With AI revolutionizing all industries, it was only a matter of time before the same happened to healthcare. And if there’s one area where artificial intelligence could bring life-changing advancements, it is healthcare.
BBC correspondents speak to three Oxford-based scientists working in academia and the commercial world about using AI in the medical field.
“After years in the laboratory, the technology is already entering hospitals to assist doctors in diagnosing and treating strokes,” said Michalis Papadakis, chief executive of Brainomix.
Papadakis continued his statement, saying, “There’s a great potential to change outcomes for stroke patients: “After every 30 minutes, a stroke patient who could have been saved dies or stays permanently disabled, not due to the stroke, but due to the reason that they’re admitted in a hospital that doesn’t have enough expertise to diagnose and pick the real life-saving treatment path.”
Brainomix’s chief executive is collecting data from brain scans to enable AI to detect what happens when someone has a stroke. His work, which uses data from thousands of brain scans, helps doctors interpret images and generates quick decisions when it is suspected that a patient has undergone a stroke.
“The outcomes have provided that the sensitivity of healthcare providers increases significantly with the use of AI tools in diagnosing and detecting stroke damage on the scans compared to how they’re doing it manually,” Michalis Papadakis said.
“At the time, the biggest challenge is getting doctors to adopt this new way of working through AI technology,” Papadakis further explains the scenario for AI in health.
For more than ten years into a project that again has depended on gathering a vast store of data, Antoniya Georgieva, the Scientific Director at Oxford Centre for Fetal Monitoring, is still some way off. Her research involves monitoring babies in the womb, and she started it from fetal monitoring units in labor wards in Oxford dating back to 1993.
According to Georgieva, her study aims to get a clear understanding of “how to automate the process where medical consultants and midwives can be alerted to the unusual cases where a baby is experiencing breathing concerns and could end up receiving brain damage.
“We are taking anxiety from having to look at complex graphs, complex stats that take time, that humans are just not so good at,” says Antoniya Georgieva while explaining the vision & mission of her research work.
“The new technology-supported system will monitor those graphs and tell the human specialists exactly when they need to be worried,” said Georgieva.
Having been observed in a hospital for ten years and learning as much as she can about medicine, Dr. Georgieva is not rushing to turn her research into a commercial venture. She implies just how multifaceted the interaction between her algorithms and the bulk of physiological data is and how long it takes to be clear that the AI system is generating some better outcomes.
Even earlier on the path from AI ideas to real-world developments is the effort of Charlotte Deane, Professor of Structural Bioinformatics at Oxford University. She, along with her doctoral students, is focusing on the problem of applying artificial intelligence in drug discovery, a lengthy and expensive trial of creating new medical products.
“Even if AI could remove 10% of the cost of a new drug, which costs as much as $2 billion and takes as much as 30 years, that would be a considerable gain,” says Pro. Deane.
According to Deane, the hype around AI isn’t helpful, but progress is being made: “The tasks are huge, as biological data is noisy and complex to deal with. However, the algorithms we now have are starting to deliver us the possibility to make a real difference.”
Things are going quite nicely for Deane’s research, except for one problem: Her doctoral students continue getting hired by the biotech firms clustered around Oxford. Still, the determined powerhouses keep putting their efforts into the global application of AI in healthcare.
Most recent research in the field of AI in Medical Diagnosis:
In recent times, Business Insider Intelligence has examined the value of AI applications in three high-value areas of medical diagnosis—imaging and interpretation, clinical decision assistance, and personalized medicine—to demonstrate how the technology can drastically improve patients’ outcomes, lower expenses, and enhance productivity.
For this, researchers from Business Insider remain focused on US healthcare systems that have effectively applied AI in these use cases to validate where and how medical specialists should implement AI technology.
Finally, they examined how a frontline US health system authenticates AI partners and internally establishes its AI strategy to offer health provider organizations a template for AI innovation.
The major companies included in this research are Arterys, Boston Gene, Allscripts, Cabell Huntington Hospital, Epic, Google, HCA Healthcare, Geisinger Health System, IBM, IDX, ICAD, Oak Street Health, Tempus, Stanford University, Verily, UCI Health System, and Yale-New Haven Hospital.
Here are some of the key takeaways from the report:
– Use of AI in interpreting imaging, clinical decision support, and precision medicine across hospitals
– Those US hospitals who haven’t applied AI are missing the tech’s gain until they don’t develop an effective AI strategy
– Those who already have implemented AI reaping the reward in medical diagnosis through driving efficiency, reducing costs, and increasing overall outcomes
– Strategies enlighten the need for an AI strategy fortified by talent, data accumulation techniques, and cooperative associations with external vendors.