What does the future hold for AI in healthcare?

Can you imagine a future where kids wear smart clothes to track their every move? It might sound like something out of sci-fi, but the romper jumpsuit being piloted in Helsinki, Copenhagen, and Pisa does just that.

The Kinematic Assessment of an Infant Suit (MAIJU) looks like typical baby clothes, but there is a fundamental difference – it is filled with sensors that assess a child’s development.

“MAIJU provides the first-of-its-kind quantitative assessment of infants’ motor abilities through the lifespan from lying on their back to fluent walking,” explains Professor Sampsa Vanhatalo, Head of the Project at the University of Helsinki. Such quantification has never been possible anywhere, not even in hospitals. Here, we present the solution for homes, which provides the only environmentally relevant context for an engine assessment.”

Vanhatalou describes the path from wishful thinking about a solution to potential clinical implementation as a “stormy road.”

“There is no shortage of dreams or technology, but we lack adequate and appropriate clinical problem data, contextually and environment-relevant data sets, and reliable clinical phenotypes of the substance, as well as appropriate legislation for products that do not follow traditional forms,” he says.

Machine learning allowed researchers in Helsinki to find latent properties in infant movement cues that could not be determined by traditional inferential planning.

“At the same time, we have to remember that AI in medical applications can only be as smart as we allow it,” Vanhatalu adds. “Real-world situations are more muddy than we hope, and the ambiguity of many clinical cases or diagnoses greatly limits our chance of building as accurate AI solutions as we hope. For example, it is not possible to train and validate a classifier for the myriad medical diagnoses that It has no clear boundaries.”

Vanhatalo also believes that the medical community needs to recognize the reasonable goals of AI.

“It is much more rewarding to train clinical decision support systems (CDSS) than training clinical decision systems,” he says. The latter is what some hope and others fear. But the responsibilities, including legal, of decisions are so great that I find it hard to see any company that would dare market such solutions. Indeed, I can already see how the legal risks arising from such obligations, even indirect or fictitious, create a bottleneck for the marketing of many good AI products.”

The vanguard of oncology

Oncology is one area of ​​medicine where artificial intelligence has great potential to revolutionize care. Professor Carol Sikora, Chief Medical Officer (CMO) at the forefront of cancer care, Rutherford HouseHe believes that machine learning can benefit clinicians by aiding in complex treatment decisions.

“A range of commercial solutions are available to identify and map nearby organs at high risk of cancer,” Sikora explains. “Accurate oncology requires the analysis of large amounts of data in an unprecedented way and we hope that AI will provide long-term benefit to the patient.”

The network of oncology centers at Rutherford Health uses the latest innovations in cancer technology, such as artificial intelligence to plan radiotherapy.

According to Sikora, machine learning could be of great benefit in improving patient selection in the future. “Artificial intelligence can drive patient understanding of the risk-benefit equation of any intervention,” he says.

Simplify artificial intelligence

But for healthcare organizations to fully de-capitalize the potential of AI, there is a need to demystify the “hype” surrounding it, according to Atif Chughtai, senior director of global healthcare and life sciences business at Software company Red Hat.

“Artificial intelligence properly applied has huge potential to save lives and manage the increasing cost of healthcare,” Shogtai says. “In the future, AI will continue to evolve and will be widely used as an assistive technology to perform tasks more accurately and efficiently with humans in the loop to make final decisions.”

He adds that in order to successfully adopt AI capabilities, organizations must introduce change at a manageable pace and work collaboratively to innovate in intelligent business processes.

“Too often, as data scientists or IT professionals, we don’t take the time to understand our clients’ business process which leads to mismanagement of change,” he says.

Vanhatalo, Sikora and Chaughtai will be speaking at the session today Unlocking the future of artificial intelligence In the HIMSS22 European Health Conference and Exhibition, taking place 14-16 June 2022.