A neural-fuzzy model of recall based on neuropathological findings in Alzheimer's patients

The problem chosen for this work is the recognition of faces. The inability to recognize familiar faces is a common symptom in many Alzheimer's patients. The author describes an initial effort to develop a computer-based model of recall using artificial neural nets and fuzzy logic. The computer-based neural net model is then altered in a way suggested by the neuropathological findings in brain autopsies of senile dementia of the Alzheimer type (SDAT) patients. The ability of the computer model to correctly recognize images during a progressive alteration of the neural net is reported. The hope is that this model can serve as a tool to study the effects of SDAT on patients and possibly lend support to the early diagnosis of the disease.<
>