Thermographic imaging of the pregnant uterus : the evaluation of a novel device and its use in the assessment of intra-uterine characteristics during pregnancy

Abstract

In humans, heat production and heat dissipation is regulated by metabolic and vascular mechanisms which maintain a delicate equilibrium between the body temperature and the surrounding environmental temperature. In pregnancy, an increase in maternal metabolic rate and heat dissipation from the foeto-placental unit are added factors resulting in heat production. Infrared thermography is a technique to record the naturally emitted infrared radiation from the area studied, making it a safe, non-invasive, non-contact device which can be used to evaluate the temperature patterns over the pregnant abdomen. The principle goal of this study was to assess the use of thermographic technology to acquire abdominal skin temperature data and explore if this can be related to ultrasound imaging measurements in pregnant women. The research was divided into four stages: (i) a preliminary pilot study to empirically assess the feasibility of taking thermographic measurements and correlating them with the deepest liquor pool measured on ultrasound. This was conducted on 46 patients and provided a basis for adjustments and refining of the methodology used in the second phase of the study; (ii) a main study which investigated a cohort of 42 healthy pregnant women of gestational age between 28 and 31 weeks. The objective was to investigate the presence and quantify any relationship between the temperature readings on thermographic imaging and the liquor pool measurements on ultrasound. The effect of maternal and foetal factors on this relationship was also analysed. Furthermore the time profile for acclimatisation of these patients to ambient temperature was explored; (iii) a second study investigated a group of 10 patients suffering from diabetes during pregnancy. The same relationships between thermography data and ultrasound measurements as in the main study were investigated together with acclimatisation profiles of these patients. Results from this cohort and the non-diabetic cohort were compared; (iv) an intra-rater and test-retest reliability study was conducted on the methodology of using the thermographic software to process image data in this obstetrics application. The results showed positive correlations between the temperature measurements carried out by thermography and the liquor pool depths on ultrasound in both healthy and diabetic patients. The application of cream on the abdominal surface was found to be a factor which affects this relationship in both groups, whilst the foetal abdominal circumference was found to have an effect in the diabetic group. The acclimatisation period required for the two groups was determined by analysing the statistical significance of the relationship over time points. A predicative model based on this relationship was generated for both groups. Both reliability studies showed excellent levels of agreement in the method of processing thermographic data by different users and by the same operator during distinct study days. The motivation for this research undertaken was to assess if thermography can be considered as a useful modality in obstetrics in humans. The study provided new insights into the way this device can potentially be used in this medical field, and makes an original contribution to the research on medical thermography by providing an added understanding to the way thermographic data can be interpreted when paralleled with conventional methods of analysis such as ultrasound.