Smarter than you think

Artificial intelligence is a branch of computer science that involves performing tasks that typically require human intelligence.

It has been around for almost 50 years now: advertising algorithms, web search, digital personal assistants, machine translators, and self-driving cars are just a few examples of what AI can do.

And in the healthcare space, the use of AI within the Internet of Medical Things (IoMT) is already having a major impact on people’s lives.

A great example is the work of Dr Kristina Lång, based at the Unilabs Mammography Unit at Skåne University Hospital in Malmö. She is looking into how AI can improve breast cancer screening in the world´s first randomised AI screening trial, involving 100,000 women.

“We hypothesise that the “interval cancer” and false positive rates will be reduced with AI-integrated mammography screening, along with a considerable reduction in the screen-reading workload,” she says. Another example is how, thanks to AI, health apps are encouraging good behaviour, “gamifying” exercise, and helping people lead healthy lives.

AI also allows healthcare professionals to better understand the day-to-day patterns of the people under their care. And with that understanding, they can provide better support. For instance, AI helps them to identify missed findings, improving the quality of their diagnostics. It also makes them more efficient, since it takes the more burdensome tasks off their plates.

AI is already being used to detect diseases such as cancer more accurately - and crucially, at earlier stages. An example: at Unilabs we are involved in a trial that aims to establish whether AI can improve mammography screening by adapting single and double reading based on AI-derived cancer-risk scores, and using AI as detection support in the screen reading. The expectation is that both false-positives and false negatives can be reduced with AI-integrated mammography screening, and that the technology will also lead to a considerable reduction in the screen-reading workload.

We are also partnering with external parties to implement AI-powered solutions. For example, we are using a state of the art platform that uses algorithms developed with advanced deep learning and it’s trained to detect prostate cancer. The algorithms automatically analyse images from tissue biopsies, providing insights to the pathologists diagnosing the case, whose assessment is vital for reaching correct treatment decisions by oncologists. These AI-generated insights include case prioritisation worklists, cancer heatmaps, tumour grading and measurement, streamlined reporting tools and lots more.

These tools benefit both the quality and efficiency of our work during the review and reporting process, but there’s much more to it. AI can help us every step of the way, from data capture to report sign-off. The so-called “workflow” solutions help us in ways that aren’t always visible. But we definitely feel the effects. At the moment, we are testing a tool for MRI in Sweden. This solution can help us improve the quality of images scanned with fast protocols and bring it to a standard-of-care level. This way, we can perform scans much faster with the same or even better quality.

Furthermore, Unilabs Portugal and a local leading software company developed an AI tool for clinical diagnosis. Since March this year, we have been running a pilot for the detection of the helicobacter pylori bacteria. So far, this tool has allowed us to process more than 2,000 samples. Once the pilot is over, the idea is to implement this platform across the rest of our network. 

AI is also related to deep learning, which involves predictive medicine: essentially, predicting what treatment protocols are most likely to succeed based on various patient attributes and the treatment complex.

The next big step is pattern recognition and human-like analysis. Pattern recognition is already used in radiomics, with initial results suggesting it can detect clinically relevant features beyond what the human eye can perceive. That could result in faster and better diagnoses, and thus better care for our patients, which is what it’s all about. 

Unilabs strongly believes in the added value that AI solutions can deliver. That’s why we are pioneers in digital pathology and radiology, with a whole department – the AI Centre of Excellence – dedicated to identifying the highest-value use cases for AI solutions. We want AI to help us solve our professional needs. 

And we are only at the beginning of this change - this is just the start. 

The future will be smarter than you think.