Study Overview

Researchers have reported on a soft add-on optical sensor designed to detect bleeding during colonoscopy, specifically in areas beyond the direct view of the colonoscope's camera.

The findings, published in Scientific Reports, detail the sensor's performance in a small live-animal study.

Addressing Colonoscopy Risks

Colonoscopy is a widely used procedure for investigating conditions like colorectal cancer. However, it carries risks such as bleeding and perforation, which can go undetected if they occur outside the camera's field of vision. Current methods, such as wireless capsule endoscopy, are not routinely used for lower gastrointestinal bleeding and cannot be integrated with standard endoscopes.

Sensor Design and Function

The developed sensor is a soft device designed to attach along the colonoscope. It continuously samples colonic fluid through a microfluidic channel. The device uses infrared and green light to monitor transmission through this channel, with reduced green-light transmission indicating the presence of blood.

Earlier ex vivo tests suggested a detection threshold of about 35% blood concentration, which improved to approximately 20% blood concentration in the in vivo porcine study.

In Vivo Evaluation

Experimental Setup

The study involved two 30 kg Yorkshire pigs across two testing sessions with two expert endoscopists. The sensor measured 10mm x 25mm x 1mm, made of polydimethylsiloxane, and featured four optical waveguides connected to a central microfluidic channel. Three sensors were mounted on a flexible sleeve behind the colonoscope tip, positioned 120.

Trial Conduct

Each endoscopist performed 15 tests, navigating the colonoscope from the rectum up to 60cm into the distal colon. Navigation times were recorded for both colonoscope-only and device-assisted trials. In bleeding trials, the sensor was monitored as the scope was retracted past a biopsy site. Intravenous blood was sometimes used to create pooling at biopsy sites when natural bleeding was difficult to sustain.

Key Findings

The sensor achieved 92% accuracy, 83% precision, 100% sensitivity, and 87% specificity in the porcine model.

• Accuracy: The sensor achieved 92% accuracy, 83% precision, 100% sensitivity, and 87% specificity in the porcine model.
• Procedure Time: Median navigation time with the sensor (16.41 seconds) was comparable to without it (17.47 seconds), indicating no significant effect on insertion time.
• Operator Workload: NASA-TLX scores suggested no significant increase in operator workload with the device.
• Safety: No unintended perforations or bleeding caused by the sensor were reported. Internal bruising in one pig was linked to the duration of testing and repeated colonoscopies, not the device.
• Challenges: False positives occurred in some no-bleeding trials, attributed to faecal matter and mucus, highlighting the importance of bowel preparation. Mucus also occasionally affected insertion pace.

Future Implications

The study suggests that a soft optical sensor has the potential to enhance colonoscopy safety by detecting bleeding outside the camera's immediate view. However, the authors emphasized that this was a small preclinical animal study.

Further testing with larger user cohorts is required before the approach can advance towards clinical application.