CE45 - Mathématique, informatique, automatique, traitement du signal pour répondre aux défis de la biologie et de la santé

Radiation dose optimization for X-ray guided procedures – OptimiX

OptimiX - Radiation dose optimization for X-ray guided procedures

The overarching goal of OptimiX project is to improve the overall radiation safety of patient and clinical staff by (i). developing novel approaches for fast and accurate radiation simulation, (ii). propose methods for optimizing an X-ray imaging device configuration to minimize the delivered dose (iii). developing radiation awareness systems using Augmented/Virtual Reality visualization to facilitate teaching, in an engaging and intuitive manner, on the behavior of ionizing radiation.

General objective and main issues

X-ray based medical imaging plays a fundamental role in several fields of medicine. However, the use of X-rays for medical purposes is associated with an inherent risk of exposing patient, interventional radiologists/surgeons and supporting medical staff members to harmful ionizing radiation. Countless studies have shown that any amount of exposure increases the risk of (i). radiation-induced tissue reactions (epilation, skin necrosis, cataracts…) and (ii). stochastic effects such as malignancies. While a patient’s exposure can be justified by the medical indication and usually occurs in a single episode, medical staff providing patient care can be exposed daily. The repetitive nature of this exposure, even to low doses, increases the risk of developing negative biological effects, and this risk increases with dose accumulated over time. Studies have documented the dosage of ionizing radiation among interventional practitioners as the highest registered for any medical staff working with X-rays. Indeed, most interventions are performed under fluoroscopy guidance (continuous X-ray imaging) with staff most of the time obliged to remain close to the patient during the procedure. Even if most of their body is shielded with lead protective clothing, the dose delivered to unprotected body parts such as hands, eyes and legs, can approach the maximum established limits. Patient dose can get alarmingly high during complex procedures involving extended fluoroscopy times and/or the acquisition of a large number of images in a given projection angle. On the other hand, studies have reported a considerable amount of unnecessary exposures resulting from a lack of awareness, a reduced concern of long-term exposure risks and a poor knowledge of the behavior of ionizing radiation.

Recent works have proposed approaches for estimating patient/staff radiation exposure in quasi real time. However, such approaches are not practical for clinical application and training sessions as they are associated with substantial approximations. Due to the high computational cost required, they rely on a large database of simulations which are pre-computed for a set of X-ray device configurations using a generic patient model. The ambition of the OptimiX project is to eliminate the inherent approximations in such approaches by developing methodologies that will enable real-time dose computation specific to a given imaging device and room layout, as well as a given patient and staff configuration.

Therefore, the overarching goal of OptimiX project is to improve the overall radiation safety of patient and clinical staff by (i). developing novel approaches for fast and accurate radiation simulation considering realistic and personalized models of patient and staff, (ii). propose methods for optimizing an X-ray imaging device configuration to minimize the delivered dose without compromising the overall clinical image quality and (iii). developing radiation awareness systems using Augmented/Virtual Reality visualization to facilitate teaching, in an engaging and intuitive manner, on the behavior of ionizing radiation, the optimal positioning of imaging device to minimize patient/staff dose and the best use of protective measures.

a. First ultra fast prediction model of the personalized dose within the patient using artificial intelligence.
b. First ultra fast prediction model of the X-ray scattering with operating room using artificial intelligence.
c. First optimization model that include the presence of protection shield

The main impact of the OptimiX project will be the improved health of surgeons, interventional radiologists, operating room staff and patients by reducing the absorption of X-ray doses to this population. Methods that will be developed in this project can be easily transferred to healthcare industry and benefit to the scientific community. Finally, OptimiX may also become an international standard to train and teach radiation protection.

In progress

X-ray based medical imaging plays a fundamental role in several fields of medicine. However, the use of X-rays for medical purposes is associated with an inherent risk of exposing patient, interventional radiologists/surgeons and supporting medical staff members to harmful ionizing radiation. Countless studies have shown that any amount of exposure increases the risk of (i). radiation-induced tissue reactions (epilation, skin necrosis, cataracts…) and (ii). stochastic effects such as malignancies. While a patient's exposure can be justified by the medical indication and usually occurs in a single episode, medical staff providing patient care can be exposed daily. The repetitive nature of this exposure, even to low doses, increases the risk of developing negative biological effects, and this risk increases with dose accumulated over time. Studies have documented the dosage of ionizing radiation among interventional practitioners as the highest registered for any medical staff working with X-rays. Indeed, most interventions are performed under fluoroscopy guidance (continuous X-ray imaging) with staff most of the time obliged to remain close to the patient during the procedure. Even if most of their body is shielded with lead protective clothing, the dose delivered to unprotected body parts such as hands, eyes and legs, can approach the maximum established limits. Patient dose can get alarmingly high during complex procedures involving extended fluoroscopy times and/or the acquisition of a large number of images in a given projection angle. On the other hand, studies have reported a considerable amount of unnecessary exposures resulting from a lack of awareness, a reduced concern of long-term exposure risks and a poor knowledge of the behavior of ionizing radiation.

Recent works have proposed approaches for estimating patient/staff radiation exposure in quasi real time. However, such approaches are not practical for clinical application and training sessions as they are associated with substantial approximations. Due to the high computational cost required, they rely on a large database of simulations which are pre-computed for a set of X-ray device configurations using a generic patient model. The ambition of the OptimiX project is to eliminate the inherent approximations in such approaches by developing methodologies that will enable real-time dose computation specific to a given imaging device and room layout, as well as a given patient and staff configuration.

Therefore, the overarching goal of OptimiX project is to improve the overall radiation safety of patient and clinical staff by (i). developing novel approaches for fast and accurate radiation simulation considering realistic and personalized models of patient and staff, (ii). propose methods for optimizing an X-ray imaging device configuration to minimize the delivered dose without compromising the overall clinical image quality and (iii). developing radiation awareness systems using Augmented/Virtual Reality visualization to facilitate teaching, in an engaging and intuitive manner, on the behavior of ionizing radiation, the optimal positioning of imaging device to minimize patient/staff dose and the best use of protective measures.

The main impact of the OptimiX project will be the improved health of surgeons, interventional radiologists, operating room staff and patients by reducing the absorption of X-ray doses to this population. Methods that will be developed in this project can be easily transferred to healthcare industry and benefit to the scientific community. Finally, OptimiX may also become an international standard to train and teach radiation protection.

Project coordinator

Monsieur Julien Bert (LABORATOIRE DE TRAITEMENT DE L'INFORMATION MÉDICALE)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

IHUS FCS IHU MIXSURG
ICube - UNISTRA Laboratoire des sciences de l'Ingénieur, de l'Informatique et de l'Imagerie (UMR 7357)
CHUB Délégation à la Recherche Clinique et à I'innovation - CHRU de Brest
UMR_S1101 LABORATOIRE DE TRAITEMENT DE L'INFORMATION MÉDICALE

Help of the ANR 497,448 euros
Beginning and duration of the scientific project: January 2019 - 36 Months

Useful links

Sign up for the latest news:
Subscribe to our newsletter