"No patient wants to stay in hospital for an unnecessarily long time."
How long a patient stays in hospital depends on many factors. An app for predicting inpatient length of stay (VWD) aims to optimise patient care and improve discharge management. Daniel Lichterfeld CEO and Founder of AICURA Medical explains how.
What does AICURA do?
AICURA develops a software platform for lifecycle management of medical data and artificial intelligence-based applications - AI apps - for the healthcare industry and its heterogeneous data silos. Our technologies bring the AI apps to the data and not the data to the AI app. This approach makes the data usable but at the same time offers the highest level of data protection. The AI apps optimize economic processes or support diagnosis and therapy decisions. In addition, with our platform, we provide players such as pharmaceutical companies or MedTech companies with a tool for the development of new therapies and for the implementation of clinical trials. Both save time and money.
A forever discussed topic is the length of stay of a patient in the hospital. Together with KMS, you have developed software for predicting inpatient length of stay, the VWD app. What is it about exactly?
It's about analyzing information such as examination or diagnosis data in relation to a patient's length of stay. We have developed an AI app that analyses the information of more than one million patients and makes a prediction about the length of stay. The aim is to be able to predict the date of a patient's discharge more accurately in order to be able to plan the discharge process at an early stage. At the moment, post-discharge treatment or rehab tends to be planned at short notice. The predictive planning made possible by the VWD App leads to considerable advantages for the hospital in organizing and managing patients, such as staff planning.
The average length of stay in Germany has remained almost constant in recent years. According to Eurostat figures, it is nine days, which is twice as long as in the Netherlands. Can the length of stay be reduced with the help of your app?
If the length of stay is half as high in the Netherlands, then there is a need for optimization. But the mere length of stay alone does not help us. With the help of the VWD app, the reasons for the longer length of stay can be analyzed. This means that the responsible staff can see much more quickly how hospital management can be improved. No patient wants to stay unnecessarily long in the hospital. There needs to be an optimal balance of therapy in and recovery from the hospital. We assume that the average length of stay can be reduced with the help of the app. How much differs from hospital to hospital.
What do the medical professions gain from the app?
First and foremost, the VWD app supports hospital staff in discharge management. But the app also supports nursing staff. It becomes comprehensible which factors have an influence on the length of stay. With the VWD App, we support organizational planning in the hospital. We do not replace the employee. We help them to use their limited time more efficiently.
Which are the technologies used?
The AI app and platform require access to the KMS data warehouse database. The AI app is developed with the help of federated learning, which means we can develop the AI app based on information from several hospitals and thus make the knowledge gained available to all hospitals in compliance with data protection laws.
A big challenge for everyone involved is the time before and after the patient is discharged from the hospital. What will soon be possible here?
Systems like Recare benefit from reliable information in order to be able to initiate processes in time. The goal is to organize post-discharge stays in such a way that they run without interruption, without the patient having to wait for days for a place.
Is this already happening or is it still a dream of the future?
The first hospitals are already being equipped with the VWD app.
What happens to the data?
The data stays where it is: The medical data never leave the hospital. Our platform receives the data locally via the KMS Data Warehouse, we process it and then delete it. This ensures 100 percent data protection.
Let's venture a look into the future: what's next?
Artificial intelligence will change healthcare in all areas. As early as 2018, McKinsey estimated the potential savings from AI in the hospital sector alone at EUR 34 billion per year. Since then, the possibilities have continued to develop in leaps and bounds. We are still at the beginning of this development. The IT landscape will also change radically in Germany. AI support in diagnosis and therapy will increase. Pharmaceutical companies will cooperate more closely with hospitals to develop modern therapies and medicines more cost-effectively. We need to network both worlds more strongly. This is a cultural issue, but also a money issue. We want to spread the knowledge in the hospitals and see ourselves as a driver of this development.
Is the cooperation with KMS a win-win for you?
Absolutely. The more than 25 years of experience of KMS with its network of over 500 hospitals, the data warehouse, and the understanding of machine data are a win-win situation for us.