AI in medicine: A win-win for doctors and patients

While the corona pandemic sped up digitization in the healthcare sector, Central Hessen had already set an early course to create the “digital medicine of the future.” Artificial intelligence (AI) is not only making medicine both more efficient and better – in fact, patients are moving to the center of healthcare offerings such as AI-supported diagnostic tools and monitoring apps. Meanwhile, doctors, nurses and therapists will have more time for patients thanks to video consultation hours and digital practice management.

Credit: Pixabay

This ‘medicine of the future’ is a collaboration: real and virtual medicine, doctors and IT experts, patients and digital apps. The result: everyone involved in healthcare will benefit. “Ultimately, the relationship between patient and doctor will improve,” says Prof. Dr. Martin Hirsch, Director of the Institute for Artificial Intelligence in Medicine at Philipps University in Marburg. This is because the new digital tools will enable doctors to identify rare diseases or atypical clinical pictures more quickly. “Doctors can take more seriously patient-measured health data, for example sending it via medical apps over their smartphones, since these physicians have learned about an AI,” Hirsch explains. As a biologist with a doctorate in neuroscience, Hirsch is building the “Center for Digital Medicine” in Marburg. Under his leadership, the web app Covid-online was developed, which serves as a digital guidance system for citizens in the region if they experience symptoms of COVID-19. Hirsch also has extensive experience with digital tools that address the need for personalized information. Among others, he founded Ada Health in Berlin, an end-user self-assessment app developer. Hirsch says, “Personalization can be done better via devices like the smartphone and with the support of AI than via static media.”

0
billion euros is the forecast market volume in the EU for digital products and services by 2025.

The forecast: enormous potential for AI in healthcare

The potential of AI in healthcare is enormous: For example, the 2020 Roland Berger study “Future of Health 2 – The Rise of Healthcare Platforms” forecasts an EU-wide market volume for digital products and services of around 232 billion euros by 2025 – an increase of almost 50 percent compared to last year’s forecast. The COVID pandemic has accelerated the digitization of the industry by at least two years, they add. The analysts at Markets-and-Markets also expect growth of more than 50 percent annually through 2025. According to their calculations, AI sales worldwide were already at two billion US dollars by 2019. 

The application of AI to improve healthcare outcomes goes well beyond clinical usage: Authors of an article published in the Harvard Business Review even argue that the greatest value of AI lies not in supporting clinical decision-making, but in cognitive support for routine interactions. Cognitive virtual assistants that are conversational in nature – with the number of functions they can perform increasing with each interaction – can relieve physicians and nurses of routine tasks and administrative processes, and assist them in the planning and execution of medical and nursing interventions. Patients, in turn, would have round-the-clock access to routine information and support through their virtual assistants.

"Artificial intelligence will fundamentally change healthcare by providing personalized advice to people who are already ill at home, without time pressure yet based on a lot of background knowledge. It then actively supports doctors in diagnosis and therapy in practices and clinics."
Prof. Dr. Martin Hirsch
Director of the Institute for Artificial Intelligence in Medicine at Philipps University in Marburg, Germany

Central Hessen as a model region for AI-supported health

With the help of AI, data that is collected by intelligent machines and smart products can be meaningfully linked and analyzed. The knowledge gained is used to create new, more efficient products and services. However, real AI does not only learn from data. Many medical conditions can be so individual that there are not enough data sets to train the AI. “If I take personalized medicine seriously, I have to teach the AI a basic form of understanding and make it capable of reasoning,” explains Hirsch. Only then can it deal appropriately with an individual situation. “Intelligence is, after all, the ability to exist in a situation that I don’t yet know,” he adds. 

While apps and other AI-based systems should benefit patients, many valid questions exist, not only regarding technological progress, but also ethics and trustworthiness, matters of responsibility and licensing, as well as the role of physicians in society and the training of future doctors. “The debate about AI in medicine must not just be a technological one, but must include all facets of medical practice,” says Hirsch. He adds that this can only be done together with experts from different disciplines and in broad discussion.

In Hirsch’s view, Central Hessen is ideally suited for establishing a model region for AI-supported health. Other medical professionals agree. “With Martin Hirsch, we have brought one of the international medical AI pioneers to Marburg,” explains Dr. Helmut Schaefer, Dean of the Department of Medicine at Philipps University Marburg. New courses of study and training formats will thus emerge in the region which are at the boundary between medicine and computer science. “Our professorship for artificial intelligence in medicine is an important building block for Philipps University in the interaction between technical innovation and social responsibility,” says University President Prof. Dr. Katharina Krause. “With the establishment of the Center for Digital Medicine, the University of Marburg is proving that it is also prepared to take on a responsible pioneering role in this area of digitization.” Dean Schaefer is certain the benefits are revolutionary. “The methods of artificial intelligence in medicine, scientifically evaluated and in the hands of trained doctors and informed patients, will lead to even better medicine.” 

Digital transformation at hospitals has only just begun

The spectrum of AI offerings ranges from telemedicine and workflow management tools for hospitals and doctors’ offices to AI software for diagnostics, robotics and AI in the operating room, and the “digitization” of patients using wearables to generate health data, as a few examples. While monitoring apps and video consultations are already part of everyday life for many patients with acute or chronic illnesses, the goal of digital transformation goes much further, explains radiologist and AI expert PD Dr. Felix Nensa of Smart Hospital Essen: “We’re not only making medicine better, but also more human.” In the Smart Hospital, for example, Nessen adds that many routine activities are delegated to various software assistants with compatible interfaces. Meanwhile, medical staff communicate digitally using tablets and messengers.

The Giessen-Marburg University Hospital (UKGM) is also on the way to using digitization to provide up-to-date and optimal care for patients. “Artificial intelligence is already making our everyday medical work easier,” says Prof. Harald Renz, Medical Director of Marburg University Hospital. “I am convinced that AI will take on far greater areas of responsibility in the future.” As one example: In an emergency, software on a tablet transmits all patient data from the ambulance directly to the hospital. This way, doctors are optimally prepared when the patient arrives. Such a smart transfer of critical medical information is just one of the digital projects that UKGM Marburg would like to tackle in the future. 

Promoting e-health and telemedicine for Central Hessen

“For me, digital transformation means progress in patient care and relief for medical staff,” explains Armin Haeuser, Managing Director of the Competence Center for Telemedicine and E-Health Hessen. His work focuses primarily on cross-sector electronic communication, new and needs-based forms of care, and need-based care chains. The goal is better networking for doctors with everyone else in medicine: other doctors, hospitals, health insurance companies, pharmacies and medical supply stores, rehab centers, care facilities and emergency services. “Telemedicine and e-health are not the sole solution for this, but they are a good and important approach,” says Häuser. After all, while the population ages, healthcare systems are suffering from a shortage of specialists, so they must work even more efficiently.

Ensure funding and consulting needs according to company size

“This is precisely why we, as the economic development agency of the state of Hessen, attach great importance to not missing out on the opportunities presented by the digital transformation in healthcare and promote the sustainable further development of Hessian companies,” says Dr. Rainer Waldschmidt, Managing Director of Hessen Trade & Invest GmbH (HTAI), the economic development agency of the state of Hessen. To achieve this, Waldschmidt and his staff keep an eye on the various funding and consulting needs: New and innovative start-ups need capital, not only when they are founded, but especially to grow on the global market. For small and medium-sized enterprises, digital skills and digitally-trained employees are a crucial competitive factor. And in industry, intelligent production processes can achieve higher productivity and efficiency. 

“Things are happening in Central Hessen” is also the perception of Prof. Dr. Martin Przewloka, Professor of Technologies and Modern Business Informatics at the Technical University of Central Hessen (THM). Przewloka, therefore, deliberately founded the start-up 1ACare, based in Giessen, to develop a sales platform that brings digitally brings together medical supply stores and their customers. “Personally, I see the biggest opportunity of digitization in the fact that it helps people cope better with their illness,” says Przewloka. Since the human brain has limited memory capacity, he believes there is nothing to stop a powerful computer from analyzing data patterns faster and better and making more accurate decisions. “We could use the machines to support us and, in turn, gain time for the actual work on the patient.” It will likely be some time before that becomes a reality at all levels in healthcare. However, the path to achieve this digitalized healthcare is mapped out – and many players in Central Hessen are right at the forefront.

 

AI and Big Data: Order in the Data Jungle

Millions of medical data and patient information are stored every day. With the help of powerful computers and intelligent algorithms, these enormous volumes of data – Big Data – can be effectively analyzed and thus provided with an optimal benefit for the patient. “Today, medical informatics is capable of processing huge amounts of data. This opens up completely new perspectives for the future of medicine. For example, we can make decisions for individual therapy from the analysis of anonymized patient databases,” explains Prof. Dr. Keywan Sohrabi from the Department of Health at THM. 

More and more patients are also willing to make their anonymized data available for research into new therapies. However, it now seems impossible for individual doctors to keep track of all the data, let alone be able to evaluate it. AI tools are thus becoming assistants: They provide valuable information in prevention and follow-up, for example, and check that nothing is overlooked. 

Giessen-based contract research organization Alcedis GmbH combines big data with clinical studies. The goal: to bring drugs to approval readiness and make them accessible to doctors and patients. “We offer our customers customized software solutions to provide them with the best possible support in evaluating study data,” says Hanno Haertlein, Managing Director of Alcedis GmbH. “Digitization helps here, for example, to make faster statements about the efficacy and safety of drugs, as certain processes are significantly accelerated.