How Artificial Intelligence will affect cardiology
Artificial Intelligence (AI) is becoming more and more widely spread in the medical industry, as well as many others. Seminars over recent years have all spoken about this subject prominently. This used to be something that we only saw in films and in dubious studies, but now we’ve seen this being developed and accepted by the Food and Drug Administration (FDA). Industries are using it when looking at data and interpreting this into back end IT processes to support the speed of information available, this is no different in the medicinal industry where it is already being utilized without us knowing.
As this is such a new aspect of technology, it will take a while for clinicians to trust, the validity of the information it is perceived to provide is something still to accept. Times are changing and the way AI can help the business side of things as well as patient care is becoming evident in the industry.
Cardiology development by the ACC
John Rumsfield, M.D., Ph. D., FACC, (American College Cardiology (ACC) chief innovation officer and professor of medicine at the University of Colorado School of Medicine) explains, “We have a huge gap between all this AI investment and how we actually take care of patients. We need to integrate it into our care because if it is not part of how we take care of patients, this isn’t going to work.” “The clinical evidence needs to be there, and right now there is way more hype for artificial intelligence. We need to build that evidence and we also need alignment with our payment models.”
He went on to explain that if AI is going to be accepted and utilised, then there is a requirement to review the system most hospitals use as their current cardiac care system, the compensation system is one of these. Unfortunately, so far, the changes required within cardiology care have been minimal in order to fully utilise AI and take on the investment considered by the healthcare system. Movements have been made by the ACC project entitled the “Roadmap for Innovation,” which Rumsfield is fully involved, a project that is attempting to partner with AI providers directly. The ACC has also tried to push applications for the ACC National Cardiovascular Data Registry (NCDR) in order to support the launch of AI within cardiac care.
By the coming together of the ACC with AI companies there have been technologies invented that will absolutely benefit the industry. Wearable devices that operate directly with an electronic medical record (EMR) have already been developed, in order to review patients at any time and integrate this with smartphones as well. These could be especially useful when working with patients who have suffered heart failure, in order to gain far more useful information that can expand the EMR. The administration behind such recordings can be reduced and therefore staff may be able to better spend their time on direct treatment.
“What I am hoping is that we can get from where we are today to actually lead the digital transformation in healthcare,” Rumsfeld explained.
Universities and the ACC have the ability to source huge amounts of data from its NCDR, data which will absolutely support AI. Trials are happening between Yale University and ACC to review and extract NCDR data alongside AI to see what conclusions can be reached. Rumsfield says, “The best technology is what they call ‘quiet technology,’ you don’t even know it’s there and it makes you more efficient and does a task. In the non-healthcare world, we don’t even realize when AI is being used.” AI is around us everywhere, it’s able to support traffic congestion by analysis of big data so it can provide live travel times or alternative routes, it’s the thing that is able to give you an accurate weather forecast on your smartphone, it is the aspect that powers online search engines like Google.
“As cardiologists, we are comfortable with advances in technology, but we need it to be efficient, useful, and make our lives better and not worse like many would argue the EMR has done, and it really should be in the background. But if it is going to run in the background, we need to know it has been clinically validated and that we have shown it is safe, effective, and actually does what you need it to do,” Rumsfeld said.
Complicated illnesses will benefit from the Big Data that AI can provide
Anthony Chang, M.D., chief artificial intelligence officer, Children’s Hospital of Orange County (CHOC) talks about complex analytical software that is also known as machine learning, which stems from AI. Anthony is a founder of AIMed, this is a company that is trying to enable links between the hospitals that are interested in AI and various companies that are able to collaborate in this area. Deep learning – the newest solution from AI is the ability to resolve issues that appear more intricate, machine learning can only provide so much before more is required.
This new “deep learning” technology works like the human brain, able to understand what errors have been made previously and take actions gained from those experiences. Patients with more complicated problems will benefit most from the technology that AI can bring.
“Deep learning is harder to do and it requires a lot more data, but I think the dividends for particularly complex situations are going to be much bigger than anything we have seen before,” Change said.
Patients who have intricate illnesses may be prescribed various medicines from physicians of various expertise, the variety of which may interact with one another. There may also be a history of treatments and procedures carried out previously that may affect the results. Deep learning systems should be able to review the data of patients, historical and current, looking at all results from ECGs, radiology reports, lab results, etc., and be able to offer solutions based on current regulations or relevant information from recent studies.
Chang advises that deep learning can take on a lot of data where you know the results but let the computer sort through it. It can then, using AI capabilities, be able to determine new ways to identify which patients are at most risk.
“This is a paradigm because we do not set the rules, we let the computers figure it out, and then we take that algorithm developed by the computer and apply it to new patients and see if it makes sense,” Chang explained. “We should not let the algorithms be the only solution, there should be cognition as well, where there can be a collaboration between humans and the machine to get the best outcome.”
Within the realm of cardiology, we know aspects are far more complex than what is simply seen. Initially, AI apps that have been developed simply use medical images to support the identification of radiological results. The way these algorithms work is unfortunately no smarter than for example asking a computer to distinguish a person or an animal from lots of images and photos.
“Cardiology is one of the best fields to use AI because it has this set of problems, like complex patients, the need for decision support, wearable technology, and the AI needed for that. Cardiology certainly has the portfolio of problems with the solutions that can be engendered by Artificial Intelligence,” Chang said.
The use of AI in cardiology now
AI already has its uses in the industry of medical imaging and cardiology. There are commercially available options of AI such as ejection fraction (EF) calculations and point-of-care ultrasound systems (POCUS), for example, the GE Healthcare Vscan. More upmarket cardiac systems such as the Philips Epiq and the Siemens SC2000 are able to use AI to detect parts of the anatomy, defining and labeling it, identifying the optimum views and measurements before the physician can view the case.
There are software providers that have AI-automated calcium scoring software for cardiac CT scans. This software is able to quantify the information very quickly and create reports that colour code by vessel section on the dataset. GE healthcare and Canon have recently both received FDA approval for AI-based CT iterative reconstruction formulas that can analyse quality images from CT scans that may not be clear. Siemens is also able to optimize CT scans, by the isocenter on the scanning bed.
AI is already being utilised in order to speed up examination times and enable numerous images to be developed from one scan, thus minimizing the time needed inside the MRI scanner, so that increased scans of patients can be carried out each day. Arterys AI-based cardiac MRI software is able to speed up the processing from the examination by automating the data.
Sanjaya Gupta M.D. (electrophysiologist at St. Luke’s Mid America Heart Institute) has been part of a team that has invented an app that uses AI in order to define the risk of atrial fibrillation (AFib) patients. It is able to filter patients into those who need anticoagulation, those who do not, and those who are contenders for left atrial appendage (LAA) occlusion, an amazing feat. This app has developed the ability to select patients for LAA occlusion procedures. Gupta said “Most importantly, this helps us identify those patients that we did not realize had a problem, that’s what is really key. This also helps us identify risky patients during their regular clinic visits and allow us to call and intervene. That is really where the next level of this is going and will make a big impact on patient care quality,”
Wearables and smartphones, again, are playing a part in recording ECG and making patients aware of arrhythmias from the AI algorithms that are able to detect and send notifications. The Apple Watch and the Kardia Alivecor are already utilizing this. Wearables and apps are just the start of cardiac monitoring, there will progressively be more progress in the point-of-care options for patients from what AI can offer. Patients that are at-risk will be identified easier, therefore a human cardiological physician can detect and therefore provide aid sooner.