AI-industrie.png

Smart Manufacturing & Materials-iStock-1135159658-BEWERKT.jpg

<p><strong>Computers that evaluate X-rays and CT scans, propose treatment plans, or monitor ICU patients at the hospital. Work is underway on this at Isala Hospital in Zwolle. “Artificial intelligence (AI) takes work off the hands of medical professionals,” says Lieke Poot, CMIO and head of clinical physics at Isala. “AI shortens waitlists and reduces healthcare costs.”</strong></p><p><img src="https://cms.aihub-oost.nl/uploads/U77_Dz4fk_XMLQ_9_Rck7_Qwya6cd5685_Tm_Fxlx_Q_Sg_Fq_I_cfd2215957.jpeg" alt="U77Dz4fkXMLQ9Rck7Qwya6cd5685TmFxlxQSgFqI.jpeg" srcset="https://cms.aihub-oost.nl/uploads/thumbnail_U77_Dz4fk_XMLQ_9_Rck7_Qwya6cd5685_Tm_Fxlx_Q_Sg_Fq_I_cfd2215957.jpeg 234w,https://cms.aihub-oost.nl/uploads/small_U77_Dz4fk_XMLQ_9_Rck7_Qwya6cd5685_Tm_Fxlx_Q_Sg_Fq_I_cfd2215957.jpeg 500w," sizes="100vw" width="509" height="339"></p><p>“Artificial intelligence (AI) works like a car with extra features. My car has cameras that monitor whether I’m driving neatly between the white lines and whether I’m keeping a safe distance from the car in front of me. I don’t have to do anything for that. In fact, I don’t even notice that my car is doing this; I don’t even have to press a button for it. But I do arrive at my destination more rested than I would in a ‘regular’ car. That’s exactly what AI does in the hospital as well: it supports people in their work.”</p><p>As a clinical physicist and CMIO, Lieke Poot is constantly on the lookout for technical innovations that can make healthcare more efficient and effective. In 2022, artificial intelligence has become an indispensable part of her field. At Isala, she can really make the most of her passion. Because although AI is offered worldwide by developers—for example, Isala has also purchased the system that helps interpret CT scans—Isala also develops its own AI applications.</p><h2>Algorithms that devise treatment plans</h2><p>“We’re currently working on a system that can suggest treatment plans for people with skin cancer. We can’t and aren’t allowed to use this in clinical practice yet, so we’re currently using it solely for scientific research. The goal of this application is to ensure that a patient always receives the best possible treatment and can more easily choose the option that suits them best.”</p><p>To explain how it works, Lieke Poot outlines the situation of a patient with skin cancer. “There are always different treatment options: various types of chemotherapy, surgery, or radiation therapy. Every person is different—consider gender, genetic characteristics, any existing conditions or diseases, and so on. From the medical literature, doctors know which options are likely to be effective and how treatments may affect other factors. For example, chemotherapy can have a negative effect on the health of someone with a muscle disorder, while localized radiation therapy is safe. New insights are constantly being gained worldwide. For a doctor, this means he or she must continuously read research studies to keep all options at their fingertips.”</p><p>That is precisely what the algorithm Isala is developing will eventually take over from doctors. “We feed the system all the information we have and develop algorithms that combine the patient’s unique data with—proven—effective treatment plans. Sometimes multiple options are possible. In that case, the patient can review the options to see which one suits him or her best.”</p><h2>Treatment on demand?</h2><p>Will the doctor’s advice be replaced by “treatment on demand”? “Not at all. Certainly not for the time being. Many AI applications aren’t foolproof yet. There are countless examples of this worldwide. Take the well-known application of facial recognition, for example. Using algorithms, cameras can determine whether someone is a man or a woman. That seemed to work perfectly—until people of color appeared in front of the camera; the system failed. What turned out to be the case? The developers had only trained the algorithm using features from white faces. That shows that a system must be made very ‘rich’ with an infinite amount of data and variables before it can handle everything.”</p><h2>Set off alarms only when truly necessary</h2><p>The latest project Lieke Poot is working on is a collaboration with Windesheim University of Applied Sciences and HKV. Thanks to years of experience in water management, the company has extensive knowledge of sensors, data, and algorithms. Windesheim has IT students who want to apply their knowledge to social issues, such as ensuring access to healthcare.</p><p>“Together, we’re developing a system to monitor patients in the neonatal unit. Newborn babies are constantly monitored there—for things like heart rate, blood pressure, temperature, and so on. There are dozens of parameters involved. So many, in fact, that alarms go off all the time,” Poot explains. “Nurses then rush to the patient. Often—fortunately—it’s a false alarm. A slight rise in temperature, for example, doesn’t have to be a cause for concern as long as the oxygen level in the blood and the body’s fluid levels are normal,” explains Lieke Poot. “We’re developing algorithms that combine all these factors, so that the alarms only go off when a truly dangerous situation is about to arise.”</p><h2>Keeping Healthcare Accessible</h2><p>Doctors’ reactions to the use of AI vary. “Some are leading the way, and some are still a bit resistant,” says Lieke Poot. “AI is changing their work, and some areas of expertise will be less necessary in the future. I can imagine doctors wondering, ‘What does this mean for my job?’ But there is—and will continue to be—a tremendous amount of work in a hospital. Especially with the aging of our population; there will be more and more patients in the coming years. It can be helpful and convenient if certain routine tasks are handled ‘automatically,’ so that doctors have their hands free for the work that only they can do. ‘Collaboration’ with smart technology keeps healthcare affordable and prevents long waitlists.”</p><h2>Identifying Healthcare Needs at an Early Stage</h2><p>Building on that, Lieke Poot advocates for a prevention system that combines data from all healthcare providers in a region. “Suppose you have data from general practitioners, the Municipal Health Service (GGD), hospitals, municipalities, and other healthcare providers all in one system; then you can identify healthcare needs at an early stage. Take, for example, a vulnerable elderly person who arrives at the hospital’s emergency room with a broken hip after a fall. That person may have already visited their general practitioner four times because they sprained their ankle in a minor accident at home. These are signs that help is needed. If we all identify this sooner, we can prevent the situation from getting worse.”</p>

Lieke Poot, Isala: "Artificial intelligence keeps healthcare accessible"

These entrepreneurs have paved the way for you

Meet our frontrunners

U77Dz4fkXMLQ9Rck7Qwya6cd5685TmFxlxQSgFqI.jpeg

<p>Computers that evaluate X-rays and CT scans, propose treatment plans, or monitor ICU patients at the hospital. Work is underway on this at Isala Hospital in Zwolle. “Artificial intelligence (AI) takes work off the hands of medical professionals,” says Lieke Poot, CMIO and head of clinical physics at Isala. “AI shortens waitlists and reduces healthcare costs.”</p>