What is a software used for in thermography?

Technological advancements have allowed thermography to transition from a qualitative to a quantitative technology, thanks to analysis software. Let’s explore what that means.

Thermography benefits from special cameras that have a sensor sensitive to infrared radiation. When we point the camera towards an object, the sensor detects the infrared energy emitted by the object. This energy is invisible to us, but the camera converts it into a visible image, but where is the software?.

If we aim the camera at a person’s skin, we capture the radiation emitted by their body. Unlike other tests that apply radiation, thermography collects the radiation emitted by the body, making it harmless and non-invasive. Each object has a different emissivity, and human skin has an emissivity of 0.98.

The process of performing human thermography is simple and quick: the site is established, the subject is acclimated to expose the skin in the region to be analyzed, and the thermography is conducted. However, the process of analyzing the images can become complicated without the use of software.

The use of software for analyzing thermographic images

As seen in previous discussions, qualitative analysis of colors can provide an initial approximation and easily educate patients or athletes about their body’s condition. However, this methodology has significant biases, mainly due to the lack of significance in color. In other words, color alone does not provide meaningful data.

To perform a more accurate analysis, we need software. This software is primarily responsible for extracting the data within the pixels of the image to provide meaningful information. We can envision the thermographic image as an Excel spreadsheet, where each pixel grid represents a temperature data point (hence the importance of having acceptable pixel resolution in the camera).

To guide the software on where to find the important data, we have two options: manual segmentation of the regions or automatic segmentation. Opting for automatic segmentation can save a lot of time.

Once the segmentation is done, we have various methods to analyze the extracted information.

Are all software methodologies the same?

Once we have the extracted data, it is important to evaluate which metrics and statistics are the most reliable and valid for analyzing the images.

In doing so, scientific evidence should guide our decision-making process.

Scientific evidence tells us that there are many influencing factors (Fernandez-Cuevas 2015). Therefore, working with absolute temperature data, such as mean, maximum, and minimum temperatures, may be less reliable than working with relative data within the image itself. Factors of influence, such as camera and environmental variations, can easily affect absolute temperature data.

Another approach is pixelgraphy, which evaluates the pixel frequency within a manually segmented region to calculate the number of pixels in three predetermined temperature zones: cold zones (28°C to 31°C), neutral zones (31°C to 33°C), and hot zones (33°C to 36°C). However, working with absolute temperature data in thermography poses challenges.

The calculation of asymmetries is another method that involves calculating the difference in mean temperatures between segmented regions within the image or compared to other images to obtain a relative data point. This methodology relies on identifying laterality and, as demonstrated by both old and recent studies, is the most reliable, valid, and established methodology for the use of thermography in humans (Formenti 2018, Zhan et al. 2023).

As seen, the segmentation of body regions to identify temperature distribution is crucial for analysis. This segmentation can be more global (e.g., all legs) or more anatomical (e.g., muscle sections). Using segmentation in anatomical regions rather than global regions, such as the average temperature of legs or trunk, is more reliable and better for detecting changes in body thermoregulation (Hillen et al. 2023).

It is important for these regions to be included in different body region protocols. This is another reason to consider software: to assess how many protocols can be analyzed and how the camera is positioned to capture regions and optimize image resolution.

Lastly, visualizing these data and metrics can be easier if different reports are configured to present the data in an accessible language.

What sets ThermoHuman apart?

ThermoHuman software allows for direct capture from the analysis window of the protocols, eliminating the intermediate step of downloading and uploading images in the process.

The analysis is automatic, segmenting over 200 body regions from 8 protocols (with the facial protocol soon to be launched), thereby improving the regional identification of potential risk areas or pathologies.

The main metrics are body asymmetries and coefficients of variation. Asymmetries are the most researched and reliable metric applied to humans. The uniqueness of this metric lies in its individualization within the image, which allows the elimination of most influencing factors. The main drawback is that it only identifies unilateral problems and not bilateral ones.

Selective reports according to the application, such as prevention, fatigue, or injury monitoring, allow for a quick analysis of individuals or groups in an intuitive way.

Furthermore, two of ThermoHuman’s major milestones in terms of business and software are having scientifically demonstrated the software’s validity, improving inter-analysis reliability, and obtaining the medical device certification, which allows for European Community marking and compliance with product quality requirements.

References

Fernández-Cuevas, I., Marins, J. C. B., Lastras, J. A., Carmona, P. M. G., Cano, S. P., García-Concepción, M. Á., & Sillero-Quintana, M. (2015). Classification of factors influencing the use of infrared thermography in humans: A review. Infrared Physics & Technology, 71, 28-55.

Formenti et al. (2018) Is the maximum value in the region of interest a reliable indicator of skin temperature? June 2018 Infrared Physics & Technology 94

Zhang, H. Y., Son, S., Yoo, B. R., & Youk, T. M. (2023). Reference Standard for Digital Infrared Thermography of the Surface Temperature of the Lower Limbs. Bioengineering, 10(3), 283.

Hillen, B., López, D. A., Marzano-Felisatti, J. M., Sanchez-Jimenez, J. L., de Anda, R. M. C. O., Nägele, M., … & Priego-Quesada, J. I. (2023). Acute physiological responses to a pyramidal exercise protocol and the associations with skin temperature variation in different body areas. Journal of Thermal Biology, 103605.