May 1959

Interpretation of Symptoms with a Data-Processing Machine

Author Affiliations

New York

From the Department of Medicine, New York Hospital-Cornell Medical Center.

AMA Arch Intern Med. 1959;103(5):776-782. doi:10.1001/archinte.1959.00270050098015

Recognizing that machines in the practice of medicine are here to stay, physicians have the obligation to learn as much of their advantages and limitations as they can comprehend. The machine described here merely correlates symptoms set down by the patient and draws conclusions on the basis of what it has "learned" from physicians. Hence it makes the same errors as the human brain which "taught" it plus others that arc inherent in its inability to initiate the thinking process. One reviewer of the paper presented below asked this important question, "What is the character of the error when a diagnosis is made which is not correct? If a patient with flat feet is simply not so diagnosed, this is one type of error, but if the machine reads, 'respiratory tuberculosis inactive,' it's another." This and many other questions properly may arise.

At the same time, the device is an