Barrett’s esophagus on AI’s radar

More than one-quarter of the U.S. adult population has Gastroesophageal Reflux Disease, or GERD, and the condition saddles as many as 20% of its sufferers with Barrett’s esophagus. The latter is a serious risk factor for esophageal cancer.

AI has shown great promise for helping endoscope operators with diagnoses and prognoses. In clinical trials, it’s flagged abnormal cells and cancers in Barrett’s patients with 90% accuracy while correctly ruling it out at an 80% clip.

With more research and experimentation, the technology can help push the field further forward, suggest two gastroenterologists with special interest in both Barrett’s and AI.

“AI may also prove to be useful in quality control, streamlining clinical work, documentation and lessening the administrative load on physicians,” they state. “Research in this area is advancing at a rapid rate, and, as the field expands, regulations and guidelines will need to be put into place to better regulate the growth and use of AI.”

The physicians, Nour Hamade, MD, of Indiana University and Prateek Sharma, MD, of the Kansas City VA Medical Center, consider AI’s present state and future potential in Barrett’s esophagus care in a literature review published by Therapeutic Advances in Gastrointestinal Endoscopy.

Among the areas of activity they spotlight:

Histopathology interpretation and AI. Diagnosing Barrett’s-associated tissue masses, or neoplasia, on histology slides is no easy task, Hamade and Sharma point out. Its very difficulty may invite AI developers to lend a hand.

“This may be especially true with low-grade dysplasia which has been shown to have a very low interobserver agreement even among expert histopathologists,” they report. One research team used a convolutional neural network to classify slide images as noncancerous, precancerous or cancerous. Their algorithm had classification accuracies of 85% to 89%.

Quality control and AI. Another group of researchers used AI to flag blind spots in upper GI endoscopies. The experimental approach also helped the team quantify the adequacy of areas visualized and measure read times, offering objective indicators of exam quality.

“Such use of AI will be beneficial moving forward to monitor and record the quality of exams as physician reimbursement rates likely will be increasingly based on outcome measures performance with the goal of providing value-based care,” Hamade and Sharma comment. “In addition, as these AI quality systems become more sophisticated, it is possible that their use will expand to other applications, such as providing real-time feedback and objective metrics that can be used to train young endoscopists.”

Proposed clinical use of AI in upper endoscopy. The set expectation within gastroenterology is that, sooner or later, AI will be used to analyze upper-endoscopy videos in real time, Hamade and Sharma note. The technology will not only flag suspicious cellular structures for the endoscopist but also measure and pre-classify lesions.

The endoscopist can then “decide if the area needs to be sampled based on the characterization provided by the machine or managed endoscopically,” they write. “It is possible that the AI system will also generate the endoscopy report at the end, including automated … measurements of hiatal hernia and so on to be reviewed by the endoscopist for verification.”

Future directions and applications. While most AI innovations within gastroenterology have focused on augmenting human image interpretation, subsequent use cases “will likely include additional applications such as streamlining endoscopic and clinical workflows by automatically documenting and interpreting clinical encounters to decrease the administrative workload on physicians.”

The authors add that AI may also help gastroenterology business managers with scheduling, billing and payment.

Hamade and Sharma conclude:

Barrett’s esophagus related dysplasia and early adenocarcinoma can present as very subtle lesions that are difficult to detect and characterize endoscopically. AI systems show promising results to detect these lesions; however, they still need further study and validation, especially in the real-world setting. Commercially developed AI will need to demonstrate cost-effective care that will provide meaningful value and impact on patient care and outcomes. The future appears extremely bright as the field continues to expand with accelerating momentum. Once clinically available, AI promises to significantly impact the field of Barrett’s esophagus detection, diagnosis and endoscopic treatment.”

Read the rest.

Around the web

The Department of Defense is gifting Case Western researchers a grant to study the use of AI in determining whether patients require surgery. 

This study brings light to the prospect of a "fully automated solution" for echocardiogram analysis, experts reported.

Experts noted a "significant" reduction in false positives and false negatives using their modified machine learning model.

Trimed Popup
Trimed Popup