Artificial intelligence could make healthcare cheaper, faster and more accessible. And help with worker shortages.

Artificial intelligence could make healthcare cheaper, faster and more accessible. And help with worker shortages.

AI could come to the rescue in a world where doctors and health practitioners are scarce.

The World Health Organization estimates there will be a shortfall of 10 million health workers by the end of the decade, mostly in low and lower-middle income countries.

But even developed countries like Australia struggle. The nation is under immense pressure with a critical shortage of workers, and an ageing population that is seeing demand for healthcare professionals rise.

According to RWC Medical , within the next three years there will be a 15% surge in demand for medical services, yet the industry is projected to be short of more than 100,000 workers by 2025.

Dr Aengus Tran — a medical doctor and world-ranked Al engineer and data scientist — thinks his startup Harrison.ai can help.

The innovative startup made LinkedIn Australia’s Top Startups list for 2023. You can check out the other companies that made the top 20 here.

Harrison.ai is a clinician-led healthcare company that uses AI to help healthcare professionals diagnose diseases, design treatments and analyse medical images.

Here, he speaks to LinkedIn News Australia about how he sees the role of AI in the healthcare sector.

Aengus, why did you and Dimitry Tran start Harrison.ai?

We are dedicated to addressing the most critical issues in our healthcare system — inequality and capacity limitations.

Many areas of clinical diagnosis and treatment are struggling to keep up with the rising demand for healthcare services, which is compounded by skill shortages.

To address these challenges, we build AI tools that speed up diagnosis, improve patient experience and reduce costs. This supports clinicians by giving them a second set of eyes, enabling them to be more consistently accurate and efficient than ever before.

Dr Aengus Tran

The company has evolved with focuses on two particular healthcare pathways — radiology and pathology. Can you explain how you're using AI in those fields?

Our mission has always been to scale global healthcare capacity through AI automation, resulting in the creation of our two solutions Annalise.ai and Franklin.ai.

Annalise.ai bring the power of AI to radiology, in partnership with I-MED Radiology Network . The first product to come from this partnership was the world’s most comprehensive AI clinical decision-support solution for chest X-rays. It is already in use by more than 350 radiologists in Australia every day and we are rolling it out to hundreds more.

More recently, we partnered with Sonic Healthcare to form Franklin.ai to develop AI-powered tools for pathology. Franklin.ai combines Harrison.ai’s clinically-led AI platform with Sonic Healthcare’s deep clinical expertise and global reach. The aim here is to support a more effective and efficient diagnosis of patients.

What illnesses or diseases can AI be used to detect? What are you currently able to help detect, and what do you think your technology can do in the future?

AI has actually been applied in healthcare for a while, but its remit has been very narrow, and only able to pick up one thing at a time. For example, one chest X-ray would need two different AI applications to detect broken ribs or pneumonia, and multiple others to pick up everything that is needed to be checked off.

Where Harrison.ai is different is that we are pioneering an A-Z solution which is called comprehensive AI.

In practice, this means that our Annalise Enterprise CTB (brain scans) can detect up to 130 radiological findings and Annalise Enterprise CXR (chest X-rays) can detect up to 124 findings.

Our technology is designed to support clinicians, so we will continue to build solutions that alleviate capacity pressure where needed.


?? Harrison.ai is one of LinkedIn's Top Startups for 2023. What other Australian startups inspire or impress you, and why? Create your own post sharing your thoughts by using this link.


What is the greater benefit of AI in healthcare? Is it being able to detect more illnesses and diseases?

AI in healthcare has the potential to do all these things, but our primary focus is improving accessibility. Clinicians can use AI to reduce time looking over scans and imaging, giving them more time to see patients.

Also, patients are typically addressed on a first-come-first-served basis, but this means that those who require urgent medical attention may not receive treatment in a timely manner. AI offers the ability to detect and triage patients efficiently, offering better standards of care to those who need it most.

We ultimately see a world where healthcare also becomes more affordable as a result of AI.
Co-Founders Dimitry Tran and Dr Aengus Tran with Sonic Healthcare CEO Dr Colin Goldschmidt. Supplied: Harrison.ai
Co-Founders Dimitry Tran and Dr Aengus Tran with Sonic Healthcare CEO Dr Colin Goldschmidt. Supplied: Harrison.ai

Will there come a time when we don’t need doctors? How will AI change the way medical professionals work?

Doctors and clinicians are and will always be in the driver's seat of medical technology developments. The role of AI tools is to augment the capabilities of clinicians rather than replace them. For us, we use AI coupled with human intelligence to enhance healthcare.

What are the dangers of AI in healthcare, or ethical issues? How do you ensure those lines aren’t crossed?

As with any industry, introducing new solutions comes with risk, but you can't afford risk in healthcare. As a result, the health tech industry moves a lot slower and is a lot more regulated to make sure we get things right —?and rightly so.

I think the bigger risk is around change management.?The risk exists if you invent AI tools and try to force them into an existing workflow with which it doesn't fit.

In this case, you fail to create any value. We solve this by working closely with clinicians.

Another potential risk with AI in healthcare is bias. If tools fail to be trained on diverse learnings they fail to pick up diverse understandings of disease and diagnosis. To mitigate bias and ensure fairness in healthcare decision-making, we start at a foundational level, training our AI algorithms on diverse and representative datasets.?

Do patients need to be worried about AI using or storing their medical data? Is it safe?

As a heavily regulated industry, any form of technology in healthcare has to go through stringent processes to be used on and with patients. For example, all our devices are classed as medical devices and have to go through many tests and regulations to ensure they are both useable and ethical.

This means we have to graduate with the same approvals as your COVID tests, pacemakers and the like. At Harrison.ai, we work collaboratively with hospitals and healthcare systems to ensure that our systems work within the correct patient data and regulatory framework.

Big tech companies like Google, Microsoft (LinkedIn's parent company) and Amazon are making inroads into AI healthcare. What does this say about the potential of the sector?

All of these companies have been involved in AI for a long time, but with the increase in AI capability, we are certainly seeing greater innovation investments being made across the board.

The rising investments in the industry are indicative of the value that AI and technology have in creating a more equitable future for our growing and ageing population. People are recognising the impact that healthtech, and specifically AI, will have on our industry and lives, for the better.

How do you find skilled workers? Is it easy to find talent trained with the AI skills you need?

Our team combines medical and engineering knowledge, with expertise in health management and artificial intelligence to continuously explore and drive innovation in the field of medical imaging.

We work very closely with our people and culture team to help find people with the right skills to help us achieve our goals. We’re very fortunate to work with incredible talent who in turn attract incredible talent!

What advice would you give to someone wanting to get a job at an AI startup?

Have a passion for the problem you’re looking to solve.

AI is a growing industry and there are many opportunities, but pick an AI startup that you believe in, and help to solve a problem you care about.

It will not only help you stand out, but will increase your happiness tenfold!

?? Harrison.ai is one of Linkedin Australia’s Top Startups for 2023 — a list of 20 young companies that are rising to the challenges of the moment by innovating and attracting top talent in 2023. Check out the full list here.

?? What Australian startups inspire or impress you, and why? Create your own post to share with your network here.


? Enjoyed this interview? Click here to subscribe?to the latest tech and startup news and analysis on LinkedIn.



Annette Tavitian

Narcissistic Abuse Recovery Counselling |Adult Childhood Trauma |

1 年

What an interesting concept!

回复
???????????????? ????????????????

???????????????? ???????????????? ?????????? ????????????... ?? ?????????????????? ???????? ?????????? ?????????? 2018! ???????? ????????, ?????????? ??????????, ???????? ????????. ?????????????????

1 年

My take on this is that provided it is responsibly and judiciously implemented, AI integration into healthcare systems holds exciting potential to make quality medical care faster, more cost-efficient and widespread over the coming years through its amazing problem-solving and decision-support capabilities. Also, freed from tedious data entry and exchanges of nonclinical information, medical professionals can focus more on direct patient care. Additionally, AI-guided triaging of symptoms and conditions could make initial care options more affordable and conveniently available, particularly in rural areas short on specialists. It's clear that AI holds exciting potential to make quality medical care faster, more cost-efficient and widespread over the coming years through its amazing problem-solving and decision-support capabilities by already being able to assist in analyzing massive datasets and medical records far beyond any individual's cognitive limitations.

Jason Clauss

???????? Founder of XenBuild - THE product visualization tool | UX/product designer

1 年

A robot couldn’t be any less effective than the average clock punching pharmastooge working at a typical McClinic.

Amanda Johnstone

Founding CEO @ Transhuman. TIME Next Gen Leader. LinkedIn Top Voice in AI. International Keynote Speaker

1 年

Great interview Marty McCarthy, you didn’t shy away from asking the hard questions. I am particularly fond of how their AI can be used to analyse medical images. I’m also excited for the future of EmotionAI and how medical professionals can understand the psycho-physiological state of the body and mind in real time to optimise patient care. Especially when the patient is unable to advocate for themselves.

要查看或添加评论,请登录

社区洞察

其他会员也浏览了