Abstract
BACKGROUND: Diabetic foot ulcers (DFUs) are a life-changing complication of diabetes. There is increasing evidence that remote plantar temperature monitoring can reduce the recurrence of DFUs. Monitoring of foot temperature once a day is the current guideline for identifying early signs of foot inflammation. However, single readings of physiological signals can increase the risk of misdiagnosis when the signals fluctuate throughout the day. OBJECTIVE: The aim of this study was to evaluate whether intraday temperature asymmetry signals were stable or varied as a function of time in individuals at risk of DFUs. METHODS: In total, 64 participants with diabetes (mean age 68, SD 13.8 y) were provided with multimodal sensory insoles (Orpyx Sensory Insoles) to monitor continuous temperature data at a frequency of once per minute at 5 plantar locations in a 90-day study window. The augmented Dickey-Fuller test was used to determine whether the temperature asymmetry signals were stationary (ie, indicating constant mean and variance over time) or nonstationary (ie, indicating time-varying behaviors or trends in the signal). RESULTS: The study included 43 participants, 1080 data days, and 5400 contralateral temperature asymmetry signals. Most (4428/5400, 82%) of the temperature asymmetry signals were nonstationary, with intraday fluctuations likely influenced by physiological and environmental factors. Of the nonstationary signals, nearly half (1948/4428, 44%) fluctuated above and below the concerning asymmetry threshold of 2.2 °C. The intraday variability underscores the potential for false-positive and false-negative hot spot detection with once-daily measurements. Substantial variability was observed in stationarity patterns both within and across participants. Notably, concerning asymmetries in nonstationary signals occurred at different time points across participants, measurement windows, and days. CONCLUSIONS: Our findings highlight the value of continuous plantar temperature monitoring for diabetic foot health management, relative to once-daily measurements. Several repeated measurements throughout the day increase confidence with regard to the accuracy of observed plantar physiology trends. Continuous monitoring may improve the accuracy of plantar temperature measurement, unlock new diagnostic capabilities, and support personalized care.