Periodic Reporting for period 3 - SINFONIA (Radiation risk appraisal for detrimental effects from medical exposure during management of patients with lymphoma or brain tumour)
Berichtszeitraum: 2023-09-01 bis 2024-08-31
The SINFONIA project addressed critical shortcomings in radiation dosimetry and risk assessment within the medical field. Specifically, it sought to overcome the limitations of current methods for estimating personalized, organ-specific doses in X-ray imaging examinations, which are essential for accurate risk appraisal. The project also developed AI-powered tools, enabling personalized dosimetry in routine clinical practice, and novel methodologies for accurately estimating the dose received by patients from nuclear medicine (NM) procedures and radiation therapy (RT). SINFONIA conducted research on individual sensitivity to radiation and susceptibility to second malignant neoplasms (SMNs) for risk appraisal. Moreover, the project focused on evaluating radiation exposure for medical workers, caregivers, and the broader public and environment. Major results also include a repository for doses, images and non-imaging data; multidisciplinary education and training; and recommendations for diagnostic radiology, NM and RT.
By providing more accurate and personalized radiation dose estimates, SINFONIA contributes to:
• Improved patient safety: Personalized dosimetry can help healthcare professionals tailor imaging and radiation treatments to individual patients, reducing the likelihood of overexposure and minimizing the risk of side effects and complications.
• Enhanced radiation protection: The project's advancements in dose estimation and risk assessment can lead to more effective measures to protect healthcare workers, caregivers, the public and environment from unnecessary exposure. SINFONIA’s research findings can help institutions implement safer working environments, potentially reducing occupational health risks and associated costs.
• Informed decision making: By providing accurate information about radiation doses, SINFONIA can empower individuals and policymakers to make informed decisions regarding the use of ionizing radiation in medical applications.
By providing more accurate and personalized radiation dose estimates, SINFONIA contributes to:
• Improved patient safety: Personalized dosimetry can help healthcare professionals tailor imaging and radiation treatments to individual patients, reducing the likelihood of overexposure and minimizing the risk of side effects and complications.
• Enhanced radiation protection: The project's advancements in dose estimation and risk assessment can lead to more effective measures to protect healthcare workers, caregivers, the public and environment from unnecessary exposure. SINFONIA’s research findings can help institutions implement safer working environments, potentially reducing occupational health risks and associated costs.
• Informed decision making: By providing accurate information about radiation doses, SINFONIA can empower individuals and policymakers to make informed decisions regarding the use of ionizing radiation in medical applications.
SINFONIA made substantial advances toward achieving its objectives. A web-based personalized dosimetry platform (i-Dose, http://idose.med.uoc.gr/(öffnet in neuem Fenster)) for patients with brain tumors and lymphomas enables accurate estimation of organ doses from diagnostic and therapeutic X-ray imaging procedures. Through various tools and AI-driven extensions, healthcare professionals can calculate patient radiation doses and radiogenic risks.
A method was devised to calculate the total radiation dose from external RT, considering the individual contributions from the treatment planning system, out-of-field radiation and imaging doses, to assess associated risks. A set of innovative digital tools was developed to enable real-time extremity dose monitoring and simulations of exposure scenarios, including tools for close-contact dose simulations between patients and caregivers. Additionally, new models were developed to assess the environmental impact of radiopharmaceuticals released into water systems, helping to quantify radiation exposures to humans and wildlife.
In the field of cancer therapy, the project concluded that patients who develop SMNs do not exhibit a higher level of intrinsic radiosensitivity. This finding helps narrow the search for predictive biomarkers. On the technological front, the creation of a secure, cloud-based repository for storing and sharing medical data was a major achievement. It also incorporates AI tools to enable the execution of algorithms for image analysis, dose estimation and other tasks.
A survey was conducted to identify gaps and good practices in education and training in the EU. SINFONIA organized six sustainable, high-level multidisciplinary training courses. Research on novel education tools has led to the development and validation of a context-aware module to allow health professionals to ask questions and receive answers based on proper and reliable educational material. The module is based on a large language model (LLM), which is an advanced technology gaining popularity and trust.
Based on SINFONIA's research, recommendations were developed for diagnostic radiology, NM and RT. These recommendations target professionals working with ionizing radiation or those involved in determining doses in medical use and assessing risks. A comprehensive communication and dissemination plan was developed and implemented. Additionally, informative materials for a non-expert audience were made accessible through different channels.
A method was devised to calculate the total radiation dose from external RT, considering the individual contributions from the treatment planning system, out-of-field radiation and imaging doses, to assess associated risks. A set of innovative digital tools was developed to enable real-time extremity dose monitoring and simulations of exposure scenarios, including tools for close-contact dose simulations between patients and caregivers. Additionally, new models were developed to assess the environmental impact of radiopharmaceuticals released into water systems, helping to quantify radiation exposures to humans and wildlife.
In the field of cancer therapy, the project concluded that patients who develop SMNs do not exhibit a higher level of intrinsic radiosensitivity. This finding helps narrow the search for predictive biomarkers. On the technological front, the creation of a secure, cloud-based repository for storing and sharing medical data was a major achievement. It also incorporates AI tools to enable the execution of algorithms for image analysis, dose estimation and other tasks.
A survey was conducted to identify gaps and good practices in education and training in the EU. SINFONIA organized six sustainable, high-level multidisciplinary training courses. Research on novel education tools has led to the development and validation of a context-aware module to allow health professionals to ask questions and receive answers based on proper and reliable educational material. The module is based on a large language model (LLM), which is an advanced technology gaining popularity and trust.
Based on SINFONIA's research, recommendations were developed for diagnostic radiology, NM and RT. These recommendations target professionals working with ionizing radiation or those involved in determining doses in medical use and assessing risks. A comprehensive communication and dissemination plan was developed and implemented. Additionally, informative materials for a non-expert audience were made accessible through different channels.
SINFONIA achieved significant progress beyond the current state of the art in several areas. The project’s dosimetry platform iDose, equipped with AI-powered tools, represents a major leap forward in personalized medicine. Research on PET–CT led to the development of novel deep-learning tools for patient-specific internal radiation dosimetry calculations. AI-enhanced algorithms were developed for the generation of 3D dose distributions in radionuclide therapy and theranostic applications. Innovative methods employing both traditional and machine-learning techniques were developed to estimate radiation doses from RT imaging. Furthermore, a methodology was developed for summation of the individual contributions of the doses rendered by treatment planning systems for either photon or proton therapy and out-of-field and imaging doses.
The project’s real-time extremity dose calculation tools for NM workers can potentially transform radiation safety practices in hospitals. The project demonstrated that exposure risks from proton therapy, an emerging cancer treatment, are low, alleviating public concerns and promoting wider adoption of the therapy. Additionally, the models developed to assess the environmental impact of radiopharmaceuticals may potentially influence regulatory policies, ensuring better protection of both human and environmental health from radiation exposure.
SINFONIA’s cloud-based repository is a secure platform for sharing and processing medical data with features specifically tailored to the needs of healthcare providers. The ability to host and run AI algorithms within the repository makes it a valuable tool for both researchers and clinicians, enabling them to automate routine tasks such as radiation dose calculations.
An AI-powered context-aware module was developed to assist healthcare professionals in efficiently finding the information they need, promoting continuous education. The tool enables users to query a LLM, whose responses are generated from a controlled and verifiable knowledge base. The methodology employs retrieval-augmented generation, a technique that enhances the LLM's output quality and reliability by leveraging a controlled and authoritative knowledge base.
In summary, SINFONIA contributes to:
• Reduced healthcare costs: More accurate dose calculations can lead to more efficient and effective treatments, potentially reducing the need for additional procedures or interventions.
• Environmental protection: SINFONIA’s models assessing the environmental impact of radiopharmaceuticals can inform policies that safeguard human and environmental health.
• Economic growth: The development and adoption of innovative radiation technology can stimulate economic growth and job creation in the healthcare and technology sectors.
• Increased public trust: By demonstrating a commitment to radiation safety and responsible use of ionizing radiation, SINFONIA can help build public trust in medical technologies.
The project’s real-time extremity dose calculation tools for NM workers can potentially transform radiation safety practices in hospitals. The project demonstrated that exposure risks from proton therapy, an emerging cancer treatment, are low, alleviating public concerns and promoting wider adoption of the therapy. Additionally, the models developed to assess the environmental impact of radiopharmaceuticals may potentially influence regulatory policies, ensuring better protection of both human and environmental health from radiation exposure.
SINFONIA’s cloud-based repository is a secure platform for sharing and processing medical data with features specifically tailored to the needs of healthcare providers. The ability to host and run AI algorithms within the repository makes it a valuable tool for both researchers and clinicians, enabling them to automate routine tasks such as radiation dose calculations.
An AI-powered context-aware module was developed to assist healthcare professionals in efficiently finding the information they need, promoting continuous education. The tool enables users to query a LLM, whose responses are generated from a controlled and verifiable knowledge base. The methodology employs retrieval-augmented generation, a technique that enhances the LLM's output quality and reliability by leveraging a controlled and authoritative knowledge base.
In summary, SINFONIA contributes to:
• Reduced healthcare costs: More accurate dose calculations can lead to more efficient and effective treatments, potentially reducing the need for additional procedures or interventions.
• Environmental protection: SINFONIA’s models assessing the environmental impact of radiopharmaceuticals can inform policies that safeguard human and environmental health.
• Economic growth: The development and adoption of innovative radiation technology can stimulate economic growth and job creation in the healthcare and technology sectors.
• Increased public trust: By demonstrating a commitment to radiation safety and responsible use of ionizing radiation, SINFONIA can help build public trust in medical technologies.