Objective
The COVIRA project (COmputer VIsion in RAdiology) aims at a substantial improvement of the quality of computer assistance in the clinical Neurosciences by providing a fundamental image interpretation tool which is a prerequisite for efficient computer assistance in neuroradiological diagnosis, in radiation therapy planning and in stereotactic neurosurgery.
Experiments have been carried out on computer vision in radiology. The prime result of the project is the hardware and software specification of a system for knowledge based interpretation of cranial magnetic resonance (MR) images as a means of computer assistance in diagnostic radiology, radiation therapy planning and stereotactic neurosurgery. This specification is based on demonstrated results from prototypical implementations and an analysis of the state of the art and of the clinical requirements. Image segmentation and interpretation results were obtained for a set of test images of tumour patients selected by medical experts from the above fields. For each patient slice, two MR spin echo images were available. For image segmentation, results were obtained for 3 schemes: one using a Canny edge detector followed by a smooth patch fitting for region finding, another using a region detection according to Nagao/Matsuyama and a Marr-Hildreth edge detector based on one of the echoes to guide region merging, and a third scheme using a multiscale approach to edge detection. The image interpretation results were based on a case model representation of clinical, anatomical, MR-Physics, and tissue parameter knowledge. Results were obtained for 2 schemes: one using a fuzzy clustering approach providing a fuzzy segmentation and a subsequent fuzzy relational matching step, and another using a blackboard approach based on the classical low, middle and high level of processing, allowing for a dynamic control with feedback and backtracking mechanisms. The results were comparatively evaluated by medical experts based on criteria of clinical usefulness.
By involving extensive expert knowledge which is represented and used in the system, a significant amount of consultation time between radiologists, neurologists, oncologist and neurosurgeons can be saved resulting in improved cost-performance in clinical decision making. In the Exploratory Action AIM, the COVIRA project will result in a specification of a system suitable for clinical use. Such a system is to be set up within an AIM main phase activity.
The project proposed here addresses workplan item T550, "Medical Image and Signal Interpretation, Pattern Recognition" and is related to workplan items T510 (Reasoning from Medical Databases), T530 (Advanced Architecture for Diagnostic Knowledge Based Systems) as well as T545 (Prototype Implementation for Medical Consultation Systems).
The low level image processing approach claims generality with respect to most classes of medical images, whereas the symbolic processing at a higher level draws upon four knowledge sources relating to the domain of cranial Magnetic Resonance (MR) images of patients with brain lesions : Anatomical, clinical, statistical (MR tissue parameters) as well as Physics (MR Scanning) knowledge.
The knowledge sources and a case model representation will be used by all partners. Furthermore, the clinical participants, representing the potential application areas (oncology, stereotactic neurosurgery and diagnostic neuroradiology) will select a number of representative clinical cases to be processed by all partners to evaluate the interpretation results against professional "manual" interpretation performance.
In accordance with the objectives of T550, the project will "identify the requirements for low to intermediate level processing of medical images" and extend to high level processing and interpretation. For this purpose low level processing algorithms will be programmed in "C" with a division of tasks, and integrated into different control structures to be pursued in parallel. These will be rated at the end of the project with regard to their performance for the standardized input data in order to arrive at a specification for an optimum system architecture to be realized in the expected AIM main phase.
These control structures include a blackboard approach with heuristic search and backtracking strategy, which makes use of a rule based "integration of primitives", and an approach for information maximization using an iconic fuzzy set representation of organs of the body. The implementation will be object oriented in "Common LISP". For intermediate level processing, use of massive parallelism will be evaluated.
Based on these parallel approaches (as recommended by CEC for the initial work), the project will relate to the further objectives of T550 by "demonstrating medical image interpretation results via a prototype system" and finally by "making recommendations with respect to possible further activities". The potential impact on clinical decision making will also be investigated.
The consortium partners have substantial experience in all above aspects.
In order to maximize probability of success, the project has been restricted to the MR imaging modality and to the evaluation of organ structures, neglecting the fact that the envisaged clinical application may also call for an investigation of the more complex vascular structures in the brain, which are likely to be the subject of other AIM projects. The project therefore has allocated an activity to the exchange of information with such projects.
Main Deliverables :
Specification report on a system for the interpretation of cranial magnetic resonance images for applications in diagnostic radiology, stereotactic neuro-surgery and radiation therapy planning, based on demonstrated results.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques. See: The European Science Vocabulary.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques. See: The European Science Vocabulary.
- natural sciences biological sciences neurobiology
- natural sciences computer and information sciences artificial intelligence computer vision
- medical and health sciences clinical medicine radiology
- natural sciences computer and information sciences artificial intelligence pattern recognition
- natural sciences computer and information sciences artificial intelligence heuristic programming
You need to log in or register to use this function
Programme(s)
Multi-annual funding programmes that define the EU’s priorities for research and innovation.
Multi-annual funding programmes that define the EU’s priorities for research and innovation.
Topic(s)
Calls for proposals are divided into topics. A topic defines a specific subject or area for which applicants can submit proposals. The description of a topic comprises its specific scope and the expected impact of the funded project.
Data not available
Calls for proposals are divided into topics. A topic defines a specific subject or area for which applicants can submit proposals. The description of a topic comprises its specific scope and the expected impact of the funded project.
Call for proposal
Procedure for inviting applicants to submit project proposals, with the aim of receiving EU funding.
Data not available
Procedure for inviting applicants to submit project proposals, with the aim of receiving EU funding.
Funding Scheme
Funding scheme (or “Type of Action”) inside a programme with common features. It specifies: the scope of what is funded; the reimbursement rate; specific evaluation criteria to qualify for funding; and the use of simplified forms of costs like lump sums.
Funding scheme (or “Type of Action”) inside a programme with common features. It specifies: the scope of what is funded; the reimbursement rate; specific evaluation criteria to qualify for funding; and the use of simplified forms of costs like lump sums.
Data not available
Coordinator
22527 Hamburg
Germany
The total costs incurred by this organisation to participate in the project, including direct and indirect costs. This amount is a subset of the overall project budget.