Building the First FDA Approved Autonomous AI in Healthcare –

Screening for early diabetic retinopathy has become more important as diagnostic testing and early treatment of diabetic macular edema is now possible with the injection of intravitreal anti-VEGF compounds as well as long-lasting intravitreal anti-VEGF inserts.

In people with diabetes, too much blood sugar can damage the tiny blood vessels at the back inner wall of the eye (retina) or block them completely. This condition is diabetic retinopathy. Sometimes, tiny bulges (microaneurysms) protrude from the vessel walls, leaking or oozing fluid and blood into the retina. This fluid can cause swelling (edema) in the central part of the retina (macula). This is a serious eye complication called diabetic macular edema that can cause vision problems or blindness.

Primary care physicians often encounter patients with diabetes and refer them to an ophthalmologist  (vitreoretinal specialist) for diagnosis and treatment.  Many patients are non-compliant with seeing an ophthalmologist, especially if they are not symptomatic. Decreases in visual acuity can be subtle and if binocular a patient has no reference to a decrease in visual acuity.  If the macular edema is present in one eye a patient becomes aware and will seek treatment. The provider will most likely be a physician or advanced practice (nurse practitioner).    Some managed care plans require patients to be seen by primary care (family practice or internal medicine).  This can cause serious delays and lead to more serious problems.

Many managed care programs measure the compliance of diabetic patients. Do they comply with a referral to an ophthalmologist?  The rating of health plans is measured by compliance rates for many diseases, including diabetes. Because of advances in electronic health records and data analytics health plans monitor if patients are non-compliant.  Health plans follow up with their members and notify them to see a competent specialist.  If they do not then home testing is available by HealPros with retinal fundus cameras. Fundus photos are obtained digitally and transmitted directly to an ophthalmologist for analysis and reporting.  Until now imaging is analyzed by an ophthalmologist.  The rate of positive findings is fairly low.  The addition of artificial intelligence can be used to screen normals, and refer positive findings to the ophthalmologist.  This allows greater efficiency and lower cost.

FDA permits marketing of artificial intelligence-based device to detect certain diabetes-related eye problems

Building the First FDA-Approved Autonomous AI in Healthcare – Dr. Michael Abramoff and Seth Rainford

Resource Center: Ambient Clinical Intelligence

Non-disclaimer notice:  Most of this article was provided by Nuance

Experience better telehealth visits

Sub-Optimal Patient-Physician Experiences Negatively Impact Everyone 

How this AI Solution Saves 7 Minutes Per Patient Encounter 

Better patient-physician experience, all around

Why Use Medical Scribes When You Don’t Have To

Cooper University Health improves operational efficiency with the Dragon Ambient eXperience (DAX

How Artificial Intelligence Drives Value for Patients and Providers 

What is Ambient Clinical Intelligence? 

Heading off the physician shortage: The role ambient clinical intelligence must play

Leveraging AI to Relieve Physician Stress in Turbulent Times 

The University of Michigan Health-West Restores the Joy of Medicine with Ambient Clinical Intelligence 

Expanding the Role of AI Technology in Healthcare 

The adoption of artificial intelligence in healthcare is beginning. The topic is a complex area with input required from siloes of knowledge spanning data science, programming, and clinical areas. Clinicians, clinical information systems, including electronic health records, and governmental regulatory bodies (including CMS, Medi-Cal, and Public health organizations will be required to communicate with interoperability standards, much like the health information exchanges designed to merge disparate systems for accessing electronic patient information.  The task will be formidable and take considerable time until it is mature, perhaps as long as ten years.. Experience with electronic health record interoperability demonstrates it has taken over ten years to become well established.

EHR interoperability has expanded in several dimensions, proprietary systems such as EPIC, and CERNER provides an interface amongst institutions that utilize their electronic health record system.  

 

Resource Center: Ambient Clinical Intelligence

Artificial intelligence and metaverse: future implications for clinical medicine and healthcare | MI10

How are artificial intelligence and metaverse related?”

The following are my thoughts, starting with an explanation of just what metaverse is purported to be, and followed by just how artificial intelligence will be interacting with metaverse.

Metaverse

Metaverse is a convergence of the real and virtual worlds made possible with technologies of extended reality (augmented and virtual types), artificial intelligence, and blockchain. The origin of the metaverse is from the 1992 science fiction novel Snow Crash in which the terms “meta” and “universe” are combined. Metaverse is perhaps a paradigm shift in just how we interact with technology (rather than any type of technology), but the true definition of metaverse remains a bit vague, reminiscent of the early discussion of the internet in the 1970s.

The metaverse is, in short, an immersive virtual world (VR, AR, and even social media) that can accommodate a person to be a persona in a myriad of networks. While the metaverse has been deployed mainly in gaming and social connection up to now, its potential use in entertainment, business, education, and training in the future is very high. Examples of early entities in the metaverse realm include the role-playing game World of Warcraft and the more recent survival game Fortnite, while the most recent entries into this domain include Microsoft, Nvidia, and Facebook (the latter even rebranded to Meta). One significant limitation of the metaverse is that the VR experience often creates a motion sickness problem for the user.

Artificial intelligence and metaverse

Artificial intelligence in the form of self-supervised learning will most likely play key roles in the portfolio of metaverse architectural elements:

• Language
  The convening of players in the metaverse will rely heavily on language processing, so artificial intelligence in the form of natural language processing will be extremely useful for this element.

• Avatars
  The players are represented by avatars, which are essentially 2D and 3D images created by artificial intelligence. As the user experience is a key part of the metaverse, artificial intelligence will be vital to improving the overall experience.

• Data
  The metaverse is also supported by data of all types in the players. The curation and management of data will be enabled by artificial intelligence in the form of machine and deep learning.

The bottom line is that there is an important role for artificial intelligence, not only in the metaverse but also in the extended reality portfolio of tools. The potential of metaverse use in clinical medicine and healthcare is virtually (excuse the unintended pun) limitless: medical education, clinical training, patient education, and performing procedures.

In short, artificial intelligence is an integral part of the metaverse and will render the entire visual experience more real and fluid to the user, including clinicians and patients with their families.

Anthony Chang M.D.

Artificial intelligence and metaverse: future implications for clinical medicine and healthcare | MI10

MI10.Updates on AI in Medicine

What is your MIQ?

Most physicians are dimly aware of artificial intelligence (machine learning) but are illiterate in terms of accessing and/or using it for their clinical practices.

Information is siloed between Data science, technical experts, and clinicians' workspace.

How does one assess how to use AI for clinical practice?  That is a large and open question. Fortunately, a roadmap has been developed by knowledgeable experts in the field.

Overview

MI10 is a healthcare artificial intelligence consulting firm dedicated to helping healthcare stakeholders efficiently implement actionable AI solutions.

MI10 has developed a revolutionary metric to define an organization’s readiness to enter into an AI/Healthcare partnership: the MIQ™. Using this metric, they enable both technology companies and healthcare providers to implement AI solutions more effectively.

The MIQ™ of a fictitious healthcare organization scored an MIQ™ of 54 points (machine or technology = 22; human = 20; and both = 12) and the results are displayed in a spider diagram (pictured).

This MIQ™ score of 54 points (again, out of a total of 100 points) will place this healthcare organization in the “AI Ready” stage category of the AI capability scale.

Your Organization's Roadmap to an AI-Enabled Future

What is your MIQTM?

MI10