An Analog Approach to Computer Diagnosis

The problem of diagnostic computation has been approached by defining each disease in terms of its symptoms and their relative significance in the characterization of that particular disease. These ``definitions'' were then converted into analogous resistive networks and were used as standards against which a group of symptoms representing an unknown disease could be compared. The specialized computer constructed for this purpose compares the unknown with each of the standards stored in the machine, and if there is a match it indicates with which standard or disease there is a match and to what extent a correlation exists between them. The machine compares the unknown with all of the standards even if a match is found early in the process so that if two or more possibilities exist they will all be indicated.