Microb-AI-ome structure

WP1 - Creation of a federated database network for international colorectal cancer data

Key to the success of Microb-AI-ome is a secured design of a federated microbiome database network (Microb-AI-net)
that will enable structuring locally clinical and microbial data while respecting international highest standards for data
privacy, clinical and microbial data treatment.  more ⇒

WP lead:

WP2 - Data collection and management

The goal is to collect and manage the clinical and gut microbial data coming from healthy participants or individuals with CRC, or at risk of developing CRC. The first step will be to aggregate data already available to the consortium partners coming from various studies for over >1,000 individuals. The second step is to extend these data by recruiting >4,000 individuals from ongoing CRC screening initiatives from Ireland, France, and to integrate data from other clinical partners of other countries.   more ⇒

WP lead:

WP3 - Federated artificial intelligence models and patient stratification apps

We will develop and implement a collection of federated CRC Profiler apps covering all steps of a patient stratification workflow, from normalisation and batch-effect correction to AI model learning and application. Further, the individual CRC Profiler apps will be chained into pipelines connected to the federated database network Microb-AI-Net and become available for clinicians as CRC Stratifier software.  more ⇒

WP lead:

WP4 - Application and validation of federated AI models

We will deploy the resulting CRC Profiler apps and CRC Stratifier software within the context of the bowel screening programmes in France and Ireland. Of the 4,000 total subjects recruited over the course of the project in WP2, the first 1,000 available data sets plus the first 2,000 sampled in WP2 (i.e. 3,000 total) will have been used for constructing the Microb-AIs in WP3 and the latter 2,000 subjects will be used for validating the patient stratification performance here. The CRC Profiler based predictions of CRC status will be then compared to the clinical findings of the screening service colonoscopy to determine the sensitivity, specificity and clinical utility of the CRC Stratifier microbiome screening tool as an adjunct or replacement for colonoscopy  more ⇒

WP lead:

WP5 - Evidence-based guidelines

In order to be translated into clinical practice, the developed AI/ML-based CRC patient stratification tools (Microb-AIs, the CRC Profiler apps and the CRC Stratifier software) need to be approved for clinical use by regulatory bodies in the long run. To facilitate and speed-up this process, we will begin with registering Microb-AI-ome technology in the recently established registry for AIs in Medicine (doi: 10.1038/s41592) followed by systematically approaching regulatory bodies with corresponding draft guideline updates.  more ⇒

WP lead:

WP6 - Legal and ethical assessment

Ethics and data protection are at the centre of Microb-AI-ome and the prime rationale behind its federated architecture. Closely linked to WPs 1 and 3, a process of ethical and data protection impact assessment and a process to ensure their implementation into the project, from Microb-AI-Net all the way to the CRC Stratifier, will be established in the project as a whole, including expertise from ethics boards and the clinical partners’ legal departments. more ⇒

WP lead:

WP7 - Dissemination & Communication

Dissemination and communication activities will address different target groups via a multitude of measures. We will inform the public, clinics, academia and stakeholders on the vision of creating the first privacy-preserving federated big data network in CRC research. Interdisciplinary training will attract young scientists and students. Patients, the medical community and possible business partners will be informed about how the novel Microb-AI-ome strategy will streamline, internationalise and optimise CRC patient stratification as well as earlier and cheaper detection of individuals at increased CRC risk. more ⇒

WP lead:

WP8 - Exploitation

We aim to maximise the acceptability and visibility of Microb-AI-ome by reaching out to potentially interested stakeholders at all levels. Therefore we will develop strategies and implement measures to achieve optimal exploitation and the best possible business potential of Microb-AI-ome’s key outputs, namely the CRC Profiler apps and the CRC Stratifier software and its utilisation as medical diagnostic devices. To this end, an exploitation strategy will be developed and implemented.  more ⇒

WP lead:

WP9 - Management

Microb-AI-ome is an ambitious interdisciplinary project and relies on an efficient project management structure. An experienced management team, responsibility split and a smooth decision making structure will enable a successful project navigation. more ⇒

WP lead: