Develop a workflow to build the bridge between AI development and clinical deployment.

---

Objective

The aim of this project is to develop a practical AI system which is technically advanced and clinically relevant. We ensure that AI solutions are validated, integrated into workflows and meet high standards of patient care. Within this project, we will develop a full-automated system which integrated the two in-house developed AI algorithms into the PACS system.

Methodology

Following steps have been successfully applied on two clinical cases:

1. Work-out a clinical challenge that can be supported by AI algorithms. This is a collaboration between the physicians and the engineer.
2. Design a scientific solution and develop an AI model which can tackle these challenges.
3. Implement these AI algorithms in a usable package such that it can be integrated in the PACS system.

Impact and future directions

Based on the results of this project, we have implemented a full automatic workflow to bring AI results integrated in the PACS system of our hospital. This assures that AI results can support physicians within their current clinical workflow without any additional burden.

General info & contact

Keywords (#): Medical imaging, AI-driven support, PACS, software tool

RADar project research lead: Dr. Nathalie Mertens

RADar project researchers: Prof. Dr. Peter De Jaeger

Principal investigator: MD Kristof De Smet

Timeline: 2024-2025

Status: Feasibility study has been finished. Clinical research will be started.

Publications / presentations: Demo at RSNA2023, Blog on NVIDIA

Partners: Sectra, NVIDIA