Dr Harro Stokman schreef een stuk op de Engelse pagina Security Magazine over het gebruik van Artificiële Intelligentie in de gezondheidszorg. Lees de originele versie van het stuk hieronder!
AI moves beyond physical security to patient care
Anxiety around artificial intelligence is at all-time high. Whether its concerns around how “The Algorithm” is shaping what we see and engage with on social media, or how facial recognition might end the idea of privacy as we know it, the technology is entwined in some of the most important questions facing society at large. But AI is not a monolith. Its use cases are as varied as there are jobs that exist in a given sector, and balancing its benefits against its costs is becoming an important equation in a lot of sectors that are embracing the technology.
The healthcare and wellbeing space is an industry grappling with this more than most. You’ll most likely have heard of AI in healthcare in the context of an algorithm correctly diagnosing a disease at a greater rate of accuracy than doctors, like this story from last year. But this is only the tip of the iceberg when it comes to AI in healthcare, with the technology being applied to diagnosing patients, monitoring their wellbeing, and even assisting with their mental health. These all take pressure off of medical professionals, freeing them up to focus on the core aspect of their roles – the actual providing of care.
Artificial intelligence can go beyond a simple yes/no diagnosis. It can also serve as an early warning system to detect health issues before they occur. While medical workers or care staff can look at a patient’s data to identify the likely cause of a problem once it presents symptoms, AI analysis of a patient’s medical history can offer a more proactive approach. Data such as lung sounds, and heart rate and oxygen saturation can be measured and tracked over time, with artificial intelligence and machine learning algorithms used to detect any deterioration before it becomes an issue. This allows for a more proactive rather than reactive approach to patient’s day to day health.
Beyond diagnosis, AI is commonly used in monitoring. Similar to how AI is used in facial recognition software found in security systems, the same “computer vision” technology can be used to identify a patient’s movements and body positions, and alert staff when patients need assistance. Here at Kepler Vision our “Kepler Night Nurse” solution is used in care homes to look after the wellbeing of patients at night. While motion sensors are common in care homes, they are unsuitable for anything but the most rudimentary of patient monitoring. Particularly at night, these systems are unable to articulate the difference between a patient in need of urgent help, or a patient just moving in their sleep – leading to false alarms that need to be checked by care staff – wasting their valuable time.
As in security systems, a connected camera system monitored by a human operator is a valid solution, but it is a huge violation of a patient’s privacy to have an “always on” camera in their room. AI powered computer vision offers a tidier solution. By using human activity recognition software, and camera systems that only activate when they detect a true problem, staff are only notified when residents need assistance. This helps residents sleep through the night without interruptions, and reduces time lost to false alarms, all while maintaining patient’s privacy.
Another application of AI in healthcare is in robotics. Even a few years ago this would have sounded like science fiction, advances in social robotics mean that machines can learn people’s interests and hold basic conversations. We’ve already seen international trials prove that patients speaking with AI powered social robots experienced boosted mental health and reduced loneliness. These machines aren’t intended to replace human carers, but they do help to fill lonely periods when staff do not have time to keep residents’ company – a problem that is only becoming more common as care staff are stretched further by aging populations.
The use of AI in healthcare faces the problems as AI in security, with data and privacy being by far the biggest challenge – sharing of live camera feeds, incident reports and any patient data accumulated required by these systems has the potential to put operators in violation of data laws if not carried out properly. Any business with AI solutions operating in this space should be adhering to these rules as a matter of course, but data protection is something to consider for any user to be aware of, no matter what sector or environment they are working in.
AI has proven uniquely suited for dealing with the major problems facing elderly care, but no matter how advanced a system is, the human component is absolutely indispensable. Every conclusion that AI systems come to still needs to be evaluated and responded to by a living breathing person, just as before. But the combination of machine and human expertise has the potential to streamline the workload of the overloaded elderly care sector, removing much of the day-to-day busy work and freeing up staff to fulfil their ultimate goal – providing the best possible care to patients.
This article was published on the website of Security Magazine