Towards applying content-based image retrieval in the clinical routine

Content-based image retrieval (CBIR) has been one the most vivid research areas in the field of computer vision, and substantial progress has been made over the last years. As such, many have argued for the use of CBIR to support medical imaging diagnosis. However, the sheer volume of data produced in radiology centers has precluded the use of CBIR in the daily routine of hospitals and clinics. This paper aims to change this status quo. We here present a solution that applies Computational Grids to significantly speed up the CBIR procedure, while preserving the security of data in the clinical routine. This solution combines texture attributes and registration algorithms that together are capable of retrieving images with greater-than-90% precision, yet running in a few minutes over the Grid, making it usable in the clinical routine.