Dynamic Adaptation of Cooperative Agents for MRI Brain Scans Segmentation

Abstract

Abstract.To cope with the difficulty of MRI brain scans automatic segmentation, we need to constrain and control the selection and the adjustment of processing tools depending on the local image characteristics. To extract domain and control knowledge from the image, we propose to use situated cooperative agents whose dedicated behavior, i.e. segmentation of one type of tissue, is dynamically adapted with respect to their position in the image. Qualitative maps are used as a common framework to represent knowledge. Constraints that drive the agents behavior, based on topographic relationships and radiometric information, are gradually gained and refined during the segmentation progress. Incremental refinement of the segmentation is obtained through the combination, distribution and opposition of solutions concurrently proposed by the agents, via respectively three types of cooperation: integrative, augmentative and confrontational. We report in detail our multi-agent approach and results obtained on MRI brain scans.

Multi-agent Approach for Image Processing: A Case Study for MRI Human Brain Scans Interpretation

“…It works locally, diffuses its partial results to its acquaintances (for instance agents dedicated to the same tissue in neighbouring regions), shares results via specific maps and coordinates its actions with other agents to reach a global common goal. On various realistic brain phantoms, we obtained results (about 84% of truth positive) comparable to other optimal methods, which rely on MRF models and include a bias field correction map, with a lower computational burden (less than 5 min to segment a complete volume) (see [4]). The present evaluation was performed on real MRI scans at 1.5 T.…”

“…To conclude, based on our experimentations with phantoms [4] and realistic MRI brain scans, situated and cooperative agents appear as an interesting framework to combine several information processing steps that are required for image interpretation.…”

“…The agents are organized into groups running under control of a group manager which ensures their proper coordination. Briefly, (details about the implementation can be found in [4]), three types of agents coexist in the system : global and local control agents and tissue dedicated agents. The role of the global control agent is to partition the data volume into adjacent territories, and then assign to each territory one local control agent.…”

fMRI Retinotopic Mapping—Step by Step

“…However, given the good agreement between the models obtained independently by different operators, the influence on the results should be small. We are currently working to improve our segmentation procedure (Richard et al, 2002), and hope to further increase the reproducibility of the obtained surface models in the near future.…”

Thirty years of artificial intelligence in medicine (AIME) conferences: A review of research themes

“…An interesting line of work involving distributed and cooperative approaches to image segmentation and interpretation was presented over the years by the research group of C. Garbay [99,[137][138][139][140][141]. Their multi-agent approach was later also integrated with classical methods for image segmentation which are based on Markov random fields [142].…”