Optical diagnosis of colorectal neoplasia: A Western perspective

Artificial intelligence for gastric cancer in endoscopy: From diagnostic reasoning to market

Recognition of gastric conditions during endoscopy exams, including gastric cancer, usually requires specialized training and a long learning curve. Besides that, the interobserver variability is frequently high due to the different morphological characteristics of the lesions and grades of mucosal inflammation. In this sense, artificial intelligence tools based on deep learning models have been developed to support physicians to detect, classify, and predict gastric lesions more efficiently. Even though a growing number of studies exists in the literature, there are multiple challenges to bring a model to practice in this field, such as the need for more robust validation studies and regulatory hurdles. Therefore, the aim of this review is to provide a comprehensive assessment of the current use of artificial intelligence applied to endoscopic imaging to evaluate gastric precancerous and cancerous lesions and the barriers to widespread implementation of this technology in clinical routine.

Rapid, High-Resolution, Label-Free, and 3-Dimensional Imaging to Differentiate Colorectal Adenomas and Non-Neoplastic Polyps With Micro-Optical Coherence Tomography

“Resect and discard” paradigm is one of the main strategies to deal with colorectal diminutive polyps after optical diagnosis. However, there are risks that unrecognized potentially malignant lesions are discarded without accurate diagnosis. The purpose of this study is to validate the potential of micro-optical coherence tomography (μOCT) to improve the diagnostic accuracy of colorectal lesions and help endoscopists make better clinical decision without additional pathology costs.

Improved Real-Time Optical Diagnosis of Colorectal Polyps Following a Comprehensive Training Program

BACKGROUND & AIMS

The optimal training method for endoscopic characterization of colorectal polyps using narrow-band imaging is uncertain, and sessile serrated lesions (SSLs) optical diagnosis data are lacking. We aimed to evaluate a comprehensive training program for real-time optical diagnosis of colorectal polyps, including SSLs.

METHODS

We performed a single-institution prospective study of 15 endoscopists trained with the Workgroup Serrated Polyps and Polyposis classification system. After the first phase of in vivo optical diagnosis, their performances were evaluated. After re-education for insufficient competency, they began the second phase. The learning curves and performance on 2 preservation and incorporation of valuable endoscopic innovations benchmarks were assessed.

RESULTS

A total of 7294 polyps, including 486 SSLs, were diagnosed in real-time. The overall accuracy improved from 73.5% in the first phase to 77.1% in the second. The accuracy with high confidence was 79.4% and 85.1% in the first and second phases, respectively. In the first and second phases, the negative predictive values for diminutive neoplastic polyps were 82.1% and 92.5%, respectively, and concordances of the surveillance intervals were 80.7% and 89.7%, respectively. Eight endoscopists achieved the preservation and incorporation of valuable endoscopic innovations benchmarks after the second phase compared with none after the first. In contrast, the high confidence rate decreased from 74.6% to 70.2% as training progressed.

CONCLUSION

A comprehensive training program for real-time optical diagnosis significantly improved performance and reduced individual variability in less-experienced endoscopists. ClinicalTrials.gov no: NCT02516748.

The Cold Revolution: How Far Can It Go?

Cold resection for small colonic polyps, and larger lesions, is being rapidly and widely adopted. Driven by an impressive safety and cost profile compared with conventional polypectomy, these advantages are offset by the limitations of smaller and shallower resection, and absent thermal effects that may permit persistence of residual neoplasia. To overcome this, optimal cold snare technique requires inclusion of a margin of normal mucosa and a piecemeal resection technique for larger polyps. This article examines the fundamentals of cold snare resection and evidence for its application, theorizes on limits to its application, and identifies areas for further research.

Optical classification of neoplastic colorectal polyps - a computer-assisted approach (the COACH study)

BACKGROUND AND AIMS

Clinical data suggest that the quality of optical diagnoses of colorectal polyps differs markedly among endoscopists. The aim of this study was to develop a computer program that was able to differentiate neoplastic from non-neoplastic polyps using unmagnified endoscopic pictures.

METHODS

During colonoscopy procedures polyp photographies were performed using the unmagnified high-definition white light and narrow band image mode. All detected polyps (n = 275) were resected and sent to pathology. Histopathological diagnoses served as the ground truth. Machine learning was used in order to generate a computer-assisted optical biopsy (CAOB) approach. In the test phase pictures were presented to CAOB in order to obtain optical diagnoses. Altogether 788 pictures were available (602 for training the machine learning algorithm and 186 for CAOB testing). All test pictures were also presented to two experts in optical polyp characterization. The primary endpoint of the study was the accuracy of CAOB diagnoses in the test phase.

RESULTS

A total of 100 polyps (of these 52% neoplastic) were used in the CAOB test phase. The mean size of test polyps was 4 mm. Accuracy of the CAOB approach was 78.0%. Sensitivity and negative predictive value were 92.3% and 88.2%, respectively. Accuracy obtained by two expert endoscopists was 84.0% and 77.0%. Regarding accuracy of optical diagnoses CAOB predictions did not differ significantly compared to experts (p = .307 and p = 1.000, respectively).

CONCLUSIONS

CAOB showed good accuracy on the basis of unmagnified endoscopic pictures. Performance of CAOB predictions did not differ significantly from experts' decisions. The concept of computer assistance for colorectal polyp characterization needs to evolve towards a real-time application prior of being used in a broader set-up.

Patient acceptance of the optical diagnosis and misdiagnosis of diminutive colorectal polyps

BACKGROUND AND AIMS

Optical diagnosis allows for real-time endoscopic assessment of colorectal polyp histology and consists of the resect and discard and diagnose and leave paradigms. This survey assessed patient acceptance of optical diagnosis and their responses to a hypothetical doomsday scenario.

METHODS

We conducted a 3-month cross-sectional survey of colonoscopy outpatients presenting to an Australian academic endoscopy center.

RESULTS

A total of 981 patients completed the survey (76.0% response rate). The 60.8% of patients who supported resect and discard were more likely to be older men who co-supported diagnose and leave. Fewer patients (49.6%) supported diagnose and leave. A family history of missed cancer diagnosis (odds ratio [OR], 0.59; P = .003) was significantly associated with rejection of resect and discard, and a personal or family history of bowel cancer (OR, 0.7; P = .04) was significantly associated with rejection of diagnose and leave. In the hypothetical scenario of a cancerous polyp incorrectly left in situ leading to stage III disease, 208 (21.2%) patients would definitely ask for financial compensation, 584 (59.5%) were unsure, and 189 (19.3%) would definitely not seek compensation. The patient-proposed median value of compensation sought was $760,000 USD ($1,000,000 AUD; $1 AUD = $0.76 USD). Notably, 18.5% would be willing to give optical diagnosis another chance after this error.

CONCLUSION

Patient support for optical diagnosis is limited, and those who are not supporters are more likely to seek financial compensation if errors occur.