The increase in the amount and availability of health data, especially medical images, combined with the emergence of new technological tools based on Artificial Intelligence (AI) and Machine Learning (ML), offer unprecedented opportunities. The use of AI in the medical field in recent years and, more specifically, in medical imaging, for decision-support in the diagnosis and follow-up of a disease, depend on both image acquisition and the interpretation of its content. The former has improved substantially, with devices acquiring data at faster rates and increased resolution, yet the latter has only recently begun to benefit from computer technology, especially through the rapid evolution of Artificial Intelligence (AI) and Machine Learning (ML). Most interpretations of medical images, like mammographies, CTs, X-rays, etc., are currently performed by clinical experts; however, image interpretation by humans is limited due to its subjectivity, large variations across interpreters, and fatigue or lack of time. Computerized tools, specifically AI-supported image analysis and ML, are key enablers to improve diagnosis, by facilitating the identification of the findings that require treatment, and by supporting the clinical experts’ workflow.
Through its Greek Branch specializing in R&D and in constant collaboration with the Italian team, Maggioli’s role in the INCISIVE project is that of the Coordinator, i.e. ensuring that its objectives, communication processes and quality procedures, are achieved. The company is also contributing to several technical tasks, regarding the design and development of the INCISIVE platform and services, as well as the overall system integration which Maggioli leads. Using Machine Learning, INCISIVE addresses challenges related to the detection of patterns in large volumes of cancer imaging data, striving to increase the interpretability of complex imaging data and support more effective decision-making for Healthcare Providers. It also addresses challenges related to data labelling and annotation, as well as availability and sharing of imaging data, so that it can be used for training and validating AI tools for improved imaging methods. The INCISIVE AI decision-support services will address four types of cancer: lung, colorectal, breast and prostate cancer and the related pilot activities of the developed services will be carried out at 8 clinical sites located in 5 countries: Greece, Italy, Spain, Cyprus, and Serbia.
The project’s two main deliverables are:
- an AI-based toolbox consisting of novel AI models, combined with a set of predictive, descriptive and prescriptive analytics, enabling the multi-modal exploration of the available data sources and integrated into 4 major AI services, one for each type of cancer addressed, as well as including a ML-based automatic annotation system to produce data for the training of algorithms in AI research;
- an interoperable pan-European federated repository of health images, that allows the sharing of data in compliance with legal, ethical, privacy and security requirements, for AI-related training and experimentation. The repository will include a High-Performance Computing as-a-Service, as well as a Models-as-a-Service functionality, thereby allowing for cost-effective performance of computationally intensive processing, without the need for maintaining expensive equipment.
INCISIVE has received funding from the European Union’s Horizon 2020 Research & Innovation Program and the consortium, coordinated by Maggioli SpA, is implemented by 26 organizations from 9 different countries: Italy, Spain, Finland, Greece, Cyprus, Serbia, Belgium, United Kingdom, and Luxembourg.
The project has just completed its first 18month period and delivered its first major milestone: the first integrated prototype of the Pan-European federated repository of cancer images and related clinical data, as well as the first prototype of the AI toolbox. It is worth-noting that the first version of the Pan-European federated repository has already been populated with large amounts of retrospective data, shared by the consortium’s data providers, in a GDPR-compliant way. The preparations for the collection and sharing of additional cancer imaging data that will be collected prospectively will soon start, thus confirming the positive prospects of the project.