Efficacy of a novel auto-fluorescence imaging system with computer-assisted color analysis for assessment of colorectal lesions

The Role of Artificial Intelligence in Prospective Real-Time Histological Prediction of Colorectal Lesions during Colonoscopy: A Systematic Review and Meta-Analysis

Artificial intelligence (AI) presents a novel platform for improving disease diagnosis. However, the clinical utility of AI remains limited to discovery studies, with poor translation to clinical practice. Current data suggests that 26% of diminutive pre-malignant lesions and 3.5% of colorectal cancers are missed during colonoscopies. The primary aim of this study was to explore the role of artificial intelligence in real-time histological prediction of colorectal lesions during colonoscopy. A systematic search using MeSH headings relating to "AI", "machine learning", "computer-aided", "colonoscopy", and "colon/rectum/colorectal" identified 2290 studies. Thirteen studies reporting real-time analysis were included. A total of 2958 patients with 5908 colorectal lesions were included. A meta-analysis of six studies reporting sensitivities (95% CI) demonstrated that endoscopist diagnosis was superior to a computer-assisted detection platform, although no statistical significance was reached (p = 0.43). AI applications have shown encouraging results in differentiating neoplastic and non-neoplastic lesions using narrow-band imaging, white light imaging, and blue light imaging. Other modalities include autofluorescence imaging and elastic scattering microscopy. The current literature demonstrates that despite the promise of new endoscopic AI models, they remain inferior to expert endoscopist diagnosis. There is a need to focus developments on real-time histological predictions prior to clinical translation to demonstrate improved diagnostic capabilities and time efficiency.

Artificial intelligence in intestinal polyp and colorectal cancer prediction

Artificial intelligence (AI) algorithms and their application to disease detection and decision support for healthcare professions have greatly evolved in the recent decade. AI has been widely applied and explored in gastroenterology for endoscopic analysis to diagnose intestinal cancers, premalignant polyps, gastrointestinal inflammatory lesions, and bleeding. Patients' responses to treatments and prognoses have both been predicted using AI by combining multiple algorithms. In this review, we explored the recent applications of AI algorithms in the identification and characterization of intestinal polyps and colorectal cancer predictions. AI-based prediction models have the potential to help medical practitioners diagnose, establish prognoses, and find accurate conclusions for the treatment of patients. With the understanding that rigorous validation of AI approaches using randomized controlled studies is solicited before widespread clinical use by health authorities, the article also discusses the limitations and challenges associated with deploying AI systems to diagnose intestinal malignancies and premalignant lesions.

Towards Photodynamic Image-Guided Surgery of Head and Neck Tumors: Photodynamic Priming Improves Delivery and Diagnostic Accuracy of Cetuximab-IRDye800CW

Fluorescence image-guided surgery (IGS) using antibody conjugates of the fluorophore IRDye800CW have revolutionized the surgical debulking of tumors. Cetuximab, an anti-epidermal growth factor receptor (EGFR) monoclonal antibody, conjugated to IRDye800CW (Cet-IRDye800) is the first molecular targeted antibody probe to be used for IGS in head and neck cancer patients. In addition to surgical debulking, Cetuximab-targeted photodynamic therapy (photoimmunotherapy; PIT) is emerging in the clinic as a powerful modality for head and neck tumor photodestruction. A plethora of other photoactivable agents are also in clinical trials for photodynamic-based therapies of head and neck cancer. Considering the vascular and stromal modulating effects of sub-therapeutic photodynamic therapy, namely photodynamic priming (PDP), this study explores the potential synergy between PDP and IGS for a novel photodynamic image-guided surgery (P-IGS) strategy. To the best of our knowledge, this is the first demonstration that PDP of the tumor microenvironment can augment the tumor delivery of full-length antibodies, namely Cet-IRDye800. In this study, we demonstrate a proof-of-concept that PDP primes orthotopic FaDu human head and neck tumors in mice for P-IGS by increasing the delivery of Cet-IRDye800 by up to 138.6%, by expediting its interstitial accumulation by 10.5-fold, and by increasing its fractional tumor coverage by 49.5% at 1 h following Cet-IRDye800 administration. Importantly, PDP improves the diagnostic accuracy of tumor detection by up to 264.2% with respect to vicinal salivary glands at 1 h. As such, PDP provides a time-to-surgery benefit by reducing the time to plateau 10-fold from 25.7 h to 2.5 h. We therefore propose that a pre-operative PDP regimen can expedite and augment the accuracy of IGS-mediated surgical debulking of head and neck tumors and reduce the time-to-IGS. Furthermore, this P-IGS regimen, can also enable a forward-looking post-operative protocol for the photodestruction of unresectable microscopic disease in the surgical bed. Beyond this scope, the role of PDP in the homogenous delivery of diagnostic, theranostic and therapeutic antibodies in solid tumors is of considerable significance to the wider community.

Artificial intelligence and machine learning in colorectal cancer

Colorectal cancer (CRC) is a heterogeneous illness characterized by various epigenetic and microenvironmental changes and is the third-highest cause of cancer-related death in the US. Artificial intelligence (AI) with its ability to allow automatic learning and improvement from experiences using statistical methods and Deep learning has made a distinctive contribution to the diagnosis and treatment of several cancer types. This review discusses the uses and application of AI in CRC screening using automated polyp detection assistance technologies to the development of computer-assisted diagnostic algorithms capable of accurately detecting polyps during colonoscopy and classifying them. Furthermore, we summarize the current research initiatives geared towards building computer-assisted diagnostic algorithms that aim at improving the diagnostic accuracy of benign from premalignant lesions. Considering the evolving transition to more personalized and tailored treatment strategies for CRC, the review also discusses the development of machine learning algorithms to understand responses to therapies and mechanisms of resistance as well as the future roles that AI applications may play in assisting in the treatment of CRC with the aim to improve disease outcomes. We also discuss the constraints and limitations of the use of AI systems. While the medical profession remains enthusiastic about the future of AI and machine learning, large-scale randomized clinical trials are needed to analyze AI algorithms before they can be used.

Current Status and Future Perspectives of Artificial Intelligence in Colonoscopy

The early endoscopic identification, resection, and treatment of precancerous adenoma and early-stage cancer has been shown to reduce not only the prevalence of colorectal cancer but also its mortality rate. Recent advances in endoscopic devices and imaging technology have dramatically improved our ability to detect colorectal lesions and predict their pathological diagnosis. In addition to this, rapid advances in artificial intelligence (AI) technology mean that AI-related research and development is now progressing in the diagnostic imaging field, particularly colonoscopy, and AIs (i.e., devices that mimic cognitive abilities, such as learning and problem-solving) already approved as medical devices are now being introduced into everyday clinical practice. Today, there is an increasing expectation that sophisticated AIs will be able to provide high-level diagnostic performance irrespective of the level of skill of the endoscopist. In this paper, we review colonoscopy-related AI research and the AIs that have already been approved and discuss the future prospects of this technology.

Artificial Intelligence in Endoscopy

Artificial intelligence (AI) is rapidly developing in various medical fields, and there is an increase in research performed in the field of gastrointestinal (GI) endoscopy. In particular, the advent of convolutional neural network, which is a class of deep learning method, has the potential to revolutionize the field of GI endoscopy, including esophagogastroduodenoscopy (EGD), capsule endoscopy (CE), and colonoscopy. A total of 149 original articles pertaining to AI (27 articles in esophagus, 30 articles in stomach, 29 articles in CE, and 63 articles in colon) were identified in this review. The main focuses of AI in EGD are cancer detection, identifying the depth of cancer invasion, prediction of pathological diagnosis, and prediction of Helicobacter pylori infection. In the field of CE, automated detection of bleeding sites, ulcers, tumors, and various small bowel diseases is being investigated. AI in colonoscopy has advanced with several patient-based prospective studies being conducted on the automated detection and classification of colon polyps. Furthermore, research on inflammatory bowel disease has also been recently reported. Most studies of AI in the field of GI endoscopy are still in the preclinical stages because of the retrospective design using still images. Video-based prospective studies are needed to advance the field. However, AI will continue to develop and be used in daily clinical practice in the near future. In this review, we have highlighted the published literature along with providing current status and insights into the future of AI in GI endoscopy.

Liquid Biopsy and Artificial Intelligence as Tools to Detect Signatures of Colorectal Malignancies: A Modern Approach in Patient's Stratification

Colorectal cancer (CRC) is the second most frequently diagnosed type of cancer and a major worldwide public health concern. Despite the global efforts in the development of modern therapeutic strategies, CRC prognosis is strongly correlated with the stage of the disease at diagnosis. Early detection of CRC has a huge impact in decreasing mortality while pre-lesion detection significantly reduces the incidence of the pathology. Even though the management of CRC patients is based on robust diagnostic methods such as serum tumor markers analysis, colonoscopy, histopathological analysis of tumor tissue, and imaging methods (computer tomography or magnetic resonance), these strategies still have many limitations and do not fully satisfy clinical needs due to their lack of sensitivity and/or specificity. Therefore, improvements of the current practice would substantially impact the management of CRC patients. In this view, liquid biopsy is a promising approach that could help clinicians screen for disease, stratify patients to the best treatment, and monitor treatment response and resistance mechanisms in the tumor in a regular and minimally invasive manner. Liquid biopsies allow the detection and analysis of different tumor-derived circulating markers such as cell-free nucleic acids (cfNA), circulating tumor cells (CTCs), and extracellular vesicles (EVs) in the bloodstream. The major advantage of this approach is its ability to trace and monitor the molecular profile of the patient's tumor and to predict personalized treatment in real-time. On the other hand, the prospective use of artificial intelligence (AI) in medicine holds great promise in oncology, for the diagnosis, treatment, and prognosis prediction of disease. AI has two main branches in the medical field: (i) a virtual branch that includes medical imaging, clinical assisted diagnosis, and treatment, as well as drug research, and (ii) a physical branch that includes surgical robots. This review summarizes findings relevant to liquid biopsy and AI in CRC for better management and stratification of CRC patients.

Standardization and implementation of fluorescence molecular endoscopy in the clinic

SIGNIFICANCE

Near-infrared fluorescence molecular endoscopy (NIR-FME) is an innovative technique allowing for in vivo visualization of molecular processes in hollow organs. Despite its potential for clinical translation, NIR-FME still faces challenges, for example, the lack of consensus in performing quality control and standardization of procedures and systems. This may hamper the clinical approval of the technology by authorities and its acceptance by endoscopists. Until now, several clinical trials using NIR-FME have been performed. However, most of these trials had different study designs, making comparison difficult.

AIM

We describe the need for standardization in NIR-FME, provide a pathway for setting up a standardized clinical study, and describe future perspectives for NIR-FME. Body: Standardization is challenging due to many parameters. Invariable parameters refer to the hardware specifications. Variable parameters refer to movement or tissue optical properties. Phantoms can be of aid when defining the influence of these variables or when standardizing a procedure.

CONCLUSION

There is a need for standardization in NIR-FME and hurdles still need to be overcome before a widespread clinical implementation of NIR-FME can be realized. When these hurdles are overcome, clinical outcomes can be compared and systems can be benchmarked, enabling clinical implementation.

Artificial Intelligence-Based Assessment of Colorectal Polyp Histology by Elastic-Scattering Spectroscopy

BACKGROUND

Colonoscopic screening and surveillance for colorectal cancer could be made safer and more efficient if endoscopists could predict histology without the need to biopsy and perform histopathology on every polyp. Elastic-scattering spectroscopy (ESS), using fiberoptic probes integrated into standard biopsy tools, can assess, both in vivo and in real time, the scattering and absorption properties of tissue related to its underlying pathology.

AIMS

The objective of this study was to evaluate prospectively the potential of ESS to predict polyp pathology accurately.

METHODS

We obtained ESS measurements from patients undergoing screening/surveillance colonoscopy using an ESS fiberoptic probe integrated into biopsy forceps. The integrated forceps were used for tissue acquisition, following current standards of care, and optical measurement. All measurements were correlated to the index pathology. A machine learning model was then applied to measurements from 367 polyps in 169 patients to prospectively evaluate its predictive performance.

RESULTS

The model achieved sensitivity of 0.92, specificity of 0.87, negative predictive value (NPV) of 0.87, and high-confidence rate (HCR) of 0.84 for distinguishing 220 neoplastic polyps from 147 non-neoplastic polyps of all sizes. Among 138 neoplastic and 131 non-neoplastic polyps ≤ 5 mm, the model achieved sensitivity of 0.91, specificity of 0.88, NPV of 0.89, and HCR of 0.83.

CONCLUSIONS

Results show that ESS is a viable endoscopic platform for real-time polyp histology, particularly for polyps ≤ 5 mm. ESS is a simple, low-cost, clinically friendly, optical biopsy modality that, when interfaced with minimally obtrusive endoscopic tools, offers an attractive platform for in situ polyp assessment.

Artificial intelligence in colonoscopy

Colorectal cancer remains a leading cause of morbidity and mortality in the United States. Advances in artificial intelligence (AI), specifically computer aided detection and computer-aided diagnosis offer promising methods of increasing adenoma detection rates with the goal of removing more pre-cancerous polyps. Conversely, these methods also may allow for smaller non-cancerous lesions to be diagnosed in vivo and left in place, decreasing the risks that come with unnecessary polypectomies. This review will provide an overview of current advances in the use of AI in colonoscopy to aid in polyp detection and characterization as well as areas of developing research.

Application of Artificial Intelligence in the Detection and Characterization of Colorectal Neoplasm

Endoscpists always have tried to pursue a perfect colonoscopy, and application of artificial intelligence (AI) using deep-learning algorithms is one of the promising supportive options for detection and characterization of colorectal polyps during colonoscopy. Many retrospective studies conducted with real-time application of AI using convolutional neural networks have shown improved colorectal polyp detection. Moreover, a recent randomized clinical trial reported additional polyp detection with shorter analysis time. Studies conducted regarding polyp characterization provided additional promising results. Application of AI with narrow band imaging in real-time prediction of the pathology of diminutive polyps resulted in high diagnostic accuracy. In addition, application of AI with endocytoscopy or confocal laser endomicroscopy was investigated for real-time cellular diagnosis, and the diagnostic accuracy of some studies was comparable to that of pathologists. With AI technology, we can expect a higher polyp detection rate with reduced time and cost by avoiding unnecessary procedures, resulting in enhanced colonoscopy efficiency. However, for AI application in actual daily clinical practice, more prospective studies with minimized selection bias, consensus on standardized utilization, and regulatory approval are needed. (Gut Liver 2021;15:-353)

Artificial Intelligence in Colorectal Cancer Screening, Diagnosis and Treatment. A New Era

The development of artificial intelligence (AI) algorithms has permeated the medical field with great success. The widespread use of AI technology in diagnosing and treating several types of cancer, especially colorectal cancer (CRC), is now attracting substantial attention. CRC, which represents the third most commonly diagnosed malignancy in both men and women, is considered a leading cause of cancer-related deaths globally. Our review herein aims to provide in-depth knowledge and analysis of the AI applications in CRC screening, diagnosis, and treatment based on current literature. We also explore the role of recent advances in AI systems regarding medical diagnosis and therapy, with several promising results. CRC is a highly preventable disease, and AI-assisted techniques in routine screening represent a pivotal step in declining incidence rates of this malignancy. So far, computer-aided detection and characterization systems have been developed to increase the detection rate of adenomas. Furthermore, CRC treatment enters a new era with robotic surgery and novel computer-assisted drug delivery techniques. At the same time, healthcare is rapidly moving toward precision or personalized medicine. Machine learning models have the potential to contribute to individual-based cancer care and transform the future of medicine.

Artificial intelligence in gastrointestinal endoscopy

Background and Aims

Artificial intelligence (AI)-based applications have transformed several industries and are widely used in various consumer products and services. In medicine, AI is primarily being used for image classification and natural language processing and has great potential to affect image-based specialties such as radiology, pathology, and gastroenterology (GE). This document reviews the reported applications of AI in GE, focusing on endoscopic image analysis.

Methods

The MEDLINE database was searched through May 2020 for relevant articles by using key words such as machine learning, deep learning, artificial intelligence, computer-aided diagnosis, convolutional neural networks, GI endoscopy, and endoscopic image analysis. References and citations of the retrieved articles were also evaluated to identify pertinent studies. The manuscript was drafted by 2 authors and reviewed in person by members of the American Society for Gastrointestinal Endoscopy Technology Committee and subsequently by the American Society for Gastrointestinal Endoscopy Governing Board.

Results

Deep learning techniques such as convolutional neural networks have been used in several areas of GI endoscopy, including colorectal polyp detection and classification, analysis of endoscopic images for diagnosis of Helicobacter pylori infection, detection and depth assessment of early gastric cancer, dysplasia in Barrett’s esophagus, and detection of various abnormalities in wireless capsule endoscopy images.

Conclusions

The implementation of AI technologies across multiple GI endoscopic applications has the potential to transform clinical practice favorably and improve the efficiency and accuracy of current diagnostic methods.

Cost savings in colonoscopy with artificial intelligence-aided polyp diagnosis: an add-on analysis of a clinical trial (with video)

BACKGROUND AND AIMS

Artificial intelligence (AI) is being implemented in colonoscopy practice, but no study has investigated whether AI is cost saving. We aimed to quantify the cost reduction using AI as an aid in the optical diagnosis of colorectal polyps.

METHODS

This study is an add-on analysis of a clinical trial that investigated the performance of AI for differentiating colorectal polyps (ie, neoplastic versus non-neoplastic). We included all patients with diminutive (≤5 mm) rectosigmoid polyps in the analyses. The average colonoscopy cost was compared for 2 scenarios: (1) a diagnose-and-leave strategy supported by the AI prediction (ie, diminutive rectosigmoid polyps were not removed when predicted as non-neoplastic), and (2) a resect-all-polyps strategy. Gross annual costs for colonoscopies were also calculated based on the number and reimbursement of colonoscopies conducted under public health insurances in 4 countries.

RESULTS

Overall, 207 patients with 250 diminutive rectosigmoid polyps (104 neoplastic, 144 non-neoplastic, and 2 indeterminate) were included. AI correctly differentiated neoplastic polyps with 93.3% sensitivity, 95.2% specificity, and 95.2% negative predictive value. Thus, 105 polyps were removed and 145 were left under the diagnose-and-leave strategy, which was estimated to reduce the average colonoscopy cost and the gross annual reimbursement for colonoscopies by 18.9% and US$149.2 million in Japan, 6.9% and US$12.3 million in England, 7.6% and US$1.1 million in Norway, and 10.9% and US$85.2 million in the United States, respectively, compared with the resect-all-polyps strategy.

CONCLUSIONS

The use of AI to enable the diagnose-and-leave strategy results in substantial cost reductions for colonoscopy.

Recent Advances and the Potential for Clinical Use of Autofluorescence Detection of Extra-Ophthalmic Tissues

The autofluorescence (AF) characteristics of endogenous fluorophores allow the label-free assessment and visualization of cells and tissues of the human body. While AF imaging (AFI) is well-established in ophthalmology, its clinical applications are steadily expanding to other disciplines. This review summarizes clinical advances of AF techniques published during the past decade. A systematic search of the MEDLINE database and Cochrane Library databases was performed to identify clinical AF studies in extra-ophthalmic tissues. In total, 1097 articles were identified, of which 113 from internal medicine, surgery, oral medicine, and dermatology were reviewed. While comparable technological standards exist in diabetology and cardiology, in all other disciplines, comparability between studies is limited due to the number of differing AF techniques and non-standardized imaging and data analysis. Clear evidence was found for skin AF as a surrogate for blood glucose homeostasis or cardiovascular risk grading. In thyroid surgery, foremost, less experienced surgeons may benefit from the AF-guided intraoperative separation of parathyroid from thyroid tissue. There is a growing interest in AF techniques in clinical disciplines, and promising advances have been made during the past decade. However, further research and development are mandatory to overcome the existing limitations and to maximize the clinical benefits.

Evaluation of autofluorescence and photodynamic diagnosis in assessment of bladder lesions

The ability to detect and diagnose bladder cancer early and precisely is crucial for effective treatment. The aim of this study is to assess the utility of optical biopsy performed with autofluorescence cystoscopy (AFC) using the Onco-LIFE system with numerical color values (NCVs) and by ALA/PDD. Histopathological examination of material obtained during TURBT and/or biopsy of the bladder was carried out in 251 patients. In the case of 35 patients, the selection of the specimen collected for histopathological examination was based using ALA/PDD. In the remaining 216 patients, tissue was collected based on the findings of AFC with NCV. Using AFC, the observed NCV ranged from 0 to 3.86; the highest mean NCV was observed in neoplastic muscle invasive lesions and was equal to 3.18. Furthermore, non-muscle invasive tumors were characterized by a mean NCV equal to 1.54. Tissue with inflammation, metaplasia, and healthy tissue demonstrated significantly lower mean NCV values. The presence of a muscle-invasive tumor increased the NCV by approximately 2.86 compared to healthy tissue. The rates of postoperative complications depend on the examining operator and are observed more often, as much as 65.7 % during ALA/PDD. AFC with NCV using the Onco-LIFE system, as well as ALA/PDD are helpful tools for early diagnosis of bladder precancerous and cancer lesions and for performing targeted biopsies. A significant correlation was found between lesion NCV index and the grade of dysplasia or tumor malignancy. Tissue with inflammation, metaplasia, and healthy tissue demonstrated significantly lower mean NCV values. AFE with NCV have a significantly higher sensitivity than specificity. Low rates of postoperative complications are correlated to the experience of the endoscopist and with AFE/NCV in comparison of ALA/PDD.

Review of computational methods for the detection and classification of polyps in colonoscopy imaging

Computer-aided diagnosis (CAD) is a tool with great potential to help endoscopists in the tasks of detecting and histologically classifying colorectal polyps. In recent years, different technologies have been described and their potential utility has been increasingly evidenced, which has generated great expectations among scientific societies. However, most of these works are retrospective and use images of different quality and characteristics which are analysed off line. This review aims to familiarise gastroenterologists with computational methods and the particularities of endoscopic imaging, which have an impact on image processing analysis. Finally, the publicly available image databases, needed to compare and confirm the results obtained with different methods, are presented.

Gastrointestinal diagnosis using non-white light imaging capsule endoscopy

Capsule endoscopy (CE) has proved to be a powerful tool in the diagnosis and management of small bowel disorders since its introduction in 2001. However, white light imaging (WLI) is the principal technology used in clinical CE at present, and therefore, CE is limited to mucosal inspection, with diagnosis remaining reliant on visible manifestations of disease. The introduction of WLI CE has motivated a wide range of research to improve its diagnostic capabilities through integration with other sensing modalities. These developments have the potential to overcome the limitations of WLI through enhanced detection of subtle mucosal microlesions and submucosal and/or transmural pathology, providing novel diagnostic avenues. Other research aims to utilize a range of sensors to measure physiological parameters or to discover new biomarkers to improve the sensitivity, specificity and thus the clinical utility of CE. This multidisciplinary Review summarizes research into non-WLI CE devices by organizing them into a taxonomic structure on the basis of their sensing modality. The potential of these capsules to realize clinically useful virtual biopsy and computer-aided diagnosis (CADx) is also reported.

Quality assurance of computer-aided detection and diagnosis in colonoscopy

Recent breakthroughs in artificial intelligence (AI), specifically via its emerging sub-field "deep learning," have direct implications for computer-aided detection and diagnosis (CADe and/or CADx) for colonoscopy. AI is expected to have at least 2 major roles in colonoscopy practice-polyp detection (CADe) and polyp characterization (CADx). CADe has the potential to decrease the polyp miss rate, contributing to improving adenoma detection, whereas CADx can improve the accuracy of colorectal polyp optical diagnosis, leading to reduction of unnecessary polypectomy of non-neoplastic lesions, potential implementation of a resect-and-discard paradigm, and proper application of advanced resection techniques. A growing number of medical-engineering researchers are developing both CADe and CADx systems, some of which allow real-time recognition of polyps or in vivo identification of adenomas, with over 90% accuracy. However, the quality of the developed AI systems as well as that of the study designs vary significantly, hence raising some concerns regarding the generalization of the proposed AI systems. Initial studies were conducted in an exploratory or retrospective fashion by using stored images and likely overestimating the results. These drawbacks potentially hinder smooth implementation of this novel technology into colonoscopy practice. The aim of this article is to review both contributions and limitations in recent machine-learning-based CADe and/or CADx colonoscopy studies and propose some principles that should underlie system development and clinical testing.

Artificial intelligence and colonoscopy: Current status and future perspectives

BACKGROUND AND AIM

Application of artificial intelligence in medicine is now attracting substantial attention. In the field of gastrointestinal endoscopy, computer-aided diagnosis (CAD) for colonoscopy is the most investigated area, although it is still in the preclinical phase. Because colonoscopy is carried out by humans, it is inherently an imperfect procedure. CAD assistance is expected to improve its quality regarding automated polyp detection and characterization (i.e. predicting the polyp's pathology). It could help prevent endoscopists from missing polyps as well as provide a precise optical diagnosis for those detected. Ultimately, these functions that CAD provides could produce a higher adenoma detection rate and reduce the cost of polypectomy for hyperplastic polyps.

METHODS AND RESULTS

Currently, research on automated polyp detection has been limited to experimental assessments using an algorithm based on ex vivo videos or static images. Performance for clinical use was reported to have >90% sensitivity with acceptable specificity. In contrast, research on automated polyp characterization seems to surpass that for polyp detection. Prospective studies of in vivo use of artificial intelligence technologies have been reported by several groups, some of which showed a >90% negative predictive value for differentiating diminutive (≤5 mm) rectosigmoid adenomas, which exceeded the threshold for optical biopsy.

CONCLUSION

We introduce the potential of using CAD for colonoscopy and describe the most recent conditions for regulatory approval for artificial intelligence-assisted medical devices.

Real-time computer-aided diagnosis of diminutive rectosigmoid polyps using an auto-fluorescence imaging system and novel color intensity analysis software

Objectives: An endoscopic technique that provides ≥90% negative predictive value (NPV) for differentiating neoplastic polyps is needed for the management of diminutive (≤5 mm) rectosigmoid polyps. This study aimed to assess whether a newly developed software can achieve ≥90% NPV for differentiating rectosigmoid diminutive polyps based on the green-to-red (G/R) ratio, obtained by dividing the green color tone intensity by the red color tone intensity on autofluorescence imaging (AFI). Methods: From December 2017 to May 2018, consecutive patients with known polyps who were scheduled for endoscopic treatment at our institution were prospectively recruited. All colorectal diminutive polyps were differentiated by computer-aided diagnosis using autofluorescence imaging (CAD-AFI) using a novel software-based automatic color intensity analysis; subsequent diagnosis was made by endoscopists based on trimodal imaging endoscopy (TME), which combines AFI, white-light imaging (WLI) and magnifying narrow-band imaging (M-NBI) findings. Thereafter, all polyps were removed endoscopically, and the histopathological diagnosis was evaluated. Results: Ninety-five patients with 258 diminutive rectosigmoid polyps and 171 diminutive non-rectosigmoid polyps were enrolled. Regarding diminutive rectosigmoid polyps, the NPV for differentiating neoplastic polyps was 93.4% (184/197) [95% confidence interval (CI), 89.0%-96.4%] with CAD-AFI and 94.9% (185/195) (95% CI, 90.8%-97.5%) with TME. The accuracy, sensitivity, specificity, and positive predictive value for differentiating diminutive rectosigmoid neoplastic polyps by CAD-AFI were 91.5%, 80.0%, 95.3% and 85.2%, respectively. Conclusions: Real-time CAD-AFI was effective for differentiating diminutive rectosigmoid polyps. This objective technology, which does not require extensive training or endoscopic expertise, can contribute to the effective management of diminutive rectosigmoid polyps.

Computer-aided diagnosis of colorectal polyps using linked color imaging colonoscopy to predict histology

We developed a computer-aided diagnosis (CAD) system based on linked color imaging (LCI) images to predict the histological results of polyps by analyzing the colors of the lesions. A total of 139 images of adenomatous polyps and 69 images of non-adenomatous polyps obtained from our hospital were collected and used to train the CAD system. A test set of LCI images, including both adenomatous and non-adenomatous polyps, was prospectively collected from patients who underwent colonoscopies between Oct and Dec 2017; this test set was used to assess the diagnostic abilities of the CAD system compared to those of human endoscopists (two experts and two novices). The accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of this novel CAD system for the training set were 87.0%, 87.1%, 87.0%, 93.1%, and 76.9%, respectively. The test set included 115 adenomatous polyps and 66 non-adenomatous polyps that were prospectively collected. The CAD system identified adenomatous or non-adenomatous polyps in the test set with an accuracy of 78.4%, a sensitivity of 83.3%, a specificity of 70.1%, a PPV of 82.6%, and an NPV of 71.2%. The accuracy of the CAD system was comparable to that of the expert endoscopists (78.4% vs 79.6%; p = 0.517). In addition, the diagnostic accuracy of the novices was significantly lower to the performance of the experts (70.7% vs 79.6%; p = 0.018). A novel CAD system based on LCI could be a rapid and powerful decision-making tool for endoscopists.

Does the Numerical Colour Value (NCV) correlate with preneoplastic and neoplastic colorectal lesions?

BACKGROUND

White light endoscopy (WLE) is the gold standard for detection of colorectal cancer. Autofluorescence endoscopy (AFE) is among the novel methods expected to increase the sensitivity and specificity of endoscopic diagnosis. The main objective of the study was to determine the diagnostic efficacy of AFE for the detection of preneoplastic and neoplastic colorectal lesions and to identify high-grade neoplasia using Numerical Colour Value (NCV).

METHODS

This retrospective study included 188 patients with colorectal mucosal lesions diagnosed on WLE and assessed using AFE; they were included in the study if a complete patient record was available (description of visualized colorectal lesions, NCV and histopathology report). The NCV was compared with the histological result.

RESULTS

Histology revealed 38 hyperplastic colon polyps, 77 low-grade dysplastic lesions, 17 high-grade dysplastic lesions, 24 adenocarcinomas and 32 inflammatory lesions. The mean NCVs of high-grade dysplasia (HGD) and adenocarcinoma were 2.24 ± 0.22 and 2.73 ± 0.16, respectively, significantly higher than the NCV of hyperplastic colon polyps (0.95 ± 0.06), low-grade dysplasia (LGD) (1.27 ± 0.05) and inflammatory lesions (1.26 ± 0.17). The NCV cut-off value for HGD and adenocarcinoma was set at 1.7. The sensitivity, specificity, PPV (positive predictive value) and NPV (negative predictive value) were 95.2%, 87.9%, 97.5%, 84.8%, respectively.

CONCLUSION

Our study showed that AFE could provide useful diagnostic information regarding preneoplastic and neoplastic colorectal lesions. Additionally, the NCV significantly correlated with the histopathology results.

A Clinical Wide-Field Fluorescence Endoscopic Device for Molecular Imaging Demonstrating Cathepsin Protease Activity in Colon Cancer

PURPOSE

Early and effective detection of cancers of the gastrointestinal tract will require novel molecular probes and advances in instrumentation that can reveal functional changes in dysplastic and malignant tissues. Here, we describe adaptation of a wide-field clinical fiberscope to perform wide-field fluorescence imaging while preserving its white-light capability for the purpose of providing wide-field fluorescence imaging capability to point-of-care microscopes.

PROCEDURES

We developed and used a fluorescent fiberscope to detect signals from a quenched probe, BMV109, that becomes fluorescent when cleaved by, and covalently bound to, active cathepsin proteases. Cathepsins are expressed in inflammation- and tumor-associated macrophages as well as directly from tumor cells and are a promising target for cancer imaging. The fiberscope has a 1-mm outer diameter enabling validation via endoscopic exams in mice, and therefore we evaluated topically applied BMV109 for the ability to detect colon polyps in an azoxymethane-induced colon tumor model in mice.

RESULTS

This wide-field endoscopic imaging device revealed consistent and clear fluorescence signals from BMV109 that specifically localized to the polypoid regions as opposed to the normal adjacent colon tissue (p < 0.004) in the murine colon carcinoma model.

CONCLUSIONS

The sensitivity of detection of BMV109 with the fluorescence fiberscope suggested utility of these tools for early detection at hard-to-reach sites. The fiberscope was designed to be used in conjunction with miniature, endoscope-compatible fluorescence microscopes for dual wide-field and microscopic cancer detection.

Comparative Validation of Polyp Detection Methods in Video Colonoscopy: Results From the MICCAI 2015 Endoscopic Vision Challenge

Colonoscopy is the gold standard for colon cancer screening though some polyps are still missed, thus preventing early disease detection and treatment. Several computational systems have been proposed to assist polyp detection during colonoscopy but so far without consistent evaluation. The lack of publicly available annotated databases has made it difficult to compare methods and to assess if they achieve performance levels acceptable for clinical use. The Automatic Polyp Detection sub-challenge, conducted as part of the Endoscopic Vision Challenge (http://endovis.grand-challenge.org) at the international conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2015, was an effort to address this need. In this paper, we report the results of this comparative evaluation of polyp detection methods, as well as describe additional experiments to further explore differences between methods. We define performance metrics and provide evaluation databases that allow comparison of multiple methodologies. Results show that convolutional neural networks are the state of the art. Nevertheless, it is also demonstrated that combining different methodologies can lead to an improved overall performance.

Effectiveness of color intensity analysis using updated autofluorescence imaging systems for serrated lesions

BACKGROUND AND AIM

We previously reported the effectiveness of color intensity analysis using autofluorescence imaging (AFI) for differentiating colorectal neoplastic lesions from non-neoplastic lesions. However, the ability of AFI systems for differentiating serrated lesions has not been evaluated. In the present study, we assessed the effectiveness of color intensity analysis using updated AFI systems for evaluating serrated lesions.

METHODS

We retrospectively reviewed the data for 48 consecutive patients with 87 serrated lesions that were examined using updated AFI systems and resected at Jikei University Hospital. The mean green/red (G/R) ratio, which is obtained by dividing the mean green color intensities by the mean red color intensities, was calculated for each serrated lesion and compared between hyperplastic polyps, sessile serrated adenomas/polyps (SSA/P) with cytological dysplasia, and SSA/P without cytological dysplasia. We also assessed the area under the receiver operating characteristic curve (AUC) for determining SSA/P (both with and without cytological dysplasia) and SSA/P with cytological dysplasia.

RESULTS

The AUC for determining SSA/P was 0.68; however, the AUC for determining SSA/P with cytological dysplasia was 0.97. With a cut-off for the G/R ratio of <0.93, the sensitivity, specificity, positive predictive value, and negative predictive value for SSA/P with cytological dysplasia were 95.5%, 91.0%, 77.8%, and 98.3%, respectively.

CONCLUSION

Color intensity analysis of serrated lesions using updated AFI systems could effectively distinguish SSA/P with cytological dysplasia from hyperplastic polyps and SSA/P without cytological dysplasia.

Endoscopic features of submucosal deeply invasive colorectal cancer with NBI characteristics : S Saito et al. Endoscopic images of early colorectal cancer

In this review, we discuss the features of conventional endoscopy, magnified endoscopy involving image enhanced endoscopy and endoscopic ultrasonography (EUS) using illustrations for submucosal deeply invasive colorectal cancer (SM-Ca). First, the typical features of SM-Ca were observed, including fold convergence, stiffness, depression (ulceration) and elevated lesions in depressed areas. Magnified endoscopic findings using NBI showed dilated, irregularly shaped micro-capillary vessels. In addition, VI and VN pits were clearly visible using crystal violet staining. In contrast, using EUS, at the third layer we found a layer that was thin compared to the surrounding normal mucosa, which suggested the existence of SM-Ca.

Advanced endoscopic features of ulcerative colitis-associated neoplasias: Quantification of autofluorescence imaging

Ulcerative colitis (UC) patients are well known to carry a higher risk of developing colorectal dysplasia/cancer. However, it is hard to detect the lesion in the early phase during colonoscopy. This pilot study was conducted to analyze the endoscopic characteristics of neoplastic lesions associated with UC using advanced imaging techniques. This is a retrospective analysis of 15 colorectal neoplastic lesion obtained from 11 UC patients during remission who underwent white-light- and advanced endoscopic imaging techniques, including chromoendoscopy, narrow-band imaging and autofluorescence imaging (AFI), and were treated with surgery. These lesions were analyzed for histology, location, size, shape, color and endoscopic features. The green/red ratio was also assessed to quantify the AFI intensity. All 11 patients had extensive colitis with the median disease duration of 14.0 years. A total of 15 lesions, consisting of 8 high-grade dysplasia and 7 cancer, was mostly located in the distal colon (86.7%, 13/15) with the mean size of 8.6 mm. The shape was protruding in 46.7% (7/15), flat elevated in 40.0% (6/15) and flat in 13.3% (2/15) and the color was red in 60.0% (9/15), same colored in 33.3% (5/15) and discolored in 6.7% (1/15). The lesion predominantly showed Kudo's neoplastic pit pattern in 86.7% (13/15; 5 type IIIL, 7 type IV and 1 type VI) on chromoendoscopy and Sano's neoplastic capillary pattern (type IIIa) in 63.6% (7/11) on narrow-band imaging, but were colored purple as neoplastic lesions in only 37.5% (3/8) on AFI. Of note, the AFI green/red ratio was significantly lower in the neoplastic lesions than UC-involved areas (p=0.00014) and UC-uninvolved areas (p=0.00651) irrespective of the lesion's size and histological type. In conclusion, endoscopic analysis based on advanced imaging, in particular AFI quantitation, may be helpful to detect early stage neoplastic lesions in long standing UC. Large-scale, prospective studies are needed.

Improved visibility of colorectal flat tumors using image-enhanced endoscopy

Colonoscopy is considered the gold standard for detecting colorectal tumors; however, conventional colonoscopy can miss flat tumors. We aimed to determine whether visualization of colorectal flat lesions was improved by autofluorescence imaging and narrow-band imaging image analysis in conjunction with a new endoscopy system. Eight physicians compared autofluorescent, narrow-band, and chromoendoscopy images to 30 corresponding white-light images of flat tumors. Physicians rated tumor visibility from each image set as follows: +2 (improved), +1 (somewhat improved), 0 (equivalent to white light), -1 (somewhat decreased), and -2 (decreased). The eight scores for each image were totalled and evaluated. Interobserver agreement was also examined. Autofluorescent, narrow-band, and chromoendoscopy images showed improvements of 63.3% (19/30), 6.7% (2/30), and 73.3% (22/30), respectively, with no instances of decreased visibility. Autofluorescence scores were generally greater than narrow-band scores. Interobserver agreement was 0.65 for autofluorescence, 0.80 for narrow-band imaging, and 0.70 for chromoendoscopy. In conclusion, using a new endoscopy system in conjunction with autofluorescent imaging improved visibility of colorectal flat tumors, equivalent to the visibility achieved using chromoendoscopy.

Endoscopic diagnosis of early Barrett's neoplasia: perspectives for advanced endoscopic technology

Barrett's esophagus (BE) is a metaplastic condition that occurs secondary to gastroesophageal reflux disease. BE is also a precursor to esophageal adenocarcinoma, which, although still rare in Japan, is one of the most rapidly increasing cancers in Western countries. However, the prevalence of gastroesophageal reflux disease has increased significantly over the past few decades in Japan, possibly leading to an incremental rise in BE and the associated inherent risk of adenocarcinoma. Given the poor prognosis of advanced-stage Barrett's adenocarcinoma, endoscopic surveillance is recommended for subjects with BE to detect early neoplasias including dysplasia. However, endoscopic identification of dysplastic lesions is still not sufficiently reliable or subjective, making targeted therapy extremely difficult. Over the past few years, improvements in image resolution, image processing software, and optical filter technology have enabled identification of dysplasia and early cancer in BE patients. We retrieved as many studies on advanced endoscopic technologies in BE as possible from MEDLINE and PubMed. The present review focuses on the emergent clinically available technologies to provide an overview of the technologies, their practical applicability, current status, and future challenges.