Artificial Intelligence in Healthcare – Driver of Digitalization?

In the fight against the virus, Germany has famously relied on lockdown – other countries such as Taiwan, Singapore or South Korea have used modern technologies for tracking contacts, movement patterns as well as warning apps to track down the COVID-19 pathogens with considerable success. This is despite the fact that in the densely populated regions of Taiwan or South Korea, many more people congregate in a small area than in this country.

One important finding for future pandemics is that we still know too little about how people move around in their everyday lives. Artificial intelligence (AI) is essential to change this. After all, movement flows can in principle be controlled with the help of AI. There are already start-ups that perfectly analyze movement flows – and help cities, companies or event organizers to direct their visitors, customers or passers-by to where they should go, i.e. to the sales booth, for example, but not to the exit.

AI makes it possible to analyze camera data and see in real time how many cyclists and pedestrians are waiting at a traffic light or passing through an intersection, how many people are standing at bus stops, boarding buses and trains, and when they change trains. Photos are not necessarily stored; all data can also be processed in cameras in compliance with data protection regulations.

AI supports politicians in the decision-making process

This data helps companies and municipalities to improve their service – in this way, cities can become digitally controlled smart cities in the future. AI makes it possible to manage a company’s parking space more intelligently. Instead of sending out police officers or the public order office because outraged residents post photos of crowds in parks or public spaces, an AI-based warning system can respond in real time to crowds or insufficient spacing.

With such a database, policy makers in the Corona pandemic could have given businesses and people much more freedom, for example, to implement spacing and sanitation rules. Artificial intelligence can also be used in hospitals in particular to recognize how streams of visitors move across the hospital grounds or how long people spend in waiting rooms. This information can then be used to calculate the likelihood of infection.

Whether smart city or smart healthcare, the principle is the same. Data-driven analysis is always better than the trial-and-error approach that our federal government has made the standard for pandemic control.

Social distancing and hygiene compliance may already be included as design principles in the years ahead. Designs and processes that are already standard in healthcare can become models for public spaces in this regard: These include the installation of suitable ventilation systems and the reduction of surfaces on which germs accumulate.

Examples of targeted use of AI and Big Data to tackle pandemics:

• AI can be used to predict virus spread and develop early warning systems by extracting information from social media platforms and news sites and providing useful information about future outbreaks in at-risk regions or new potential clusters as epidemiological forecasts for professionals and policy makers.
• Smart watches, cell phones, or wearables can be used for diagnosis, contact tracking, and efficient AI-assisted quarantine surveillance.
• AI can support the development of vaccines, therapies, and medicines and inspire physicians, pharmaceutical researchers, and scientists to achieve results even faster. For example, AI is capable of permanently searching for publications on important scientific topics. Consequently, it is feasible to evaluate a situation faster and more soundly, so that, for example, a lot of money could be saved by a quick program stop and new findings can be accessed more quickly.
• In the pre-selection of theoretical molecules, AI is an effective helper: researchers may save themselves many experiments and can start right away at a higher level.

Pharmaceutical research becomes faster and saves costs through AI

In pharmaceutical research in particular, great hopes rest on the use of artificial intelligence for the future For example, two research-based companies – Sumitomo Dainippon Pharma of Osaka and Exscientia, based in Oxford – have produced a drug candidate called DSP-1181 through AI, which is intended to be effective against obsessive-compulsive disorders. The targeted indication, which is being tested in the first Phase I clinical trial in Japan, is the treatment of obsessive-compulsive disorder. Psychiatry and neurology are among the research priorities of the Japanese partner. Sumitomo Dainippon has provided its experience and knowledge in monoamine GPCR drug discovery for the project.

Exscientia contributed an AI platform for drug discovery to the project. Novel compounds are automatically designed and prioritized for synthesis by AI systems using algorithms. This allows compounds with the desired criteria to be quickly identified. The exploration research phase on DSP-1181 has taken less than a year – less than a quarter of the average 4.5 years achieved using conventional research techniques.

Another use case by research-based pharmaceutical companies is for clinical trials, for example. Health data could be collected using blockchain techniques and shared only where desired and necessary. At the same time, patients’ privacy and particularly sensitive information can be protected even better. This further strengthens patients’ trust in research. Clinical trials for rare diseases in particular, for which only very few participants are eligible, could benefit from this. The time required for drug development could be reduced, and patients could benefit more quickly from new treatment options.

AI in healthcare in Germany still in its infancy

Many things are therefore possible when it comes to digitization in healthcare, and Germany would do well to learn from successful countries with an open mind. Estonia, for example, is considered a pioneer in digitization in healthcare. Estonian citizens have electronic patient records that are secured with blockchain technology (Blockchain in Medicine & Healthcare). Other areas identified there as areas of focus for possible further use include: Telemedicine, personalized medicine or the archiving of medical records – always also involving data security and data integrity.

Given the dimension of the challenge in the digitization of healthcare, all existing players should work together in the best possible way. The use of professional interim managers, which is already standard in industry, is probably one of them and is also likely to increase significantly in the pharmaceutical industry, medical technology and the hospital sector in the coming years – otherwise the problems will not be solved at all or much too slowly. There is an objective shortage of experts who are able to manage the complexity of new AI processes due to their diverse skills and wealth of experience.

Authors:

• Ralf  H. KOMOR
  
• Tasso A. ENZWEILER
  Enzweiler, who holds a doctorate in economics and an international MBA (Kellogg/WHU), is an interim manager, long-standing managing director and author of a textbook (Springer Gabler). Enzweiler has accompanied numerous projects in the healthcare industry. His focus is on close cooperation with the C-level, based on professional competencies such as Marketing, Sales & Communication. (Website)

Ralf H. KOMOR

Diplom-Wirtschaftsingenieur Ralf H. KOMOR, ist Executive Interim Manager auf C-Level, zertifizierter Beirat, Most Trusted Adviser für die Beratung von Familienunternehmen und Springer Autor. Er entwickelt neue Geschäftsmodelle und Vertriebsstrategien für Startups, Scaleups und den Mittelstand und begleitet aktiv Veränderungsprozesse in Menschen und Organisationen.