AI-Driven Radiology Transformations in Rural Tanzania: A Case Study

In rural Tanzania, where medical resources are limited and diseases like tuberculosis (TB) pose a significant threat, the integration of artificial intelligence (AI) in radiology is garnering attention. A pilot project initiated by Dr. Amir H. Bigdeli, Chief Physician of Radiology at Artemed Clinic in Munich, aims to enhance diagnostic capabilities in the region's healthcare system, particularly at St. Walburg's Hospital in Nyangao.

The pilot project began in September 2019 when Dr. Bigdeli visited the hospital and observed that its analog x-ray machine was over 50 years old, rendering it almost unusable. "Functional deficiencies, such as a defective cassette holder, made basic examinations impossible," he stated in an interview with the German Roentgen Society (DRG) in June 2025. This dire situation highlighted the pressing need for modern medical technology in areas where healthcare specialists are scarce.

Tuberculosis remains one of the deadliest infectious diseases globally, particularly in under-resourced regions. According to the World Health Organization (WHO), TB is responsible for approximately 1.5 million deaths each year, with a large proportion of these occurring in low-income countries where diagnostic facilities are inadequate (WHO, 2023).

Recognizing the limitations of traditional diagnostic methods in rural settings, Dr. Bigdeli and his team turned to AI as a scalable solution. "What is needed in such regions is scalable, robust diagnostic support, and this is precisely where AI comes into play," Dr. Bigdeli explained. The project involved modernizing the hospital's x-ray technology and implementing the AI software, CXR, developed by the Romanian startup Rayscape. This AI tool is CE-certified and can identify 148 thoracic pathologies, including TB.

The project faced several hurdles, particularly regarding data protection and infrastructure. Dr. Bigdeli noted, "Patient rights must be respected in Tanzania, especially when it comes to cloud-based image analysis." To address these concerns, the team collaborated with deepc, a technology partner, to ensure compliance with ISO/IEC 27001:2022 standards. The deepcOS platform pseudonymizes patient data onsite and transmits encrypted x-ray images to the cloud for analysis, ensuring patient confidentiality while leveraging AI capabilities.

The first live training session for local radiographers took place in May 2025, marking a significant milestone for the initiative. During this session, the AI successfully detected a case of TB that had previously been overlooked by non-specialists, illustrating the potential of AI to augment human expertise in medical diagnostics.

Dr. Bigdeli emphasized that AI does not aim to replace radiologists but rather to enhance their capabilities. "AI can help where we can't be physically present. It can save lives—not as competition but as an extension of our expertise," he remarked.

This project exemplifies how technology can bridge the gap in healthcare delivery in underserved areas. According to Dr. Benedikt Lenz from the Artemed Foundation, which supports the initiative, the integration of AI not only improves diagnostic accuracy but also promotes a more ethical and globally connected approach to healthcare. "Our knowledge and technology should reach beyond our clinical boundaries," Lenz stated.

The Nyangao project serves as a model for similar initiatives in other regions facing healthcare disparities. The implications of this approach could extend far beyond Tanzania, potentially influencing healthcare delivery models in various low-resource settings worldwide.

In conclusion, the integration of AI in radiology, as demonstrated in Nyangao, represents a significant advancement in addressing public health challenges in developing countries. As AI continues to evolve, its role in enhancing the capabilities of healthcare professionals around the globe will likely expand, paving the way for more equitable healthcare solutions in the future.