New model developed to estimate radiation skin doses during CT-guided interventional ...( A new model that would allow interve...),Health

A new model that would allow interventional radiologists (radiologists who specialize in fine needle aspiration, fine needle biopsy and radiofrequency ablation) to better estimate patient radiation skin doses during CT guided interventional procedures has been developed according to a study performed at the Agios Savvas and Konstantopoulio Hospitals in Athens, Greece.

"It is clear that skin doses in CT guided interventional procedures can become very high. Even for skin doses around 1 Gy, the prospect of repeating a procedure makes the determination of peak skin dose crucial for avoiding radiation injuries," said Ioannis A. Tsalafoutas, PhD, Virginia Tsapaki, PhD, Charikleia Triantopoulou, MD, John Papailiou MD, Christina Pouli, MD, Virginia Kouridou, MSc, and Ioanna Fagadaki, MD, authors of the study.

The theoretical model that was developed "considers the skin dose resulting from each CT slice, utilizing data that is already stored along with CT images. The skin doses calculated with this model were compared with those measured using films positioned under patients that underwent CT guided interventional procedures. The results indicate that peak skin doses can be estimated accurately using the new theoretical model that provides a base for skin dose estimation in real time," said Dr. Tsalafoutas and colleagues.

"It is important for CT interventional radiologists to be able to monitor the radiation skin dose to their patient and optimize their techniques, so as to avoid skin injuries and minimize the probability of radiation induced carcinogenesis. The first step toward this goal is to understand the risk, to quantify it and to identify factors that affect it in order to be able to reduce it," said Dr. Tsalafoutas and colleagues.

The theoretical method developed by Dr. Tsalafoutas and his colleagues could possibly lead to the development of specialized software for skin estimation in real time which "would be a signific
'/>"/>