User login
The rheumatologist of the future will see patients who have been assessed and triaged with artificial intelligence utilizing data from remote kiosk-placed ultrasound scanners and physician-directed algorithms. Practices will be broadly fueled by AI, which will screen charts, produce notes, handle prior authorizations and insurance issues, aid in earlier diagnoses, find patients for clinical trials, and maybe even suggest the next best therapy for individual patients.
Such is the future envisioned by Alvin F. Wells, MD, PhD, and John J. Cush, MD, who discussed the current and forthcoming reach of AI — and their own uses of it — at the 2024 Rheumatology Winter Clinical Symposium.
“We’re not at the stage where ChatGPT and AI can tell us what the next best therapy is, but we’re getting there,” said Dr. Cush, a rheumatologist based in Dallas and executive director of RheumNow.com. For now, he said, “AI affords us a truly big-time increase in efficiency. It helps you deal with your time constraints in managing information overload and task overload.”
At a time when “PubMed doubles every 73 days ... and it’s getting harder and harder to stay abreast,” for example, new applications such as Scite, SciSpace, and Consensus can help curate, focus, and analyze the literature to match one’s own clinical interests. Such review tools are “just now getting into play and are evolving,” Dr. Cush said, noting that many but not all of them are based on ChatGPT, OpenAI’s chatbot that had a over 100 million users by January 2023 — just over a month after its version 3.5 was released.
For Dr. Wells, a rheumatologist and Midwest Region director in the department of rheumatology for the Advocate Health Medical Group in Franklin, Wisconsin, clinician-developed algorithms are helping his group assess patients — often remotely — and triage them to be seen fairly immediately by a rheumatologist versus in 4-6 weeks or in several months. “You can use AI to guide your access,” he said.
A patient “with a family history of RA, sed rate above 50, and osteopenia on x-rays” would be seen within a week, for example, while “another patient who’s had a [positive] ANA with no other symptoms, and maybe a family history, might be seen in 4-6 weeks,” said Dr. Wells, sharing his belief that “there is not a shortage of rheumatologists, [but a] shortage of using rheumatologists efficiently.”
AI for Improving Workflow
Current and future advances will enrich the intersection of AI and virtual medicine and improve outcomes and the rheumatologist-patient interaction, Dr. Wells said, pointing to research presented at the American College of Rheumatology (ACR) 2023 annual meeting on the use of computer vision technology for remotely assessing disease activity in rheumatoid arthritis (RA).
In the proof-of-concept “MeFisto” study, 28 patients with RA used an app that enabled computer vision inference of hand motion data. Upon recording, an algorithm tracked the mean degree change of joint angle on flexion and the mean time to maximal flexion for each joint.
The researchers found a strong correlation between flexion of the distal interphalangeal (DIP) joint and the Disease Activity Score in 28 joints, the Swollen Hand Joint Count, and the Tender Hand Joint Count. DIP flexion was found to be a significant predictor of low disease activity/remission and high disease activity, the researchers reported in their abstract.
“This blows you away — that a single camera on [one’s] smartphone can look at the manipulation of a hand … and that AI can tell me, there’s a chance this might be an inflammatory arthritis,” said Dr. Wells, noting that researchers are also developing ways to detect joint swelling in RA by AI.
AI can also be used for remote ultrasound scanning in RA, as evidenced by use of the ARTHUR system in Europe, he said. Developed by the Danish company ROPCA, the ARTHUR technology (Rheumatoid Arthritis Ultrasound Robot) interacts directly with the patient who has new joint pain or established RA to capture ultrasound images in grayscale and color flow of 11 joints per hand. AI analyzes the images and creates a report for the specialist.
“They’re trying to get a foothold in the US,” Dr. Wells said, sharing his prediction that similar technology will someday be seen not only in pharmacies but also — in support of equitable access — in locations such as grocery stores. “Again,” he said, “nothing will replace us. I’m taking all [such] information and saying, who needs to be seen in 7 days and who can wait.”
AI for Writing, for Improving Practice and Patient Care
To manage his “task overload,” Dr. Cush uses ChatGPT for jobs such as first drafts of articles and making PowerPoint slides. It must be used cautiously for medical writing, however, as inaccuracies and false data/fabricated information — some of which has been coined AI “hallucinations” — are not uncommon.
“It’s very good at manuscript drafts, at generating bibliographies … it can do systematic reviews, it can do network meta-analyses, and it can find trends and patterns that can very helpful when it comes to writing. But you have to know how it’s a tool, and how it can hurt you,” he said.
Researchers recently reported asking ChatGPT to write an editorial about “how AI may replace the rheumatologist in editorial writing,” Dr. Cush noted. ChatGPT was “very politically correct,” he quipped, because it wrote that AI is “a tool to help the rheumatologist, but not replace him.”
Publishers want to preserve human intelligence — critical thinking and the ability to interpret, for instance — and most of the top medical journals (those most often cited) have issued guidance on the use of generative AI. “One said AI can’t be attributed as an author because being an author carries with it accountability of the work, and AI can’t take responsibility,” Dr. Cush said. Journals also “are saying you can use AI but you have to be totally transparent about it … [how it’s used] has to be very well spelled out.”
In practice, chatbots can be used for summarizing medical records, drafting post-visit summaries, collecting patient feedback, reminding about vaccinations, and performing administrative functions. “It’s really limitless as to what chatbots can do,” Dr. Cush said. “The question is, [what is] really going to help you?”
Much of the research submitted for presentation at major rheumatology meetings over the years has had questionable real-world utility and value, he said. But in the future this will likely change. “Take the PsA [psoriatic arthritis] patient who hasn’t responded to methotrexate or apremilast [Otezla]. There are [so many] choices, and there really isn’t a clear one. Shouldn’t data guide us on whether an IL-23 is better than a JAK, or maybe a JAK preferred over a TNF for some reason?” Dr. Cush said. “That’s what we’re hoping will happen down the line.”
More realistic AI-guided clinical scenarios for now include the following: AI screens the chart of a 68-year-old with RA on methotrexate and etanercept who is following up, and retrieves pieces of history — an elevated C-reactive protein 3 months ago, for instance, and diverticulosis 5 years ago. “AI tells you, based on this, he may have active disease, and here are three medications covered by his insurance,” Dr. Wells said.
Or, in the case of a 58-year-old patient with RA who has scheduled a virtual follow-up visit after having been on methotrexate and hydroxychloroquine for 12 weeks, AI detects a low platelet count in her previsit labs and also sees that she received an MMR booster 5 weeks ago at a local CVS Minute Clinic. AI retrieves for the rheumatologist a review article about thrombocytopenic purpura after MMR vaccination.
AI for Drug Development, Clinical Trials
Dr. Cush is following with keen interest the integration of AI into the process of drug development, from drug discovery and biomarker evaluation to clinical trial efficiency and patient recruitment, as well as marketing. “A lot hasn’t been ‘rolled out’ or shown to us, but there’s a lot going on … everyone is investing,” he said. “The number one challenge is regulatory: How will the [Food and Drug Administration] handle AI-generated data sets or AI-generated or monitored trials?”
The FDA is working to ensure quality and utility of data and is rapidly “approving AI algorithms for use in medicine and healthcare,” he said.
AI’s ability to identify patients in populations can not only facilitate earlier diagnoses but can accelerate patient recruitment for clinical trials, Dr. Cush emphasized. He pointed to research presented at the ACR 2021 annual meeting in which a machine-learning algorithm was used with electronic health records in the United Kingdom to estimate the probability of a patient’s being diagnosed with axial spondyloarthritis (axSpA).
AI identified 89 best clinical predictors (out of 820 analyzed). When applying these predictors to the population, AI was able to differentiate patients with axSpA from healthy controls with a sensitivity of 75%, a specificity of 96%, and a positive predictive value of 81%. Such an application of AI “is ideal … It would make clinical trials more streamlined and productive,” he said.
The extent to which AI will lead to cost savings — in the pharmacology arena, for instance, or for Well’s medical group — is unknown, Dr. Cush and Dr. Wells said. And, of course, there are concerns about potential bias and abuse of AI. “The worry,” Dr. Cush said, “is, who’s watching?”
Dr. Wells disclosed that he has research support and has served as a member of advisory boards and/or speaker bureaus for 17 different pharmaceutical or medical technology companies. Dr. Cush disclosed relationships with AbbVie, Amgen, Bristol-Myers Squibb, Novartis, Sanofi, and UCB.
The rheumatologist of the future will see patients who have been assessed and triaged with artificial intelligence utilizing data from remote kiosk-placed ultrasound scanners and physician-directed algorithms. Practices will be broadly fueled by AI, which will screen charts, produce notes, handle prior authorizations and insurance issues, aid in earlier diagnoses, find patients for clinical trials, and maybe even suggest the next best therapy for individual patients.
Such is the future envisioned by Alvin F. Wells, MD, PhD, and John J. Cush, MD, who discussed the current and forthcoming reach of AI — and their own uses of it — at the 2024 Rheumatology Winter Clinical Symposium.
“We’re not at the stage where ChatGPT and AI can tell us what the next best therapy is, but we’re getting there,” said Dr. Cush, a rheumatologist based in Dallas and executive director of RheumNow.com. For now, he said, “AI affords us a truly big-time increase in efficiency. It helps you deal with your time constraints in managing information overload and task overload.”
At a time when “PubMed doubles every 73 days ... and it’s getting harder and harder to stay abreast,” for example, new applications such as Scite, SciSpace, and Consensus can help curate, focus, and analyze the literature to match one’s own clinical interests. Such review tools are “just now getting into play and are evolving,” Dr. Cush said, noting that many but not all of them are based on ChatGPT, OpenAI’s chatbot that had a over 100 million users by January 2023 — just over a month after its version 3.5 was released.
For Dr. Wells, a rheumatologist and Midwest Region director in the department of rheumatology for the Advocate Health Medical Group in Franklin, Wisconsin, clinician-developed algorithms are helping his group assess patients — often remotely — and triage them to be seen fairly immediately by a rheumatologist versus in 4-6 weeks or in several months. “You can use AI to guide your access,” he said.
A patient “with a family history of RA, sed rate above 50, and osteopenia on x-rays” would be seen within a week, for example, while “another patient who’s had a [positive] ANA with no other symptoms, and maybe a family history, might be seen in 4-6 weeks,” said Dr. Wells, sharing his belief that “there is not a shortage of rheumatologists, [but a] shortage of using rheumatologists efficiently.”
AI for Improving Workflow
Current and future advances will enrich the intersection of AI and virtual medicine and improve outcomes and the rheumatologist-patient interaction, Dr. Wells said, pointing to research presented at the American College of Rheumatology (ACR) 2023 annual meeting on the use of computer vision technology for remotely assessing disease activity in rheumatoid arthritis (RA).
In the proof-of-concept “MeFisto” study, 28 patients with RA used an app that enabled computer vision inference of hand motion data. Upon recording, an algorithm tracked the mean degree change of joint angle on flexion and the mean time to maximal flexion for each joint.
The researchers found a strong correlation between flexion of the distal interphalangeal (DIP) joint and the Disease Activity Score in 28 joints, the Swollen Hand Joint Count, and the Tender Hand Joint Count. DIP flexion was found to be a significant predictor of low disease activity/remission and high disease activity, the researchers reported in their abstract.
“This blows you away — that a single camera on [one’s] smartphone can look at the manipulation of a hand … and that AI can tell me, there’s a chance this might be an inflammatory arthritis,” said Dr. Wells, noting that researchers are also developing ways to detect joint swelling in RA by AI.
AI can also be used for remote ultrasound scanning in RA, as evidenced by use of the ARTHUR system in Europe, he said. Developed by the Danish company ROPCA, the ARTHUR technology (Rheumatoid Arthritis Ultrasound Robot) interacts directly with the patient who has new joint pain or established RA to capture ultrasound images in grayscale and color flow of 11 joints per hand. AI analyzes the images and creates a report for the specialist.
“They’re trying to get a foothold in the US,” Dr. Wells said, sharing his prediction that similar technology will someday be seen not only in pharmacies but also — in support of equitable access — in locations such as grocery stores. “Again,” he said, “nothing will replace us. I’m taking all [such] information and saying, who needs to be seen in 7 days and who can wait.”
AI for Writing, for Improving Practice and Patient Care
To manage his “task overload,” Dr. Cush uses ChatGPT for jobs such as first drafts of articles and making PowerPoint slides. It must be used cautiously for medical writing, however, as inaccuracies and false data/fabricated information — some of which has been coined AI “hallucinations” — are not uncommon.
“It’s very good at manuscript drafts, at generating bibliographies … it can do systematic reviews, it can do network meta-analyses, and it can find trends and patterns that can very helpful when it comes to writing. But you have to know how it’s a tool, and how it can hurt you,” he said.
Researchers recently reported asking ChatGPT to write an editorial about “how AI may replace the rheumatologist in editorial writing,” Dr. Cush noted. ChatGPT was “very politically correct,” he quipped, because it wrote that AI is “a tool to help the rheumatologist, but not replace him.”
Publishers want to preserve human intelligence — critical thinking and the ability to interpret, for instance — and most of the top medical journals (those most often cited) have issued guidance on the use of generative AI. “One said AI can’t be attributed as an author because being an author carries with it accountability of the work, and AI can’t take responsibility,” Dr. Cush said. Journals also “are saying you can use AI but you have to be totally transparent about it … [how it’s used] has to be very well spelled out.”
In practice, chatbots can be used for summarizing medical records, drafting post-visit summaries, collecting patient feedback, reminding about vaccinations, and performing administrative functions. “It’s really limitless as to what chatbots can do,” Dr. Cush said. “The question is, [what is] really going to help you?”
Much of the research submitted for presentation at major rheumatology meetings over the years has had questionable real-world utility and value, he said. But in the future this will likely change. “Take the PsA [psoriatic arthritis] patient who hasn’t responded to methotrexate or apremilast [Otezla]. There are [so many] choices, and there really isn’t a clear one. Shouldn’t data guide us on whether an IL-23 is better than a JAK, or maybe a JAK preferred over a TNF for some reason?” Dr. Cush said. “That’s what we’re hoping will happen down the line.”
More realistic AI-guided clinical scenarios for now include the following: AI screens the chart of a 68-year-old with RA on methotrexate and etanercept who is following up, and retrieves pieces of history — an elevated C-reactive protein 3 months ago, for instance, and diverticulosis 5 years ago. “AI tells you, based on this, he may have active disease, and here are three medications covered by his insurance,” Dr. Wells said.
Or, in the case of a 58-year-old patient with RA who has scheduled a virtual follow-up visit after having been on methotrexate and hydroxychloroquine for 12 weeks, AI detects a low platelet count in her previsit labs and also sees that she received an MMR booster 5 weeks ago at a local CVS Minute Clinic. AI retrieves for the rheumatologist a review article about thrombocytopenic purpura after MMR vaccination.
AI for Drug Development, Clinical Trials
Dr. Cush is following with keen interest the integration of AI into the process of drug development, from drug discovery and biomarker evaluation to clinical trial efficiency and patient recruitment, as well as marketing. “A lot hasn’t been ‘rolled out’ or shown to us, but there’s a lot going on … everyone is investing,” he said. “The number one challenge is regulatory: How will the [Food and Drug Administration] handle AI-generated data sets or AI-generated or monitored trials?”
The FDA is working to ensure quality and utility of data and is rapidly “approving AI algorithms for use in medicine and healthcare,” he said.
AI’s ability to identify patients in populations can not only facilitate earlier diagnoses but can accelerate patient recruitment for clinical trials, Dr. Cush emphasized. He pointed to research presented at the ACR 2021 annual meeting in which a machine-learning algorithm was used with electronic health records in the United Kingdom to estimate the probability of a patient’s being diagnosed with axial spondyloarthritis (axSpA).
AI identified 89 best clinical predictors (out of 820 analyzed). When applying these predictors to the population, AI was able to differentiate patients with axSpA from healthy controls with a sensitivity of 75%, a specificity of 96%, and a positive predictive value of 81%. Such an application of AI “is ideal … It would make clinical trials more streamlined and productive,” he said.
The extent to which AI will lead to cost savings — in the pharmacology arena, for instance, or for Well’s medical group — is unknown, Dr. Cush and Dr. Wells said. And, of course, there are concerns about potential bias and abuse of AI. “The worry,” Dr. Cush said, “is, who’s watching?”
Dr. Wells disclosed that he has research support and has served as a member of advisory boards and/or speaker bureaus for 17 different pharmaceutical or medical technology companies. Dr. Cush disclosed relationships with AbbVie, Amgen, Bristol-Myers Squibb, Novartis, Sanofi, and UCB.
The rheumatologist of the future will see patients who have been assessed and triaged with artificial intelligence utilizing data from remote kiosk-placed ultrasound scanners and physician-directed algorithms. Practices will be broadly fueled by AI, which will screen charts, produce notes, handle prior authorizations and insurance issues, aid in earlier diagnoses, find patients for clinical trials, and maybe even suggest the next best therapy for individual patients.
Such is the future envisioned by Alvin F. Wells, MD, PhD, and John J. Cush, MD, who discussed the current and forthcoming reach of AI — and their own uses of it — at the 2024 Rheumatology Winter Clinical Symposium.
“We’re not at the stage where ChatGPT and AI can tell us what the next best therapy is, but we’re getting there,” said Dr. Cush, a rheumatologist based in Dallas and executive director of RheumNow.com. For now, he said, “AI affords us a truly big-time increase in efficiency. It helps you deal with your time constraints in managing information overload and task overload.”
At a time when “PubMed doubles every 73 days ... and it’s getting harder and harder to stay abreast,” for example, new applications such as Scite, SciSpace, and Consensus can help curate, focus, and analyze the literature to match one’s own clinical interests. Such review tools are “just now getting into play and are evolving,” Dr. Cush said, noting that many but not all of them are based on ChatGPT, OpenAI’s chatbot that had a over 100 million users by January 2023 — just over a month after its version 3.5 was released.
For Dr. Wells, a rheumatologist and Midwest Region director in the department of rheumatology for the Advocate Health Medical Group in Franklin, Wisconsin, clinician-developed algorithms are helping his group assess patients — often remotely — and triage them to be seen fairly immediately by a rheumatologist versus in 4-6 weeks or in several months. “You can use AI to guide your access,” he said.
A patient “with a family history of RA, sed rate above 50, and osteopenia on x-rays” would be seen within a week, for example, while “another patient who’s had a [positive] ANA with no other symptoms, and maybe a family history, might be seen in 4-6 weeks,” said Dr. Wells, sharing his belief that “there is not a shortage of rheumatologists, [but a] shortage of using rheumatologists efficiently.”
AI for Improving Workflow
Current and future advances will enrich the intersection of AI and virtual medicine and improve outcomes and the rheumatologist-patient interaction, Dr. Wells said, pointing to research presented at the American College of Rheumatology (ACR) 2023 annual meeting on the use of computer vision technology for remotely assessing disease activity in rheumatoid arthritis (RA).
In the proof-of-concept “MeFisto” study, 28 patients with RA used an app that enabled computer vision inference of hand motion data. Upon recording, an algorithm tracked the mean degree change of joint angle on flexion and the mean time to maximal flexion for each joint.
The researchers found a strong correlation between flexion of the distal interphalangeal (DIP) joint and the Disease Activity Score in 28 joints, the Swollen Hand Joint Count, and the Tender Hand Joint Count. DIP flexion was found to be a significant predictor of low disease activity/remission and high disease activity, the researchers reported in their abstract.
“This blows you away — that a single camera on [one’s] smartphone can look at the manipulation of a hand … and that AI can tell me, there’s a chance this might be an inflammatory arthritis,” said Dr. Wells, noting that researchers are also developing ways to detect joint swelling in RA by AI.
AI can also be used for remote ultrasound scanning in RA, as evidenced by use of the ARTHUR system in Europe, he said. Developed by the Danish company ROPCA, the ARTHUR technology (Rheumatoid Arthritis Ultrasound Robot) interacts directly with the patient who has new joint pain or established RA to capture ultrasound images in grayscale and color flow of 11 joints per hand. AI analyzes the images and creates a report for the specialist.
“They’re trying to get a foothold in the US,” Dr. Wells said, sharing his prediction that similar technology will someday be seen not only in pharmacies but also — in support of equitable access — in locations such as grocery stores. “Again,” he said, “nothing will replace us. I’m taking all [such] information and saying, who needs to be seen in 7 days and who can wait.”
AI for Writing, for Improving Practice and Patient Care
To manage his “task overload,” Dr. Cush uses ChatGPT for jobs such as first drafts of articles and making PowerPoint slides. It must be used cautiously for medical writing, however, as inaccuracies and false data/fabricated information — some of which has been coined AI “hallucinations” — are not uncommon.
“It’s very good at manuscript drafts, at generating bibliographies … it can do systematic reviews, it can do network meta-analyses, and it can find trends and patterns that can very helpful when it comes to writing. But you have to know how it’s a tool, and how it can hurt you,” he said.
Researchers recently reported asking ChatGPT to write an editorial about “how AI may replace the rheumatologist in editorial writing,” Dr. Cush noted. ChatGPT was “very politically correct,” he quipped, because it wrote that AI is “a tool to help the rheumatologist, but not replace him.”
Publishers want to preserve human intelligence — critical thinking and the ability to interpret, for instance — and most of the top medical journals (those most often cited) have issued guidance on the use of generative AI. “One said AI can’t be attributed as an author because being an author carries with it accountability of the work, and AI can’t take responsibility,” Dr. Cush said. Journals also “are saying you can use AI but you have to be totally transparent about it … [how it’s used] has to be very well spelled out.”
In practice, chatbots can be used for summarizing medical records, drafting post-visit summaries, collecting patient feedback, reminding about vaccinations, and performing administrative functions. “It’s really limitless as to what chatbots can do,” Dr. Cush said. “The question is, [what is] really going to help you?”
Much of the research submitted for presentation at major rheumatology meetings over the years has had questionable real-world utility and value, he said. But in the future this will likely change. “Take the PsA [psoriatic arthritis] patient who hasn’t responded to methotrexate or apremilast [Otezla]. There are [so many] choices, and there really isn’t a clear one. Shouldn’t data guide us on whether an IL-23 is better than a JAK, or maybe a JAK preferred over a TNF for some reason?” Dr. Cush said. “That’s what we’re hoping will happen down the line.”
More realistic AI-guided clinical scenarios for now include the following: AI screens the chart of a 68-year-old with RA on methotrexate and etanercept who is following up, and retrieves pieces of history — an elevated C-reactive protein 3 months ago, for instance, and diverticulosis 5 years ago. “AI tells you, based on this, he may have active disease, and here are three medications covered by his insurance,” Dr. Wells said.
Or, in the case of a 58-year-old patient with RA who has scheduled a virtual follow-up visit after having been on methotrexate and hydroxychloroquine for 12 weeks, AI detects a low platelet count in her previsit labs and also sees that she received an MMR booster 5 weeks ago at a local CVS Minute Clinic. AI retrieves for the rheumatologist a review article about thrombocytopenic purpura after MMR vaccination.
AI for Drug Development, Clinical Trials
Dr. Cush is following with keen interest the integration of AI into the process of drug development, from drug discovery and biomarker evaluation to clinical trial efficiency and patient recruitment, as well as marketing. “A lot hasn’t been ‘rolled out’ or shown to us, but there’s a lot going on … everyone is investing,” he said. “The number one challenge is regulatory: How will the [Food and Drug Administration] handle AI-generated data sets or AI-generated or monitored trials?”
The FDA is working to ensure quality and utility of data and is rapidly “approving AI algorithms for use in medicine and healthcare,” he said.
AI’s ability to identify patients in populations can not only facilitate earlier diagnoses but can accelerate patient recruitment for clinical trials, Dr. Cush emphasized. He pointed to research presented at the ACR 2021 annual meeting in which a machine-learning algorithm was used with electronic health records in the United Kingdom to estimate the probability of a patient’s being diagnosed with axial spondyloarthritis (axSpA).
AI identified 89 best clinical predictors (out of 820 analyzed). When applying these predictors to the population, AI was able to differentiate patients with axSpA from healthy controls with a sensitivity of 75%, a specificity of 96%, and a positive predictive value of 81%. Such an application of AI “is ideal … It would make clinical trials more streamlined and productive,” he said.
The extent to which AI will lead to cost savings — in the pharmacology arena, for instance, or for Well’s medical group — is unknown, Dr. Cush and Dr. Wells said. And, of course, there are concerns about potential bias and abuse of AI. “The worry,” Dr. Cush said, “is, who’s watching?”
Dr. Wells disclosed that he has research support and has served as a member of advisory boards and/or speaker bureaus for 17 different pharmaceutical or medical technology companies. Dr. Cush disclosed relationships with AbbVie, Amgen, Bristol-Myers Squibb, Novartis, Sanofi, and UCB.
FROM RWCS 2024