A collaborative study of 3,518 patients in several medical centers was designed to determine the sensitivity and specificity of thermography as compared to mammographic and clinical examinations. For 862 patients who had breast surgery or a biopsy, the sensitivity and specificity was 3 2 3 and .784 respectively for physical examinations, .721 and .798 for thermographic reading with clinical data, and .703 and .SO7 for mammography. Of additional interest was the observation that each of the procedures (physical examination, mammography, and thermography) selects different subgroups of cases. The results indicate that further evaluation of thermography as a screening technique for breast cancer is warranted. Such evaluation should be i n the form of a controlled clinical trial.
OR THE PAST SEVERAL YEARS THERE HAS BEEN
F an increased interest in the use of thermography in the detection of breast cancer.2, This technique is based on the observation first made by Lawson that breast cancer was associated with an elevation of temperature of the skin over the lesion.4 Electronic apparatus, utilizing an infrared camera, became available which permitted the transfer of the skin's infrared heat pattern onto Polaroid film, producing a permanent record of this pattern by scanning the breasts. T h e interpretation of the thermogram depends upon the comparison of various shades ranging from black to white, with the lighter areas indicating the higher temperatures and the darker areas the lower. A calibration scale, imprinted on the thermogram, permits the classification of the dark and light areas into various temperature groups.
This technique has been adopted by several medical centers. By 1965, it became clear that thermography might be a useful tool in the screening of population groups for breast