Exhibit:

Pneumo. AI

Lender: Competence Platform KI.NRW and Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS

AI-based Diagnostic Support for Pneumonia

How can artificial intelligence support medical diagnostics? The AI demonstrator “Pneumo.AI” offers an interactive experience, illustrating how medical professionals can use AI to assist in identifying pneumonia, a type of lung infection. Visitors can try out the demonstrator themselves at pneumoai.ki.nrw and decide based on X-ray images: Is it pneumonia, or is the lung healthy?

Additionally, the demonstrator provides insights into the decision-making process of the AI system. Users can explore how the data is processed and verified.

How does Pneumo.AI work?

The AI system analyzes X-ray images (thoracic scans) of patients using advanced technologies like computer vision and deep learning based on convolutional neural networks (CNNs). These algorithms are specifically designed to efficiently process large image datasets.

Moreover, the system is optimized by integrating prior knowledge and expert insights, making Pneumo.AI particularly data-efficient.

Interactive Experience for Visitors:

Game Mode: Visitors can make their own diagnoses based on X-ray images.

Comparison with Medical Experts: At the end of each game round, participants learn how their results compare to those of radiology professionals supported by AI.

Artificial Intelligence in Medicine

Pneumonia, an acute inflammation of the lower respiratory tract, remains a serious global health issue. In developing countries, it continues to be one of the leading causes of illness and death among children under the age of five.

AI systems like Pneumo.AI can help medical professionals make faster and more precise diagnoses. However, decision-making authority always remains with human experts—AI tools function purely as support, not replacements.

A careful and sensitive approach to medical data is paramount in this context—data protection and security take the highest priority.