Pregnant vs. non-pregnant women: review of the literature on thermography
Today we bring you some studies on pregnant women in which it is evident that this state modifies the temperature of the skin compared to women who are not in a pregnancy process.
It is more than known that pregnant women have a higher metabolic rate, that is, they have a higher energy expenditure to maintain basic functions (Butte, N.F., 2005). In thermography, since 1990, Ernst Beinder et al., showed us the evidence of the increase in this metabolic rate by measuring the temperature of the skin of acral regions (mainly hands). Thus, they evidenced that, in pregnant women, the difference in temperature between forearms and hands was greater and that the hands suffered a lower thermal drop due to occlusion and cooling than non-pregnant women, as shown in Figure 1.
|Thermal difference hand-forearm (º C)||Decrease in temperature with occlusion (º C)||Decrease in temperature with cooling (º C)|
(n = 8)
|0,6 ± 1,2||0,4 ± 0,3||0,5 ± 0,3|
(n = 11)
|2,6 ± 0,8*||0,2 ± 0,1**||0,1 ± 0,1**|
Falzon, O. (2018) and collaborators demonstrated that by acquiring a sequence of thermal images, dynamic thermography can discover patterns of activity of special relevance related to maternal and fetal health. However, temporary recordings of regions of the human body, such as the abdomen, are likely to be affected by movements caused by breathing and other involuntary activities. This can make the manual extraction of temperatures in a thermal imaging sequence intractable, especially as the number of images in the thermal video increases and the number of regions of interest (ROI) to consider increases. Image registration can solve this problem by aligning the corresponding spatial points in the image sequence, thus facilitating automated temperature extraction and analysis.
Specifically, not only are the pixel intensities in a thermal image sequence expected to change due to movements of the ROI, but also due to the thermal patterns themselves, which can change considerably throughout the sequence. Algorithms for recording thermal images are not readily available, and studies involving monitoring of temperature changes in ROIs on the skin are based on the presence of hot spots on the skin (Simões, R. 2011) or in the placement of physical markers (Simões, R. 2012) to help in ROI localization. Without the use of external surface markers, the homogeneity of regions such as the abdomen, together with their non-rigid movements, make it difficult to consistently identify distinctive features locally throughout the image sequence.
As can be seen in image 1, in the work of Falzon, O. (2018), propose a triangulation-based video registration technique that uses affine transformations for the automated and marker-free recording of dynamic thermal sequences of the abdominal region of pregnant and non-pregnant women. While the proposal is particularly relevant to obstetrics, the same procedure can be extended to other applications requiring ROI recording of the human body that lack distinctive features and exhibit non-rigid movements.
To facilitate temperature control within a region of interest, Ciantar, A. (2018) and collaborators have developed an automated thermal video recording process that can compensate for the mother’s breathing movements, even if these they are non-rigid in nature, as in the case of the abdomen. In image 2, we can see a record lasting one hour, where the temperature variations in any spatial location of the abdomen can be controlled considering the corresponding pixel or group of pixels, which remain unchanged throughout the entire sequence of images.
A pregnant woman shows an increased general metabolic rate, estimated at 375 kJ per day (first trimester), 1200 kJ per day (second trimester), 1950 kJ per day (third trimester), which is reflected in a significant thermal increase, both in acral regions and in the belly region. Therefore, the caloric intake recommendations, as well as periodic measurements with thermography, should be reviewed frequently to maintain correct health, both in the baby and in the mother.
Beinder E, Huch A, Huch R. Peripheral skin temperature and microcirculatory reactivity during pregnancy. A study with thermography. J Perinat Med. 1990;18(5):383-90. doi: 10.1515/jpme.1922.214.171.1243. PMID: 2292760.
Butte NF, King JC. Energy requirements during pregnancy and lactation. Public Health Nutr. 2005 Oct;8(7A):1010-27. doi: 10.1079/phn2005793. PMID: 16277817.
Ciantar A, Falzon O, Sammut L, Schembri M, Baron YM, Calleja-Agius J, Pierre Demicoli P, Camilleri KP. Registration of Dynamic Thermography Data of the Abdomen of Pregnant and Non-Pregnant Women. Annu Int Conf IEEE Eng Med Biol Soc. 2018 Jul;2018:5668-5671.
Falzon O, Ciantar A, Sammut L, Schembri M, Baron YM, Calleja-Agius J, Demicoli P, Kenneth Camilleri P. Principal Component Analysis of Dynamic Thermography Data from Pregnant and Non-Pregnant Women. Annu Int Conf IEEE Eng Med Biol Soc. 2018 Jul;2018:5664-5667.
Simões R, Nogueira-Silva C. The use of medical thermal imaging in obstetrics. Computational Vision and Medical
Image Processing: VipIMAGE. 2011:285.
Simões R, Nogueira-Silva C, Vardasca R. Thermal skin reference values in healthy late pregnancy. Journal of
Thermal Biology. 2012, 37(8):608-14.
If you have any comments or clarifications, do not hesitate to send us a message. We will be delighted to read you.