The demands on healthcare systems are at an all time high. As a result providers are increasingly turning to Artificial Intelligence (AI) and Machine Learning (ML) technologies in a bid to help alleviate some of the pressures they face, and to improve patient care.
The ongoing effects of the pandemic are placing huge pressures on healthcare delivery at a time when there is a growing need to deliver value-based care. This coupled with growing patient populations with chronic conditions and/or co-morbidities, staff and skills shortages, and increased strains on resources, are all resulting in healthcare providers increasingly looking towards AI technologies to fill the gaps in patient care. The opportunity is significant, outlooks for the sector expect the global AI healthcare market to grow from USD $4.9 billion in 2020 and reach USD $45.2 billion by 2026.
Technologies that can automate workflows and processes, and artificial intelligence solutions designed to assist diagnostic processes and help streamline treatment and care provision have become a crucial element of any care system. Seeing AI and ML as a strategic division within the delivery of patient care are now, more than ever, fundamental to successful healthcare provision.
Artificial intelligence simplifies the lives of patients, doctors and hospital administrators by performing tasks that are typically done by humans, but in less time and at a fraction of the cost. The applications are extensive, and already AI is being applied across a huge range of healthcare applications, from increasing the accuracy of detecting and diagnosing cancer, to tailoring treatment plans to meet the specific needs of individual patients.
The Impact of AI on Patient Care
The impact of AI on patient care was the topic of discussion for members of our HealthTech Networking Club at our recent monthly HealthTech RapidConf events.
Looking to the future optimistically, there are countless possibilities for the support of AI. Currently, machines are only as good as we understand biology to be. Therefore, the focus should be on enabling clinicians to do a better job and as a result improve care.
This concept was supported by panellist, Dr. Lynda Chin, who states that, “patient care is not an engineering problem, computers don’t treat patients”.
Many factors influence decision making, and generally, in medicine, not all information is complete therefore approaching it as an engineering problem, or a technical problem, is not as straightforward. In medicine, it is about bending rules, and human judgment and intuition are important.
Dr. Joanna Elmore adds that an area where AI can be very powerful and helpful in medicine is when there is variability. She uses a personal story of cancer screening which resulted in 3 different diagnoses. In healthcare, variability is a significant issue and delivering a consistent standard of care is problematic. For example, one pathologist could interpret a case of invasive melanoma in one way, but when presented with a similar case a year later, would they interpret that in the exact same way? Generally, the answer is no, and less than half the time they’ll offer a different diagnosis. AI on the other hand will consistently provide the same diagnosis, every time. It will be more reproducible!
Also, during the session, Dr. Anthony Chang tackled the negative presumptions about AI stemming from the incorrect forecast of AI replacing doctors as inaccurate. There is a need for conversational translation between data scientists and medical doctors. Clinicians can use AI to learn more about themselves and connect the dots by working backward to make breakthroughs in diagnoses.
This is a sentiment echoed by Dr. Chin. AI will make the jobs of cardiologists and radiologists more meaningful and productive by reducing the burden placed upon them. For example, a task such as reading an EKG study, could ideally be tasked to an AI application. The ability for the technology to apply an extensive library of learned experience to the reading, means that it is well provisioned to identify, and flag, abnormalities quickly and more accurately. The time saved by automating such tasks, means that physicians are empowered to use their time in more effective ways and ultimately provide a much higher overall standard of care.
How AI can Improve Patient Care
There is an emphasis on AI levelling the playing field for patient care which will improve overall outcomes. AI should help physicians make data-driven decisions, with decision support.
Dr. Chang highlights where machine learning has been applied to covid testing to improve processing rates. In one recent study, AI was able to detect Covid 26% faster than lateral flow tests.
He also cautions that we must focus on adopting AI for the long-term with a vision of improving things for the next generation of clinicians, while still supporting this generation of clinicians. One aspect of this is to use machine learning to address current biases which exist within healthcare data. In the future, the requirement to deliver equitable care can be made easier by AI and ML.
According to Balint Bene, CEO of specialist HealthTech accelerator Bene : Studio, AI should be used in ways that enhance the patient experience, improve physician time management and leverage tools to gain efficiencies – let doctors be doctors!
Want to be involved in future discussions of our HealthTech Networking club?
The invite-only networking club connects leading HealthTech companies, investors, healthcare providers and other key experts/players in the Healthtech industry providing private networking opportunities.
The networking club is free to join simply apply for an invite here.
