“Do you think you can get cancer following an imaging test like a CT scan?” asks Andreas Jahnen, a senior research associate within the "IT for Innovative Services" (ITIS) department at LIST. Jahnen has been part of an EU-wide study which aimed to assess the risk of cancer development resulting from radiation exposure during medical procedures, particularly in paediatric cases. He developed a software which was designed to extract and process data from medical imaging systems across Europe. The results of the study have been published in Nature Medicine.
More than one million European children undergo computed tomography (CT) scans every year. Exposure to moderate to high doses of ionizing radiation is known to increase the risk of blood-related cancers. However, the risks associated with the dose levels used in CT examinations are still not well understood. The prevailing belief has mostly been that of a linear relationship— more dose equating to more risk, and less dose correlating with lower risk. This linear non-threshold model formed the basis of radiation protection strategies. The backdrop to the study in question found its roots in the historical evolution of radiation risk perception.
EPI-CT, short for Epidemiological Study to Quantify Risks for Paediatric Computerized Tomography, ran between 2012 and 2017. Funded by the European Union under the Seventh Framework Programme (FP7), the project narrowed its focus to paediatric patients, who are more radiosensitive due to their developing physiology. The project aimed to understand the risks associated with the radiation exposure in medical procedures, especially CT scans, in this demographic.
The study followed a group of 948,174 individuals from nine European countries (the EPI-CT cohort) who had undergone CT scans before the age of 22 and found a connection between the cumulative radiation dose and the risk of developing blood-related cancers.
The EPI-CT cohort was created by extracting patient information from the Radiology Information System (RIS) of 276 hospitals in nine countries. A total of 1,430,454 examinations were extracted from the RIS, which included administrative details but lacked technical parameters like scanner model and exposure indicators. To address this, data from other sources, such as Picture Archiving and Communication Systems (PACS) and national surveys, were used. PACS, introduced in the mid-1990s, systematically recorded machine settings in the Digital Imaging and Communications in Medicine (DICOM) header of each image.
“The enormity of the task required a substantial dataset. With a goal to study almost one million paediatric patients, the software streamlined the collection and processing of information from hospitals across the nine countries,” Jahnen said.
The software connected with hospital information systems, extracting DICOM metadata from PACS databases. This data, crucial for organ dose reconstruction, went beyond conventional approaches, offering a more holistic view of radiation's impact on specific organs.
The successful culmination of EPI-CT in 2017 marked a significant milestone, in that it contributed to a growing body of evidence which emphasizes the importance of continually justifying the use of paediatric CT examinations and optimizing radiation doses to minimize potential health risks.
“The success of EPI-CT has paved the way for HARMONIC, a follow-up project under the Horizon 2020 initiative,” Jahnen adds. This ongoing project focuses on enhancing our comprehension of the prolonged health consequences resulting from medical exposure to ionizing radiation in children. “As for our software, it continues to play its role as a facilitator in collecting and processing data for assessing the risks.”