CadFlow efficiency and market impact improvements
Finished
Project duration: 1.12.2023 – 19.04.2024
Total grant approved: EUR 122.800,00
Total eligible expenses: EUR 145.000,00
Support upon a restricted call to enhance system robustness, IP and patent application placement analysis, insights into marketing strategy, and legal advice for product localization in different markets.
CadFlow - A novel AI system for computer computer-aided detection and identification of cardiovascular stenosis
Finished
Project duration: 1.06.2022 – 31.05.2023
Total grant approved: EUR 198.305,00
Total eligible expenses: EUR 233.300,00
The project “A novel AI system for computer-aided detection and identification of cardiovascular stenosis” fuses advanced AI algorithms with medical expertise to leverage cardiovascular healthcare. Developed through a collaboration between a multidisciplinary team of cardiothoracic surgeons, cardiologists, scientists, software engineers, AI/ML, and data experts, CadFlow aims to support the diagnosis of coronary artery disease and personalized optimal treatment indication.
Project outcomes include:
Algorithm Design and Development:
Custom-developed algorithms for assessment automatization of coronarography image data and SyntaxScore calculation, involving coronary artery segmentation, branch identification, stenosis detection, characterization, and scoring their contribution to SyntaxScore.
RichUI Development: The RichUI application, securely deployed in the Azure cloud, allows medical experts to verify, modify, and enhance each AI-generated result to provide transparency, trust, and clinical relevance.
Cloud Deployment:
The application was designed with modern microservice architecture to facilitate efficient AI processing and DICOM data management, as well as to provide scalability and flexibility for further development. It also utilizes external services for base functionalities like user management, data storage, backups, etc.
DICOM Dataset Acquisition and Curation:
A comprehensive DICOM dataset was acquired, analyzed, curated, and cleaned for algorithm development and evaluation.
Feedback from Medical Experts:
Through RichUI real-time feedback on algorithms results has been collected from medical experts and utilized to refine the AI model to ensure clinical compliance and viability.
Throughout the project, milestones have been accomplished, the project results have been disseminated through international conferences and peer-reviewed publications, and collaboration with the medical community has been strengthened. CadFlow paves the way for enhanced cardiac care, combining AI capabilities with clinical expertise for improved patient outcomes and informed treatment strategies.
Cadflow API Exposure – AI based Computer Aided Medical System Capabilities Exposure with API Service
Finished
Project no.: 2022/347151
Total grant approved: EUR 234.700,00
Total eligible expenses: EUR 199.495,00
Project duration: 1.04.2023 – 31.03.2024
The project aims to extend CadFlow by developing capabilities related to computer-aided personalized treatments, integration with external databases, services for ground truth data acquisition, subscription management, and extending the user interface to support a variety of customer types. Another aim of this project is to further develop and refine AI algorithms for SynstaxScore calculation as well as perform a beta test, final validation, and release.
This innovation will empower medical professionals, researchers, AI engineers, and hospitals to collaborate effectively, ultimately improving the accuracy and quality of CVD diagnosis and treatment.
Conferences
December 2023
Presentation at IEEE EMBS Data Science in Malta
7-9 December 2023 Katarzyna Heryan and Witold Serwatka presented our poster entitled Automated Stenosis Detection in Coronarography Image Data at IEEE EMBS Data Science and Engineering in Healthcare, Medicine and Biology in Malta.
September 2023
Presentation at IEEE EMBC Conference in Sydney, Australia
25 July 2023 Witold Serwatka presented our paper entitled A novel framework for differentiating vessel-like objects in coronarography images: IEEE EMBC Conference: 45th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Sydney, Australia panel: Biomedical Imaging and Image Processing
November 2022
Participation in Hungary Cardiological Event 2022, CadFlow presentation
Our presence at Hungarian Cardiological Convent 17-19 November 2022 was a meaningful experience. We aimed to foster relationships and broaden our network, particularly through engaging with the Polish Investment and Trade Agency (PAIH).
During the event, we had the opportunity to present our CadFlow ideas to medical experts and showcase our TeleDICOM solution through practical use-cases. Our engagement with PAIH aimed to establish a supportive network for promoting our solutions globally,
benefiting from their focus on Polish-origin businesses.
Additionally, we forged connections with sales representatives specialized in medical product distribution. Our participation in Hungary Cardiological 2022 facilitated valuable interactions, expanded our network, and contributed to the promotion and potential growth of our offerings within the cardiology field.
