HealthAgents: distributed multi-agent brain tumor diagnosis and prognosis
Applied Intelligence
| Hi-index | 0.00 |
The authors describe the implementation of SISMAPEO, a database system designed to manage brain electrical activity mapping (BEAM) information which enables practitioners to carry out statistical analysis on groups and individuals based on topography maps. In order to evaluate their system they followed a methodology in which each patient is exposed to eyes-closed (OC) and eyes-open (OA) activation methods. Twenty 4-second epochs with a subset of 18 electrodes selected from the 10-20 International System referential montage are recorded. The register is transformed to the frequency domain via a FFT algorithm and then it is analyzed using statistical parameters for each typical band of the EEG spectrum. They present comparisons based on the t-score test, between two infant groups: a group with learning disabilities and a control. An analysis of the resulting coloured brain-mapping studies is presented. These patients show no evidence of abnormality in any imaging study such as CT or NMR. That is why they are using alternative ways of evaluating disorders such as neurophysiologic ones. At this point, they concentrate only on brain mapping records but the SISMAPEO system is planned to manage other relevant records, e.g., EMG or evoked potentials. They are carrying out clinical validations on infants with language and learning disorders.