User login
Dear colleagues,
Innovation is the livelihood of our field, driving major advances in endoscopy and attracting many of us to Gastroenterology. From the development of endoscopic retrograde cholangiopancreatography to the wide-spread adoption of third space endoscopy, we continue to push the boundaries of our practice. But what is the next big disruption in GI, and how will it impact us? Dr. Jeremy Glissen Brown discusses the application of artificial intelligence in GI highlighting its promise but also raising important questions. Dr. Raman Muthusamy elaborates on single-use endoscopes – are they the wave of the future in preventing infection and meeting patient preference? Or will their long-term cost and environmental impact limit their use? I welcome your own thoughts on disruptive innovation in Gastroenterology – share with us on Twitter @AGA_GIHN and by email at [email protected].
Gyanprakash A. Ketwaroo, MD, MSc, is an associate professor of medicine, Yale University, New Haven, Conn., and chief of endoscopy at West Haven (Conn.) VA Medical Center. He is an associate editor for GI & Hepatology News.
The AI revolution, with some important caveats
BY JEREMY R. GLISSEN BROWN, MD, MSC
In 2018, Japan’s Pharmaceutical and Medical Device Agency approved the first artificial intelligence (AI)–based tool, a computer-aided diagnosis system (CADx) for use in clinical practice.1 Since that time, we have seen regulatory approval for a variety of deep learning and AI-based tools in endoscopy and beyond. In addition, there has been an enormous amount of commercial and research interest in AI-based tools in clinical medicine and gastroenterology, and it is almost impossible to open a major gastroenterology journal or go to an academic conference without encountering a slew of AI-based projects.
Many thought and industry leaders say that we are in the midst of an AI revolution in gastroenterology. Indeed, we are at a period of unprecedented growth for deep learning and AI for several reasons, including a recent shift toward data-driven approaches, advancement of machine-learning techniques, and increased computing power. There is, however, also an unprecedented amount of scrutiny and thoughtful conversation about the role AI might play in clinical practice and how we use and regulate these tools in the clinical setting. We are thus in a unique position to ask ourselves the essential question: “Are we on the cusp of an AI revolution in gastroenterology, or are we seeing the release of medical software that is perhaps at best useful in a niche environment and at worse a hype-driven novelty without much clinical benefit?” We will use the most popular use-case, computer aided detection (CADe) of polyps in the colon, to explore this question. In the end, I believe that deep-learning technology will fundamentally change the way we practice gastroenterology. However, this is the perfect time to explore what this means now, and what we can do to shape what it will mean for the future.
CADe: Promise and questions
CADe is a computer vision task that involves localization, such as finding a polyp during colonoscopy and highlighting it with a hollow box. CADe in colonoscopy is perhaps the most well-studied application of deep learning in GI endoscopy to date and is furthest along in the development-implementation pipeline. Because of this, it is an ideal use-case for examining both the evidence that currently supports its use as well as the questions that have come up as we are starting to see CADe algorithms deployed in clinical practice. It is honestly astounding to think that, just 5 years ago, we were talking about CADe as a research concept. While early efforts applying traditional machine learning date back at least to the 1990s, we started to see prospective studies of CADe systems with undetectable or nearly undetectable latency in 2019.2 Since that time we have seen the publication of at least 10 randomized clinical trials involving CADe.
CADe clearly has an impact on some of the conventional quality metrics we use for colonoscopy. While there is considerable heterogeneity in region and design among these trials, most show a significant increase in adenoma detection rate (ADR) and adenomas per colonoscopy. Tandem studies show decreases in adenoma miss rate, and at least one study showed a decrease in sessile serrated lesion miss rate as well. In one of the first randomized, controlled trials across multiple endoscopy centers in Italy, Repici and colleagues showed an increase in ADR from 40.4% in the control group to 54.8% in the CADe group (RR, 1.30; 95% confidence interval, 1.14-1.45).3 Because of pioneering trials such as this one, there are currently several CADe systems that have received regulatory approval in Europe, Asia, and the United States and are being deployed commercially.
It is also clear that the technology is there. In clinical practice, the Food and Drug Administration–approved systems work smoothly, with little to no detectable latency and generally low false-positive and false-negative rates. With clinical deployment, however, we have seen the emergence of healthy debate surrounding every aspect of this task-specific AI. On the development side, important questions include transparency of development data, ensuring that algorithm development is ethical and equitable (as deep learning is susceptible to exacerbating human biases) and methods of data labeling. On the deployment level, important concerns include proper regulation of locked versus “open” algorithms and downstream effects on cost.
In addition, with CADe we have seen a variety of clinical questions crop up because of the novelty of the technology. These include the concern that the increase in ADR we have seen thus far is driven in large part by diminutive and small adenomas (with healthy debate in turn as to these entities’ influence on interval colorectal cancer rates), the effect CADe might have on fellowship training to detect polyps with the human eye, and whether the technology affects sessile serrated lesion detection rates or not. The great thing about such questions is that they have inspired novel research related to CADe in the clinical setting, including how CADe affects trainee ADR, how CADe affects gaze patterns, and how CADe affects recommended surveillance intervals.
CADx, novel applications, and the future
Though there is not space to expand in this particular forum, it is safe to say that with the advancement of CADx in endoscopy and colonoscopy, we have seen similar and novel questions come up. The beautiful thing about all of this is that we are just scratching the surface of what is achievable with deep learning. We have started to see novel projects utilizing deep-learning algorithms, from detecting cirrhosis on ECG to automatically classifying stool consistency on the Bristol Stool Scale from pictures of stool. I ultimately do think that the deployment of AI tools will fundamentally change the way we practice and think about gastroenterology. We are at an incredibly exciting time where we as physicians have the power to shape what that looks like, how we think about AI deployment and regulation and where we go from here.
Dr. Glissen Brown is with the division of gastroenterology and hepatology at Duke University Medical Center, Durham, N.C. He has served as a consultant for Medtronic.
References
1. Aisu N et al. PLOS Digital Health. 2021 Jan 18. doi: 10.1371/journal.pdig.0000001.
2. Wang P et al. Gut. 2019 Oct;68(10):1813-9.
3. Repici A et al. Gastroenterology. 2020 Aug;159(2):512-20.e7.
What’s the future of single-use endoscopes?
BY V. RAMAN MUTHUSAMY, MD, MAS
Single-use endoscopes have been proposed as a definitive solution to the risk of endoscope-transmitted infections. While these infections have been reported for several decades, they have traditionally been associated with identified breaches in the reprocessing protocol. In 2015, numerous cases of duodenoscope-transmitted infections were reported after endoscopic retrograde cholangiopancreatography (ERCP) procedures. Many, if not most, of these cases were not associated with identified deviations from standard high-level disinfection protocols and occurred at high-volume experienced facilities. A subsequent FDA postmarket surveillance study found contamination rates were linked with potentially pathogenic bacteria in approximately 5% of duodenoscopes. Thus, amid growing concerns about the ability to adequately clean these complex devices, these events prompted the development of single-use duodenoscopes. Given the multifactorial causes leading to contaminated duodenoscopes, the advantages of such single-use devices are their ability to ensure the elimination of the potential of infection transmission as these devices are never reused. In addition to this primary benefit, the ability to create single-use devices could lead to more easily available specialty scopes and allow variations in endoscope design that could improve ergonomics. Single-use devices may also expand the ability to provide endoscopic services by eliminating the need for device reprocessing equipment at low-volume sites. However, several concerns have been raised regarding their use, especially if it were to become widespread. These include issues of device quality and performance (potentially leading to more failed cases or adverse events), cost, their environmental impact and current uncertainty regarding their indications for use. Furthermore, new alternatives such as reusable devices with partially disposable components or future low-temperature sterilization options may minimize the need for such devices. We will briefly discuss these issues in more detail below.
Given that nearly all cases of GI device–transmitted infections where standard reprocessing protocols were followed have occurred in duodenoscopes, I will focus on single-use duodenoscopes in this article. It is important that we reassure our patients and colleagues that standard reprocessing appears to be extremely effective with all other types of devices, including elevator containing linear echoendoscopes. Studies investigating the causes of why duodenoscopes have primarily been associated with device-transmitted outbreaks have focused on the complexity of the elevator including its recesses, fixed end-cap and wire channels. However, culturing has shown that up to one-third of contamination may occur in the instrument channels or in the region of the biopsy cap, leading to some potential residual sites of infection even when newly developed reusable devices with disposable elevators/end-caps are utilized.1 Another challenge with reprocessing is the ability to prove residual contamination does not exist. While culturing the devices after reprocessing is most used, it should be noted many sites with outbreaks failed to culture the culprit bacteria from the devices as accessing the sites of contamination can be challenging. The use of other markers of residual contamination such as ATP and tests for residual blood/protein have yielded variable results. Specifically, ATP testing has not correlated well with culture results but may be helpful in assessing the quality of manual cleaning.2
These challenges have made the concept of single-use devices more appealing given the lack of a need reprocess devices or validate cleaning efficacy. Currently, there are two FDA-approved devices on the market, but the published literature to date has largely involved one of these devices. To date, in four published studies that have assessed the clinical performance of single-use duodenoscopes in over 400 patients, procedural success rates have ranged from 91% to 97% with adverse event rates and endoscopist satisfaction scores comparable to reusable devices. Most of these users were expert biliary endoscopists and more data are needed regarding the performance of the device in lower-volume and nonexpert users. While indications for use in these studies have varied, I feel that there are four potential scenarios to utilize these devices: in patients with known multidrug-resistant organisms undergoing ERCP; to facilitate logistics/operations when a reusable device is not available; in critically ill patients who would not tolerate a scope-acquired infection; and in procedures associated with a risk of bacteremia.
While preliminary data suggest single-use duodenoscopes are safe and effective in expert hands, concerns exist regarding their implementation more broadly into clinical practice. First, the devices cost between $1,500-3,000, making them impractical for many health systems. One study estimated the break-even cost of the device to be $800-1,300 based on variation in site volume and device contamination rates.3 However, it should be noted that current enhanced reprocessing protocols for reusable devices may add an additional $75,000-$400,000 per year based on center volume.4 In the United States, there is currently payment by federal and some commercial payors that cover part or all of the device cost, but whether this will continue long-term is unclear. In addition, there is significant concern regarding the environmental impact of a broader mover to single-use devices. Reprocessing programs do exist for these devices, but detailed analyses regarding the environmental effects of a strategy using single-use versus reusable devices and the waste generated from each are needed.
Finally, while primarily created to avoid device-related infection transmission, other benefits can be realized with single-use devices. The potential for ergonomic enhancements (variable handle sizes or shaft stiffness, R- and L-handed scopes) as well as the creation of specialty devices (extra-long or thin devices, devices with special optical or rotational capabilities) may become more feasible with a single-use platform. Finally, the pace of endoscopic innovation and refinement is likely to quicken with a single use platform, and new advancements can be incorporated in a timelier manner.
Conclusion
In summary, I believe single-use devices offer the potential to improve the safety of endoscopic procedures as well as improve procedural access, enhance ergonomics, and foster and expedite device innovation. However, reductions in cost, refining their indications, and developing recycling programs to minimize their environmental impact will be essential before more widespread adoption is achieved.
Dr. Muthusamy is a professor of clinical medicine at the University of California, Los Angeles, and the medical director of endoscopy at the UCLA Health System. He reported relationships with Medtronic, Boston Scientific, Motus GI, Endogastric Solutions, and Capsovision.
References
1. Bartles RL et al. Gastrointest Endosc. 2018 Aug;88(2):306-13.e2.
2. Day LW et al. Gastrointest Endosc. 2021 Jan;93(1):11-33.e6.
3. Bang JY et al. Gut. 2019 Nov;68(11):1915-7.
4. Bomman S et al. Endosc Int Open. 2021 Aug 23;9(9):E1404-12.
Dear colleagues,
Innovation is the livelihood of our field, driving major advances in endoscopy and attracting many of us to Gastroenterology. From the development of endoscopic retrograde cholangiopancreatography to the wide-spread adoption of third space endoscopy, we continue to push the boundaries of our practice. But what is the next big disruption in GI, and how will it impact us? Dr. Jeremy Glissen Brown discusses the application of artificial intelligence in GI highlighting its promise but also raising important questions. Dr. Raman Muthusamy elaborates on single-use endoscopes – are they the wave of the future in preventing infection and meeting patient preference? Or will their long-term cost and environmental impact limit their use? I welcome your own thoughts on disruptive innovation in Gastroenterology – share with us on Twitter @AGA_GIHN and by email at [email protected].
Gyanprakash A. Ketwaroo, MD, MSc, is an associate professor of medicine, Yale University, New Haven, Conn., and chief of endoscopy at West Haven (Conn.) VA Medical Center. He is an associate editor for GI & Hepatology News.
The AI revolution, with some important caveats
BY JEREMY R. GLISSEN BROWN, MD, MSC
In 2018, Japan’s Pharmaceutical and Medical Device Agency approved the first artificial intelligence (AI)–based tool, a computer-aided diagnosis system (CADx) for use in clinical practice.1 Since that time, we have seen regulatory approval for a variety of deep learning and AI-based tools in endoscopy and beyond. In addition, there has been an enormous amount of commercial and research interest in AI-based tools in clinical medicine and gastroenterology, and it is almost impossible to open a major gastroenterology journal or go to an academic conference without encountering a slew of AI-based projects.
Many thought and industry leaders say that we are in the midst of an AI revolution in gastroenterology. Indeed, we are at a period of unprecedented growth for deep learning and AI for several reasons, including a recent shift toward data-driven approaches, advancement of machine-learning techniques, and increased computing power. There is, however, also an unprecedented amount of scrutiny and thoughtful conversation about the role AI might play in clinical practice and how we use and regulate these tools in the clinical setting. We are thus in a unique position to ask ourselves the essential question: “Are we on the cusp of an AI revolution in gastroenterology, or are we seeing the release of medical software that is perhaps at best useful in a niche environment and at worse a hype-driven novelty without much clinical benefit?” We will use the most popular use-case, computer aided detection (CADe) of polyps in the colon, to explore this question. In the end, I believe that deep-learning technology will fundamentally change the way we practice gastroenterology. However, this is the perfect time to explore what this means now, and what we can do to shape what it will mean for the future.
CADe: Promise and questions
CADe is a computer vision task that involves localization, such as finding a polyp during colonoscopy and highlighting it with a hollow box. CADe in colonoscopy is perhaps the most well-studied application of deep learning in GI endoscopy to date and is furthest along in the development-implementation pipeline. Because of this, it is an ideal use-case for examining both the evidence that currently supports its use as well as the questions that have come up as we are starting to see CADe algorithms deployed in clinical practice. It is honestly astounding to think that, just 5 years ago, we were talking about CADe as a research concept. While early efforts applying traditional machine learning date back at least to the 1990s, we started to see prospective studies of CADe systems with undetectable or nearly undetectable latency in 2019.2 Since that time we have seen the publication of at least 10 randomized clinical trials involving CADe.
CADe clearly has an impact on some of the conventional quality metrics we use for colonoscopy. While there is considerable heterogeneity in region and design among these trials, most show a significant increase in adenoma detection rate (ADR) and adenomas per colonoscopy. Tandem studies show decreases in adenoma miss rate, and at least one study showed a decrease in sessile serrated lesion miss rate as well. In one of the first randomized, controlled trials across multiple endoscopy centers in Italy, Repici and colleagues showed an increase in ADR from 40.4% in the control group to 54.8% in the CADe group (RR, 1.30; 95% confidence interval, 1.14-1.45).3 Because of pioneering trials such as this one, there are currently several CADe systems that have received regulatory approval in Europe, Asia, and the United States and are being deployed commercially.
It is also clear that the technology is there. In clinical practice, the Food and Drug Administration–approved systems work smoothly, with little to no detectable latency and generally low false-positive and false-negative rates. With clinical deployment, however, we have seen the emergence of healthy debate surrounding every aspect of this task-specific AI. On the development side, important questions include transparency of development data, ensuring that algorithm development is ethical and equitable (as deep learning is susceptible to exacerbating human biases) and methods of data labeling. On the deployment level, important concerns include proper regulation of locked versus “open” algorithms and downstream effects on cost.
In addition, with CADe we have seen a variety of clinical questions crop up because of the novelty of the technology. These include the concern that the increase in ADR we have seen thus far is driven in large part by diminutive and small adenomas (with healthy debate in turn as to these entities’ influence on interval colorectal cancer rates), the effect CADe might have on fellowship training to detect polyps with the human eye, and whether the technology affects sessile serrated lesion detection rates or not. The great thing about such questions is that they have inspired novel research related to CADe in the clinical setting, including how CADe affects trainee ADR, how CADe affects gaze patterns, and how CADe affects recommended surveillance intervals.
CADx, novel applications, and the future
Though there is not space to expand in this particular forum, it is safe to say that with the advancement of CADx in endoscopy and colonoscopy, we have seen similar and novel questions come up. The beautiful thing about all of this is that we are just scratching the surface of what is achievable with deep learning. We have started to see novel projects utilizing deep-learning algorithms, from detecting cirrhosis on ECG to automatically classifying stool consistency on the Bristol Stool Scale from pictures of stool. I ultimately do think that the deployment of AI tools will fundamentally change the way we practice and think about gastroenterology. We are at an incredibly exciting time where we as physicians have the power to shape what that looks like, how we think about AI deployment and regulation and where we go from here.
Dr. Glissen Brown is with the division of gastroenterology and hepatology at Duke University Medical Center, Durham, N.C. He has served as a consultant for Medtronic.
References
1. Aisu N et al. PLOS Digital Health. 2021 Jan 18. doi: 10.1371/journal.pdig.0000001.
2. Wang P et al. Gut. 2019 Oct;68(10):1813-9.
3. Repici A et al. Gastroenterology. 2020 Aug;159(2):512-20.e7.
What’s the future of single-use endoscopes?
BY V. RAMAN MUTHUSAMY, MD, MAS
Single-use endoscopes have been proposed as a definitive solution to the risk of endoscope-transmitted infections. While these infections have been reported for several decades, they have traditionally been associated with identified breaches in the reprocessing protocol. In 2015, numerous cases of duodenoscope-transmitted infections were reported after endoscopic retrograde cholangiopancreatography (ERCP) procedures. Many, if not most, of these cases were not associated with identified deviations from standard high-level disinfection protocols and occurred at high-volume experienced facilities. A subsequent FDA postmarket surveillance study found contamination rates were linked with potentially pathogenic bacteria in approximately 5% of duodenoscopes. Thus, amid growing concerns about the ability to adequately clean these complex devices, these events prompted the development of single-use duodenoscopes. Given the multifactorial causes leading to contaminated duodenoscopes, the advantages of such single-use devices are their ability to ensure the elimination of the potential of infection transmission as these devices are never reused. In addition to this primary benefit, the ability to create single-use devices could lead to more easily available specialty scopes and allow variations in endoscope design that could improve ergonomics. Single-use devices may also expand the ability to provide endoscopic services by eliminating the need for device reprocessing equipment at low-volume sites. However, several concerns have been raised regarding their use, especially if it were to become widespread. These include issues of device quality and performance (potentially leading to more failed cases or adverse events), cost, their environmental impact and current uncertainty regarding their indications for use. Furthermore, new alternatives such as reusable devices with partially disposable components or future low-temperature sterilization options may minimize the need for such devices. We will briefly discuss these issues in more detail below.
Given that nearly all cases of GI device–transmitted infections where standard reprocessing protocols were followed have occurred in duodenoscopes, I will focus on single-use duodenoscopes in this article. It is important that we reassure our patients and colleagues that standard reprocessing appears to be extremely effective with all other types of devices, including elevator containing linear echoendoscopes. Studies investigating the causes of why duodenoscopes have primarily been associated with device-transmitted outbreaks have focused on the complexity of the elevator including its recesses, fixed end-cap and wire channels. However, culturing has shown that up to one-third of contamination may occur in the instrument channels or in the region of the biopsy cap, leading to some potential residual sites of infection even when newly developed reusable devices with disposable elevators/end-caps are utilized.1 Another challenge with reprocessing is the ability to prove residual contamination does not exist. While culturing the devices after reprocessing is most used, it should be noted many sites with outbreaks failed to culture the culprit bacteria from the devices as accessing the sites of contamination can be challenging. The use of other markers of residual contamination such as ATP and tests for residual blood/protein have yielded variable results. Specifically, ATP testing has not correlated well with culture results but may be helpful in assessing the quality of manual cleaning.2
These challenges have made the concept of single-use devices more appealing given the lack of a need reprocess devices or validate cleaning efficacy. Currently, there are two FDA-approved devices on the market, but the published literature to date has largely involved one of these devices. To date, in four published studies that have assessed the clinical performance of single-use duodenoscopes in over 400 patients, procedural success rates have ranged from 91% to 97% with adverse event rates and endoscopist satisfaction scores comparable to reusable devices. Most of these users were expert biliary endoscopists and more data are needed regarding the performance of the device in lower-volume and nonexpert users. While indications for use in these studies have varied, I feel that there are four potential scenarios to utilize these devices: in patients with known multidrug-resistant organisms undergoing ERCP; to facilitate logistics/operations when a reusable device is not available; in critically ill patients who would not tolerate a scope-acquired infection; and in procedures associated with a risk of bacteremia.
While preliminary data suggest single-use duodenoscopes are safe and effective in expert hands, concerns exist regarding their implementation more broadly into clinical practice. First, the devices cost between $1,500-3,000, making them impractical for many health systems. One study estimated the break-even cost of the device to be $800-1,300 based on variation in site volume and device contamination rates.3 However, it should be noted that current enhanced reprocessing protocols for reusable devices may add an additional $75,000-$400,000 per year based on center volume.4 In the United States, there is currently payment by federal and some commercial payors that cover part or all of the device cost, but whether this will continue long-term is unclear. In addition, there is significant concern regarding the environmental impact of a broader mover to single-use devices. Reprocessing programs do exist for these devices, but detailed analyses regarding the environmental effects of a strategy using single-use versus reusable devices and the waste generated from each are needed.
Finally, while primarily created to avoid device-related infection transmission, other benefits can be realized with single-use devices. The potential for ergonomic enhancements (variable handle sizes or shaft stiffness, R- and L-handed scopes) as well as the creation of specialty devices (extra-long or thin devices, devices with special optical or rotational capabilities) may become more feasible with a single-use platform. Finally, the pace of endoscopic innovation and refinement is likely to quicken with a single use platform, and new advancements can be incorporated in a timelier manner.
Conclusion
In summary, I believe single-use devices offer the potential to improve the safety of endoscopic procedures as well as improve procedural access, enhance ergonomics, and foster and expedite device innovation. However, reductions in cost, refining their indications, and developing recycling programs to minimize their environmental impact will be essential before more widespread adoption is achieved.
Dr. Muthusamy is a professor of clinical medicine at the University of California, Los Angeles, and the medical director of endoscopy at the UCLA Health System. He reported relationships with Medtronic, Boston Scientific, Motus GI, Endogastric Solutions, and Capsovision.
References
1. Bartles RL et al. Gastrointest Endosc. 2018 Aug;88(2):306-13.e2.
2. Day LW et al. Gastrointest Endosc. 2021 Jan;93(1):11-33.e6.
3. Bang JY et al. Gut. 2019 Nov;68(11):1915-7.
4. Bomman S et al. Endosc Int Open. 2021 Aug 23;9(9):E1404-12.
Dear colleagues,
Innovation is the livelihood of our field, driving major advances in endoscopy and attracting many of us to Gastroenterology. From the development of endoscopic retrograde cholangiopancreatography to the wide-spread adoption of third space endoscopy, we continue to push the boundaries of our practice. But what is the next big disruption in GI, and how will it impact us? Dr. Jeremy Glissen Brown discusses the application of artificial intelligence in GI highlighting its promise but also raising important questions. Dr. Raman Muthusamy elaborates on single-use endoscopes – are they the wave of the future in preventing infection and meeting patient preference? Or will their long-term cost and environmental impact limit their use? I welcome your own thoughts on disruptive innovation in Gastroenterology – share with us on Twitter @AGA_GIHN and by email at [email protected].
Gyanprakash A. Ketwaroo, MD, MSc, is an associate professor of medicine, Yale University, New Haven, Conn., and chief of endoscopy at West Haven (Conn.) VA Medical Center. He is an associate editor for GI & Hepatology News.
The AI revolution, with some important caveats
BY JEREMY R. GLISSEN BROWN, MD, MSC
In 2018, Japan’s Pharmaceutical and Medical Device Agency approved the first artificial intelligence (AI)–based tool, a computer-aided diagnosis system (CADx) for use in clinical practice.1 Since that time, we have seen regulatory approval for a variety of deep learning and AI-based tools in endoscopy and beyond. In addition, there has been an enormous amount of commercial and research interest in AI-based tools in clinical medicine and gastroenterology, and it is almost impossible to open a major gastroenterology journal or go to an academic conference without encountering a slew of AI-based projects.
Many thought and industry leaders say that we are in the midst of an AI revolution in gastroenterology. Indeed, we are at a period of unprecedented growth for deep learning and AI for several reasons, including a recent shift toward data-driven approaches, advancement of machine-learning techniques, and increased computing power. There is, however, also an unprecedented amount of scrutiny and thoughtful conversation about the role AI might play in clinical practice and how we use and regulate these tools in the clinical setting. We are thus in a unique position to ask ourselves the essential question: “Are we on the cusp of an AI revolution in gastroenterology, or are we seeing the release of medical software that is perhaps at best useful in a niche environment and at worse a hype-driven novelty without much clinical benefit?” We will use the most popular use-case, computer aided detection (CADe) of polyps in the colon, to explore this question. In the end, I believe that deep-learning technology will fundamentally change the way we practice gastroenterology. However, this is the perfect time to explore what this means now, and what we can do to shape what it will mean for the future.
CADe: Promise and questions
CADe is a computer vision task that involves localization, such as finding a polyp during colonoscopy and highlighting it with a hollow box. CADe in colonoscopy is perhaps the most well-studied application of deep learning in GI endoscopy to date and is furthest along in the development-implementation pipeline. Because of this, it is an ideal use-case for examining both the evidence that currently supports its use as well as the questions that have come up as we are starting to see CADe algorithms deployed in clinical practice. It is honestly astounding to think that, just 5 years ago, we were talking about CADe as a research concept. While early efforts applying traditional machine learning date back at least to the 1990s, we started to see prospective studies of CADe systems with undetectable or nearly undetectable latency in 2019.2 Since that time we have seen the publication of at least 10 randomized clinical trials involving CADe.
CADe clearly has an impact on some of the conventional quality metrics we use for colonoscopy. While there is considerable heterogeneity in region and design among these trials, most show a significant increase in adenoma detection rate (ADR) and adenomas per colonoscopy. Tandem studies show decreases in adenoma miss rate, and at least one study showed a decrease in sessile serrated lesion miss rate as well. In one of the first randomized, controlled trials across multiple endoscopy centers in Italy, Repici and colleagues showed an increase in ADR from 40.4% in the control group to 54.8% in the CADe group (RR, 1.30; 95% confidence interval, 1.14-1.45).3 Because of pioneering trials such as this one, there are currently several CADe systems that have received regulatory approval in Europe, Asia, and the United States and are being deployed commercially.
It is also clear that the technology is there. In clinical practice, the Food and Drug Administration–approved systems work smoothly, with little to no detectable latency and generally low false-positive and false-negative rates. With clinical deployment, however, we have seen the emergence of healthy debate surrounding every aspect of this task-specific AI. On the development side, important questions include transparency of development data, ensuring that algorithm development is ethical and equitable (as deep learning is susceptible to exacerbating human biases) and methods of data labeling. On the deployment level, important concerns include proper regulation of locked versus “open” algorithms and downstream effects on cost.
In addition, with CADe we have seen a variety of clinical questions crop up because of the novelty of the technology. These include the concern that the increase in ADR we have seen thus far is driven in large part by diminutive and small adenomas (with healthy debate in turn as to these entities’ influence on interval colorectal cancer rates), the effect CADe might have on fellowship training to detect polyps with the human eye, and whether the technology affects sessile serrated lesion detection rates or not. The great thing about such questions is that they have inspired novel research related to CADe in the clinical setting, including how CADe affects trainee ADR, how CADe affects gaze patterns, and how CADe affects recommended surveillance intervals.
CADx, novel applications, and the future
Though there is not space to expand in this particular forum, it is safe to say that with the advancement of CADx in endoscopy and colonoscopy, we have seen similar and novel questions come up. The beautiful thing about all of this is that we are just scratching the surface of what is achievable with deep learning. We have started to see novel projects utilizing deep-learning algorithms, from detecting cirrhosis on ECG to automatically classifying stool consistency on the Bristol Stool Scale from pictures of stool. I ultimately do think that the deployment of AI tools will fundamentally change the way we practice and think about gastroenterology. We are at an incredibly exciting time where we as physicians have the power to shape what that looks like, how we think about AI deployment and regulation and where we go from here.
Dr. Glissen Brown is with the division of gastroenterology and hepatology at Duke University Medical Center, Durham, N.C. He has served as a consultant for Medtronic.
References
1. Aisu N et al. PLOS Digital Health. 2021 Jan 18. doi: 10.1371/journal.pdig.0000001.
2. Wang P et al. Gut. 2019 Oct;68(10):1813-9.
3. Repici A et al. Gastroenterology. 2020 Aug;159(2):512-20.e7.
What’s the future of single-use endoscopes?
BY V. RAMAN MUTHUSAMY, MD, MAS
Single-use endoscopes have been proposed as a definitive solution to the risk of endoscope-transmitted infections. While these infections have been reported for several decades, they have traditionally been associated with identified breaches in the reprocessing protocol. In 2015, numerous cases of duodenoscope-transmitted infections were reported after endoscopic retrograde cholangiopancreatography (ERCP) procedures. Many, if not most, of these cases were not associated with identified deviations from standard high-level disinfection protocols and occurred at high-volume experienced facilities. A subsequent FDA postmarket surveillance study found contamination rates were linked with potentially pathogenic bacteria in approximately 5% of duodenoscopes. Thus, amid growing concerns about the ability to adequately clean these complex devices, these events prompted the development of single-use duodenoscopes. Given the multifactorial causes leading to contaminated duodenoscopes, the advantages of such single-use devices are their ability to ensure the elimination of the potential of infection transmission as these devices are never reused. In addition to this primary benefit, the ability to create single-use devices could lead to more easily available specialty scopes and allow variations in endoscope design that could improve ergonomics. Single-use devices may also expand the ability to provide endoscopic services by eliminating the need for device reprocessing equipment at low-volume sites. However, several concerns have been raised regarding their use, especially if it were to become widespread. These include issues of device quality and performance (potentially leading to more failed cases or adverse events), cost, their environmental impact and current uncertainty regarding their indications for use. Furthermore, new alternatives such as reusable devices with partially disposable components or future low-temperature sterilization options may minimize the need for such devices. We will briefly discuss these issues in more detail below.
Given that nearly all cases of GI device–transmitted infections where standard reprocessing protocols were followed have occurred in duodenoscopes, I will focus on single-use duodenoscopes in this article. It is important that we reassure our patients and colleagues that standard reprocessing appears to be extremely effective with all other types of devices, including elevator containing linear echoendoscopes. Studies investigating the causes of why duodenoscopes have primarily been associated with device-transmitted outbreaks have focused on the complexity of the elevator including its recesses, fixed end-cap and wire channels. However, culturing has shown that up to one-third of contamination may occur in the instrument channels or in the region of the biopsy cap, leading to some potential residual sites of infection even when newly developed reusable devices with disposable elevators/end-caps are utilized.1 Another challenge with reprocessing is the ability to prove residual contamination does not exist. While culturing the devices after reprocessing is most used, it should be noted many sites with outbreaks failed to culture the culprit bacteria from the devices as accessing the sites of contamination can be challenging. The use of other markers of residual contamination such as ATP and tests for residual blood/protein have yielded variable results. Specifically, ATP testing has not correlated well with culture results but may be helpful in assessing the quality of manual cleaning.2
These challenges have made the concept of single-use devices more appealing given the lack of a need reprocess devices or validate cleaning efficacy. Currently, there are two FDA-approved devices on the market, but the published literature to date has largely involved one of these devices. To date, in four published studies that have assessed the clinical performance of single-use duodenoscopes in over 400 patients, procedural success rates have ranged from 91% to 97% with adverse event rates and endoscopist satisfaction scores comparable to reusable devices. Most of these users were expert biliary endoscopists and more data are needed regarding the performance of the device in lower-volume and nonexpert users. While indications for use in these studies have varied, I feel that there are four potential scenarios to utilize these devices: in patients with known multidrug-resistant organisms undergoing ERCP; to facilitate logistics/operations when a reusable device is not available; in critically ill patients who would not tolerate a scope-acquired infection; and in procedures associated with a risk of bacteremia.
While preliminary data suggest single-use duodenoscopes are safe and effective in expert hands, concerns exist regarding their implementation more broadly into clinical practice. First, the devices cost between $1,500-3,000, making them impractical for many health systems. One study estimated the break-even cost of the device to be $800-1,300 based on variation in site volume and device contamination rates.3 However, it should be noted that current enhanced reprocessing protocols for reusable devices may add an additional $75,000-$400,000 per year based on center volume.4 In the United States, there is currently payment by federal and some commercial payors that cover part or all of the device cost, but whether this will continue long-term is unclear. In addition, there is significant concern regarding the environmental impact of a broader mover to single-use devices. Reprocessing programs do exist for these devices, but detailed analyses regarding the environmental effects of a strategy using single-use versus reusable devices and the waste generated from each are needed.
Finally, while primarily created to avoid device-related infection transmission, other benefits can be realized with single-use devices. The potential for ergonomic enhancements (variable handle sizes or shaft stiffness, R- and L-handed scopes) as well as the creation of specialty devices (extra-long or thin devices, devices with special optical or rotational capabilities) may become more feasible with a single-use platform. Finally, the pace of endoscopic innovation and refinement is likely to quicken with a single use platform, and new advancements can be incorporated in a timelier manner.
Conclusion
In summary, I believe single-use devices offer the potential to improve the safety of endoscopic procedures as well as improve procedural access, enhance ergonomics, and foster and expedite device innovation. However, reductions in cost, refining their indications, and developing recycling programs to minimize their environmental impact will be essential before more widespread adoption is achieved.
Dr. Muthusamy is a professor of clinical medicine at the University of California, Los Angeles, and the medical director of endoscopy at the UCLA Health System. He reported relationships with Medtronic, Boston Scientific, Motus GI, Endogastric Solutions, and Capsovision.
References
1. Bartles RL et al. Gastrointest Endosc. 2018 Aug;88(2):306-13.e2.
2. Day LW et al. Gastrointest Endosc. 2021 Jan;93(1):11-33.e6.
3. Bang JY et al. Gut. 2019 Nov;68(11):1915-7.
4. Bomman S et al. Endosc Int Open. 2021 Aug 23;9(9):E1404-12.
