Different techniques for endoscopic hemostasis are available and for most indications, a combination of two modalities is recommended. Endoscopic management of recurrent or persistent bleeding is technically more challenging and associated with increased mortality (up to 10%, especially in patients with high age and comorbidities). In this situation, endoscopic hemostasis using an over-the-scope clip has shown to be superior to standard treatment for recurrent peptic ulcer bleeding in the upper gastrointestinal tract. Recent studies also have shown superiority for first-line over-the-scope clip treatment of non-variceal upper gastrointestinal bleeding in high-risk patients. In this review, management of gastrointestinal bleeding is summarized based on current guidelines and current literature.