Frequent monitoring and tight metabolic control of blood glucose levels can reduce microvascular complications and subsequent co-morbidities in patients with diabetes. Self-monitoring with finger sticks provides intermittent data at best, and results in poor compliance. We report on a minimally invasive system that continually measures glucose flux through ultrasonically permeated skin. Ten patients with diabetes were enrolled in a clinical study to determine correlation between data collected by glucose biosensors placed over ultrasonically treated skin sites (two per patient), and blood glucose readings were taken every 20 min over an 8-h period. Glucose flux biosensors measured amperometric current proportional to hydrogen peroxide level, generated from catalytic conversion of glucose by glucose oxidase; the sensor was coupled to the skin by a thin hydrogel containing an osmotic extraction buffer, creating a gradient for glucose transport through the skin. The biosensors were attached to small portable meters that recorded time, current, and temperature readings every 5 s. At the conclusion of the study period, meter recordings were downloaded for data processing. Skin sites were examined for irritation due to biosensor contact. Data from glucose biosensors with completed data sets had a correlation coefficient of 0.84, and 95% of the data pairs (n = 241) were in the A + B region of a Clarke error grid. Ultrasonic pretreatment lasting about 10 s resulted in improved conductance in all patients. No patients complained of pain or irritation at any time during the study. Continuous monitoring of glucose flux through ultrasonically permeable skin is safe and feasible.

Liquid crystal (L.C.) thermographic study of 105 women with abnormal breast characteristics was undertaken to evaluate the efficiency of L.C. thermography in breast cancer detection. Six general thermographic signs were noted to occur in breasts affected by breast cancer. These signs were derived from L.C. thermograms of 17 patients with histologically proven carcinoma. Detection of palpable malignancies with L.C. thermography had a true‐positive rate of 82.3%. The false‐positive rate was 13.6%, and only one of 17 histologically proven malignancies gave no thermographic signs of malignancy. Liquid crystal thermograms of 197 apparently healthy women with no breast abnormalities were classified according to pattern type. Six distinct thermal pattern types were characterized with 3 subgroups distinguishable in each of 3 vascular pattern groups. The L.C. pattern type was studied as a function of age, past pregnancies, previous lactation, use or non‐use of oral contraceptives, and breast size. In general, women under 30 years of age, women with several past pregnancies, women who had lactated more than 10 months' total lifetime, women using oral contraceptives, and women with large breasts had a greater occurence of the vascular patterns. Women with abnormal breast characteristics (n = 100) and known lesions (fibrocystic disease, fibroadenoma, or carcinoma) were classified according to L.C. thermal pattern type. As expected, women with malignant lesions had a higher percentage of the vascular thermal pattern types.

No abstract