Development of an AI-based algorithm for contrast agent dose reduction in magnetic resonance imaging

Informacje na temat projektu

DAICAR

Identyfikator umowy o grant: 101114169

DOI

10.3030/101114169

Projekt został zamknięty

Data podpisania przez KE 1 Czerwca 2023

Data rozpoczęcia 1 Lipca 2023

Data zakończenia 30 Czerwca 2024

Finansowanie w ramach

• European innovation ecosystems

Koszt całkowity

Brak danych

Wkład UE

€ 75 000,00

Koordynowany przez

RELIOS.VISION GMBH
Germany

Periodic Reporting for period 1 - DAICAR (Development of an AI-based algorithm for contrast agent dose reduction in magnetic resonance imaging)

Okres sprawozdawczy: 2023-07-01 do 2024-06-30

Podsumowanie kontekstu i ogólnych celów projektu

The SmartContrast project aims to use AI algorithms to enhance the contrast agent signal in brain MRI, thereby enabling contrast agent reduction and/or improved diagnostics.
International medical guidelines (such as the contrast agent guidelines of the European Contrast Media Safety Committee of the European Society for Urogenital Radiology, ESUR) recommend using the minimum effective dose to reduce the amount of contrast agent used. The rationale behind this is that MRI contrast agents can accumulate in the environment and drinking water, as well as in the human body, with unknown consequences. Furthermore, in cases of renal dysfunction, some of these agents can lead to certain diseases of the skin and organs, and, like any other foreign substance injected into the body, they may also cause anaphylactic and allergic reactions.
As the demand for MRI contrast agents continues to rise rather than decrease, it is necessary to find a solution to ensure the safety of both patients and the environment.

Prace wykonane od początku projektu do końca okresu sprawozdawczego oraz najważniejsze dotychczasowe rezultaty

As part of the project, AI algorithms were clinically evaluated to enhance MRI contrast agent (CA) signal on cranial MRI, i.e. reducing the CA dose (SmartContrast) and/or improving diagnostics (GadoBoost). A homepage was created to present the company and upcoming products. Collaborations with industry partners have been initiated for future market entry following MDR approval, and preparation of MDR documentation is underway. Investors have been secured for the company's development, two IP transfer agreements signed to transfer IP from the university to the company, and a company-owned European patent filed. The technology behind SmartContrast was published as a preprint on arXiv and submitted to a scientific journal. Additionally, the company participated in a startup accelerator program, attending intensive courses on commercialization, certification, clinical studies, and copyright.

In a study on SmartContrast for CA reduction, radiologists could not distinguish between real and artificial images. One reader detected the same number of metastases in both images, while another found more in the real images, but only for very small metastases. For metastases 5mm or larger, no difference was observed. In a study on GadoBoost for signal enhancement, one reader found significantly more metastases in AI-enhanced images compared to standard dose images.

Innowacyjność oraz oczekiwany potencjalny wpływ (w tym dotychczasowe znaczenie społeczno-gospodarcze i szersze implikacje społeczne projektu)

Results and Potential Impacts
The project provided significant insights into the potential of AI algorithms to enhance MRI contrast agent signals, leading to reduced contrast agent doses or improved diagnostic outcomes.

Potential Impacts
Clinical Impact: AI-enhanced contrast agent signals could reduce the amount of contrast agent needed, minimizing patient exposure to harmful substances and reducing environmental impact.
Economic Impact: Reduced contrast agent use could save costs for healthcare providers and reduce the need for additional imaging, lowering overall healthcare costs.
Regulatory and Market Impact: Successful AI algorithms could set new standards in imaging protocols, influencing guidelines on contrast agent use and promoting AI-integrated diagnostic procedures.

Key Needs for Further Uptake and Success
Further Research: Additional clinical trials are needed to validate findings across diverse populations, readers, and imaging conditions.
Clinical Integration: Real-world demonstration projects are essential to showcase the practical benefits of AI tools.
Market Access: Strategies to access markets and secure partnerships with healthcare providers and imaging centers are necessary.
Regulatory Support: A regulatory framework that accredits AI-enhanced imaging processes is crucial. Collaboration with regulatory bodies is needed to refine standards for AI in medical imaging.
Intellectual Property: Protecting IP is vital for maintaining a competitive advantage.
Finance and Investment: Funding is needed for further research, product development, and market entry.