An Overview of Thermal Infrared Imaging-Based Screenings during Pandemic Emergencies

The topical application of different galenic formulations can alter the thermographic images of skin: Limitations for public thermal screening on infection control situations

BACKGROUND

To analyze whether the application of topical formulas as cosmetics or sunscreens could affect the skin thermographic readings in terms of infection control in pandemic situations.

METHODS

The temperature of the skin of the dorsal region of the back and the face of 20 volunteers was followed after the application of 6 different types of gels, sunscreens, and make-up under controlled temperature and humidity conditions. High-resolution thermographic images were analyzed to calculate the temperature of treated skin compared to skin free of topical products.

RESULTS

The application of hydroalcoholic gel resulted in a mean drop of more than 2°C just after 1 minute followed by organic sunscreens until 1.7°C. Recovery was observed progressively until minute 9. Color make-up type formulas, rich in iron oxide as well as sunscreens with mineral filters had little or no effect on the skin thermal response.

CONCLUSIONS

It is possible to alter the skin temperature almost immediately by using hydroalcoholic gels and sunscreen cosmetics. So, it is possible to produce false negative data in the readings of patients screened thermically.

Setting up a three‐stage pre‐endoscopy triage during the coronavirus disease 2019 pandemic: A multicenter observational study

Objectives

Between May and July 2021, the coronavirus disease 2019 (COVID‐19) pandemic led to a sharp surge in community transmission in Taiwan. We present a three‐stage restructuring process of pre‐endoscopy triage at the beginning of the pandemic, which can support urgent endoscopic procedures while protecting endoscopy staff.

Methods

The pre‐endoscopy triage framework was set up with three checkpoints at the hospital entrance, outpatient department, and endoscopy unit, with a specific target patient population and screening methods. Relevant data included the number of endoscopic procedures performed, outpatient department visits, and performing screening methods such as temperature measurement, travel, occupation, contact, and clustering history checking, polymerase chain reaction assay, and rapid antigen test.

Results

Forehead temperature measurement and verification of travel, occupation, contact, and clustering history provided rapid, easy, and early mass screening of symptomatic patients at the hospital entrance. During the pandemic, outpatient department visits and endoscopic procedures decreased by 37% and 64%, respectively. The pre‐endoscopy screening methods used displayed regional variations in COVID‐19 prevalence. Among 16 endoscopy units with a community prevalence of ≥ 31.04 cases per 100,000 residents, 12 (75%) used polymerase chain reaction assay and four (25%) used rapid antigen test to identify asymptomatic patients before endoscopy. Of 6540 pre‐endoscopy screening patients, 15 (0.23%) tested positive by laboratory testing. No endoscopy‐related nosocomial COVID‐19 infections were reported during the pandemic.

Conclusions

We present a three‐stage pre‐endoscopy triage based on the local laboratory capacity, medical resources, and community prevalence. These measures could be useful during the COVID‐19 pandemic.

Measurement and Processing of Thermographic Data of Passing Persons for Epidemiological Purposes

Non-contact temperature measurement of persons during an epidemic is the most preferred measurement option because of the safety of personnel and minimal possibility of spreading infection. The use of infrared (IR) sensors to monitor building entrances for infected persons has seen a major boom between 2020 and 2022 due to the COVID-19 epidemic, but with questionable results. This article does not deal with the precise determination of the temperature of an individual person but focuses on the possibility of using infrared cameras for monitoring the health of the population. The aim is to use large amounts of infrared data from many locations to provide information to epidemiologists so they can have better information about potential outbreaks. This paper focuses on the long-term monitoring of the temperature of passing persons inside public buildings and the search for the most appropriate tools for this purpose and is intended as the first step towards creating a useful tool for epidemiologists. As a classical approach, the identification of persons based on their characteristic temperature values over time throughout the day is used. These results are compared with the results of a method using artificial intelligence (AI) to evaluate temperature from simultaneously acquired infrared images. The advantages and disadvantages of both methods are discussed.

Can Adipose Tissue Influence the Evaluation of Thermographic Images in Adolescents?

Infrared thermography (IRT) is a technology easy to use for clinical purposes as a pre-diagnostic tool for many health conditions. However, the analysis process of a thermographic image needs to be meticulous to make an appropriate decision. The adipose tissue is considered a potential influence factor in the skin temperature (Tsk) values obtained by IRT. This study aimed to verify the influence of body fat percentage (%BF) on Tsk measured by IRT in male adolescents. A total of 100 adolescents (16.79 ± 0.97 years old and body mass index of 18.41 ± 2.32 kg/m²) was divided into two groups through the results of a dual-energy X-ray absorptiometry analysis: obese (n = 50, %BF 30.21 ± 3.79) and non-obese (n = 50, %BF 11.33 ± 3.08). Thermograms were obtained by a FLIR T420 infrared camera and analyzed by ThermoHuman^® software version 2.12, subdividing the body into seven regions of interest (ROI). The results showed that obese adolescents presented lower mean Tsk values than the non-obese for all ROIs (p < 0.05), with emphasis on the global Tsk (0.91 °C) and anterior (1.28 °C) and posterior trunk (1.18 °C), with "very large" effect size values. A negative correlation was observed in all the ROI (p < 0.01), mainly in the anterior (r = -0.71, p < 0.001) and posterior trunk (r = -0.65, p < 0.001). Tables of thermal normality were proposed for different ROIs according to the classification of obesity. In conclusion, the %BF affects the registered Tsk values in male Brazilian adolescents assessed by IRT.

A survey on computer vision based human analysis in the COVID-19 era

The emergence of COVID-19 has had a global and profound impact, not only on society as a whole, but also on the lives of individuals. Various prevention measures were introduced around the world to limit the transmission of the disease, including face masks, mandates for social distancing and regular disinfection in public spaces, and the use of screening applications. These developments also triggered the need for novel and improved computer vision techniques capable of ( i ) providing support to the prevention measures through an automated analysis of visual data, on the one hand, and ( ii ) facilitating normal operation of existing vision-based services, such as biometric authentication schemes, on the other. Especially important here, are computer vision techniques that focus on the analysis of people and faces in visual data and have been affected the most by the partial occlusions introduced by the mandates for facial masks. Such computer vision based human analysis techniques include face and face-mask detection approaches, face recognition techniques, crowd counting solutions, age and expression estimation procedures, models for detecting face-hand interactions and many others, and have seen considerable attention over recent years. The goal of this survey is to provide an introduction to the problems induced by COVID-19 into such research and to present a comprehensive review of the work done in the computer vision based human analysis field. Particular attention is paid to the impact of facial masks on the performance of various methods and recent solutions to mitigate this problem. Additionally, a detailed review of existing datasets useful for the development and evaluation of methods for COVID-19 related applications is also provided. Finally, to help advance the field further, a discussion on the main open challenges and future research direction is given at the end of the survey. This work is intended to have a broad appeal and be useful not only for computer vision researchers but also the general public.

Can the body mass index influence the skin temperature of adolescents assessed by infrared thermography?

Infrared thermography (IRT) is a technology that has been used as an auxiliary tool in the diagnostic process of several diseases and in sports monitoring to prevent injuries. However, the evaluation of a thermogram can be influenced by several factors that need to be understood and controlled to avoid a misinterpretation of the thermogram and, consequently, an inappropriate clinical action. Among the possible factors that can affect IRT are anthropometric factors, especially those related to body composition. Based on these, our objective was to verify the influence of Body Mass Index (BMI) on skin temperature (Tsk) in male adolescents. One hundred male adolescents (age: 16.83 ± 1.08 years; body mass: 66.51 ± 13.35 kg; height: 1.75 ± 7.04 m and BMI: 21.57 ± 4.06 kg/m²) were evaluated and divided into three groups, based on the World Health Organization (WHO) proposed classification ranges: underweight (n = 33), normal weight (n = 34) and overweight/obesity (n = 34). Thermograms were obtained using the FLIR T420 thermal imager after a period of acclimatization of the subjects in a controlled environment (temperature: 21.3 ± 0.7 °C and humidity: 55.3 ± 2.2%); they were evaluated using the ThermoHuman® software, integrating the original regions of interest (ROI) into seven larger ROIs. The results showed that underweight individuals had higher Tsk values than normal weight and overweight/obese individuals for all evaluated ROIs, and overweight/obese individuals had lower Tsk values than normal weight individuals for most evaluated ROIs, except for arms region. BMI showed a correlation of -0.68 and -0.64 for the anterior and posterior regions of the trunk, respectively. Thermal normality tables were proposed for various ROIs according to BMI classification. Our study demonstrated that BMI can affect the Tsk values assessed by IRT and needs to be considered to interpret the thermograms.

Relative Stability of Regional Facial and Ocular Temperature Measurements in Healthy Individuals

Purpose

Non-contact measurement of facial temperature using infrared thermography has been used for mass screening of body temperature during a pandemic. We investigated the relative stability of temperature measurement in different facial regions of healthy individuals.

Methods

Twenty healthy subjects underwent two experiments. In the first experiment, subjects washed their faces with a 20°C wet towel for 1 minute. Temperature changes compared to baseline in the forehead, cornea, inner canthus, and outer canthus were determined using an infrared camera for 10 minutes. In the second experiment, lubricating eye drops at 20°C were instilled over one eye. Temperature changes in the same regions of interest were monitored for 5 minutes.

Results

Baseline temperatures before face washing in the forehead and cornea, inner canthus, and outer canthus of the right eye were 33.4°C ± 0.8°C (mean ± SD), 33.3°C ± 0.8°C, 34.3°C ± 0.7°C, and 32.8°C ± 0.7°C, respectively. Reductions in temperature due to face washing were most significant for the forehead and least significant for the cornea. One minute after face washing, the corresponding changes were -2.8°C ± 0.6°C, -0.3°C ± 0.6°C, -0.6°C ± 0.7°C, and -0.9°C ± 0.7°C for the forehead, cornea, inner canthus, and outer canthus, respectively. After administering the eye drops, no significant temperature changes were observed.

Conclusions

When facial temperature was exogenously cooled, the cornea had the most stable temperature readings.

Translational Relevance

When using infrared thermography to screen facial temperature, the measurement of corneal temperature is probably a better representative if the stability of temperature readings is critical.

Healthcare Supply Chain Management under COVID-19 Settings: The Existing Practices in Hong Kong and the United States

COVID-19 is recognized as an infectious disease generated by serious acute respiratory syndrome coronavirus 2. COVID-19 has rapidly spread all over the world within a short time period. Due to the coronavirus pandemic transmitting quickly worldwide, the impact on global healthcare systems and healthcare supply chain management has been profound. The COVID-19 outbreak has seriously influenced the routine and daily operations of healthcare facilities and the entire healthcare supply chain management and has brough about a public health crisis. As making sure the availability of healthcare facilities during COVID-19 is crucial, the debate on how to take resilience actions for sustaining healthcare supply chain management has gained new momentum. Apart from the logistics of handling human remains in some countries, supplies within the communities are urgently needed for emergency response. This study focuses on a comprehensive evaluation of the current practices of healthcare supply chain management in Hong Kong and the United States under COVID-19 settings. A wide range of different aspects associated with healthcare supply chain operations are considered, including the best practices for using respirators, transport of life-saving medical supplies, contingency healthcare strategies, blood distribution, and best practices for using disinfectants, as well as human remains handling and logistics. The outcomes of the conducted research identify the existing healthcare supply chain trends in two major Eastern and Western regions of the world, Hong Kong and the United States, and determine the key challenges and propose some strategies that can improve the effectiveness of healthcare supply chain management under COVID-19 settings. The study highlights how to build resilient healthcare supply chain management preparedness for future emergencies.

Estimation of Heart Rate Variability Parameters by Machine Learning Approaches Applied to Facial Infrared Thermal Imaging

Heart rate variability (HRV) is a reliable tool for the evaluation of several physiological factors modulating the heart rate (HR). Importantly, variations of HRV parameters may be indicative of cardiac diseases and altered psychophysiological conditions. Recently, several studies focused on procedures for contactless HR measurements from facial videos. However, the performances of these methods decrease when illumination is poor. Infrared thermography (IRT) could be useful to overcome this limitation. In fact, IRT can measure the infrared radiations emitted by the skin, working properly even in no visible light illumination conditions. This study investigated the capability of facial IRT to estimate HRV parameters through a face tracking algorithm and a cross-validated machine learning approach, employing photoplethysmography (PPG) as the gold standard for the HR evaluation. The results demonstrated a good capability of facial IRT in estimating HRV parameters. Particularly, strong correlations between the estimated and measured HR (r = 0.7), RR intervals (r = 0.67), TINN (r = 0.71), and pNN50 (%) (r = 0.70) were found, whereas moderate correlations for RMSSD (r = 0.58), SDNN (r = 0.44), and LF/HF (r = 0.48) were discovered. The proposed procedure allows for a contactless estimation of the HRV that could be beneficial for evaluating both cardiac and general health status in subjects or conditions where contact probe sensors cannot be used.

Low-cost thermal imaging with machine learning for non-invasive diagnosis and therapeutic monitoring of pneumonia

Rapid screening and early treatment of lung infection are essential for effective control of many epidemics such as Coronavirus Disease 2019 (COVID-19). Recent studies have demonstrated the potential correlation between lung infection and the change of back skin temperature distribution. Based on these findings, we propose to use low-cost, portable and rapid thermal imaging in combination with image-processing algorithms and machine learning analysis for non-invasive and safe detection of pneumonia. The proposed method was tested in 69 subjects (30 normal adults, 11 cases of fever without pneumonia, 19 cases of general pneumonia and 9 cases of COVID-19) where both RGB and thermal images were acquired from the back of each subject. The acquired images were processed automatically in order to extract multiple location and shape features that distinguish normal subjects from pneumonia patients at a high accuracy of 93 % . Furthermore, daily assessment of two pneumonia patients by the proposed method accurately predicted the clinical outcomes, coincident with those of laboratory tests. Our pilot study demonstrated the technical feasibility of portable and intelligent thermal imaging for screening and therapeutic assessment of pneumonia. The method can be potentially implemented in under-resourced regions for more effective control of respiratory epidemics.

COVID-19 classification using thermal images

SIGNIFICANCE

There is a scarcity of published research on the potential role of thermal imaging in the remote detection of respiratory issues due to coronavirus disease-19 (COVID-19). This is a comprehensive study that explores the potential of this imaging technology resulting from its convenient aspects that make it highly accessible: it is contactless, noninvasive, and devoid of harmful radiation effects, and it does not require a complicated installation process.

AIM

We aim to investigate the role of thermal imaging, specifically thermal video, for the identification of SARS-CoV-2-infected people using infrared technology and to explore the role of breathing patterns in different parts of the thorax for the identification of possible COVID-19 infection.

APPROACH

We used signal moment, signal texture, and shape moment features extracted from five different body regions of interest (whole upper body, chest, face, back, and side) of images obtained from thermal video clips in which optical flow and super-resolution were used. These features were classified into positive and negative COVID-19 using machine learning strategies.

RESULTS

COVID-19 detection for male models [receiver operating characteristic (ROC) area under the ROC curve (AUC) = 0.605 95% confidence intervals (CI) 0.58 to 0.64] is more reliable than for female models (ROC AUC = 0.577 95% CI 0.55 to 0.61). Overall, thermal imaging is not very sensitive nor specific in detecting COVID-19; the metrics were below 60% except for the chest view from males.

CONCLUSIONS

We conclude that, although it may be possible to remotely identify some individuals affected by COVID-19, at this time, the diagnostic performance of current methods for body thermal imaging is not good enough to be used as a mass screening tool.

Comparative Analysis of Human Body Temperatures Measured with Noncontact and Contact Thermometers

Body temperature measurement is one of the basic methods in clinical diagnosis. The problems of thermometry—interpretation of the accuracy and repeatability of various types of thermometers—are still being discussed, especially during the current pandemic in connection with the SARS-CoV-2 virus responsible for causing the COVID-19 disease. The aim of the study was to compare surface temperatures of the human body measured by various techniques, in particular a noncontact thermometer (infrared) and contact thermometers (mercury, mercury-free, electronic). The study included 102 randomly selected healthy women and men (age 18–79 years). The Bland–Altman method was used to estimate the 95% reproducibility coefficient, i.e., to assess the degree of conformity between different attempts. Temperatures measured with contact thermometers in the armpit are higher than temperatures measured without contact at the frontal area of the head. The methods used to measure with contact thermometers and a noncontact infrared thermometer statistically showed high measurement reliability. In order to correctly interpret the result of measuring human body temperature, it is necessary to indicate the place of measurement and the type of thermometer used.

Pyrometry with flexible infrared fibers for temperature-controlled laser surgery

Pyrometry is widely used in science, medicine, and industry to measure the surface temperature of objects in a non-contact way. IR fibers are an ideal solution for the flexible delivery of thermal radiation emitted from objects inside a complex structure like internal organs inside the human body. Silver halide polycrystalline infrared fibers (PIR) are transparent in a spectral range of 3 - 18 µm, matching perfectly with the spectra of black body radiation for temperatures ranging from 20°C to 200°C. These fibers are non-toxic and allow small bending radii. They could become critical components in pyrometric systems for temperature-controlled laser surgeries. Here we discuss the ability of the PIR fibers for simultaneous laser power delivery and real-time temperature monitoring in laser surgery applications and demonstrate two different setups for this purpose.

Progress on Infrared Imaging Technology in Animal Production: A Review

Infrared thermography (IRT) imaging technology, as a convenient, efficient, and contactless temperature measurement technology, has been widely applied to animal production. In this review, we systematically summarized the principles and influencing parameters of IRT imaging technology. In addition, we also summed up recent advances of IRT imaging technology in monitoring the temperature of animal surfaces and core anatomical areas, diagnosing early disease and inflammation, monitoring animal stress levels, identifying estrus and ovulation, and diagnosing pregnancy and animal welfare. Finally, we made prospective forecast for future research directions, offering more theoretical references for related research in this field.

SARS-CoV-2 (Covid-19) workplace temperature screening: Seasonal concerns for thermal detection in northern regions

Workplace temperature screening has become standard practice during the SARS-CoV-2 pandemic. The objective was to determine the consistency of four temperature devices during exposure to simulated and actual environmental conditions reflective of a workplace. An infrared (IR) digital thermometer (accuracy(A)±0.2), IR laser thermometer (A±1), and thermal imaging camera (A±0.3) were used to measure forehead and tympanic (digital only) temperatures. The first experiment was conducted in a controlled simulated environment (-20 to 20 °C) with three participants (32-YOF, 27-YOM, 20-YOF). The second experiment used actual outdoor conditions (-0.48 to 45.6 °C) with two participants (32-YOF, 27-YOM). The tympanic measurement was the least impacted by environmental temperature (mean(±SD)): simulated (36.8(±0.18) °C) and actual (36.9(±0.16) °C). The thermal imaging camera had the lowest RMSE values (0.81-0.97 °C), with outdoor temperatures ranging from 0 to 45 °C. Environmental temperature influenced forehead temperature readings and required a resting period in a thermoneutral environment (5-9 min (-20 to -10 °C) to immediate (15-20 °C)).

Can body mass index influence the skin temperature of adolescents? A preliminary study with the use of infrared thermography

Abstract Infrared thermography (IRT) has been used to assess skin temperature (Tsk), especially during the COVID-19 pandemic, as an important tool in medical screening not only of the general population, but also of young athletes. However, the subcutaneous adipose tissue can act as an insulator when the Tsk is assessed by IRT, modifying the normal Tsk data and leading to their misinterpretation. Considering that the body mass index (BMI) is an important predictor of obesity, the objective of this study was to verify if the Tsk measured by IRT is affected by the BMI in adolescents. A preliminary study was carried out being four participants intentionally selected, all 16 years old, each one classified in a different BMI range according to the criteria of the World Health Organization for the adolescent population: underweight, healthy weight, overweight and obesity. Four thermograms of each participant were recorded and the ThermoHuman® software was used to evaluate 82 regions of interest (ROI), which were integrated into 6 body regions. Using healthy weight subjects as a reference, it was found a progressive reduction in Tsk in all ROI compared to overweight and obese participants, with emphasis on the anterior region of the trunk (3.04% and 6.69% less respectively), and an increase in the Tsk of all body regions for the underweight subject. There are indications that BMI can influence the Tsk value in adolescents and should be taken into account when analyzing thermograms for a correct evaluation of thermal normality.

The environment has effects on infrared temperature screening for COVID-19 infection

Infrared temperature measurement is a common form of mass screening for febrile illnesses such as COVID-19 infection. Efficacy of infrared monitoring is debated, and external factors can affect accuracy. We determine that outside temperature, wind, and humidity can affect infrared temperature measurements and partially account for inaccurate results.

Regions of interest selection and thermal imaging data analysis in sports and exercise science: a narrative review

OBJECTIVE

Infrared thermography (IRT) is a non-invasive, contactless and low-cost technology that allows recording of the radiating energy that is released from a body, providing an estimate of its superficial temperature. Thanks to the improvement of infrared thermal detectors, this technique is widely used in the biomedical field to monitor the skin temperature for different purposes (e.g. assessing circulatory diseases, psychophysiological state, affective computing). Particularly, in sports and exercise science, thermography is extensively used to assess sports performance, to investigate superficial vascular changes induced by physical exercise, and to monitor injuries. However, the methods of analysis employed to treat IRT data are not standardized, and hence introduce variability in the results.

APPROACH

This review focuses on the methods of analysis currently used for thermal imaging in sports and exercise science.

MAIN RESULTS

Firstly, the procedures employed for the selection of regions of interest (ROIs) from anatomical body districts are reviewed, paying attention also to the potentialities of morphing algorithms to increase the reproducibility of thermal results. Secondly, the statistical approaches utilized to characterize the temperature frequency and spatial distributions within ROIs are investigated, showing their strengths and weaknesses. Moreover, the importance of employing tracking methods to analyze the temporal thermal oscillations within ROIs is discussed. Thirdly, the capability of employing procedures of investigation based on machine learning frameworks on thermal imaging in sports science is examined.

SIGNIFICANCE

Finally, some proposals to improve the standardization and the reproducibility of IRT data analysis are provided, in order to facilitate the development of a common database of thermal images and to improve the effectiveness of IRT in sports science.

Infrared Thermal Imaging of Patients With Acute Upper Respiratory Tract Infection: Mixed Methods Analysis

Background

Upper respiratory tract infection is a common disease of the respiratory system. Its incidence is very high, and it can even cause pandemics. Infrared thermal imaging (IRTI) can provide an objective and quantifiable reference for the visual diagnosis of people with acute respiratory tract infection, and it can function as an effective indicator of clinical diagnosis.

Objective

The aims of this study are to observe and analyze the infrared expression location and characteristics of patients with acute upper respiratory tract infection through IRTI technology and to clearly express the quantification of temperature, analyze the role of IRTI in acute upper respiratory tract diagnostic research, and understand the impact of IRTI in qualitative and quantitative research.

Methods

From December 2018 to February 2019, 154 patients with acute upper respiratory tract infection were randomly selected from the emergency department of the First Affiliated Hospital of Guangzhou Medical University. Among these patients, 73 were men and 81 were women. The subjects were divided into two groups according to the presence of fever, namely, fever and nonfever groups. Qualitative and quantitative analyses of the infrared thermal images were performed to compare the results before and after application of the technology.

Results

Using the method described in this paper, through the analysis of experimental data, we elucidated the role of IRTI in the diagnosis of acute upper respiratory tract infection, and we found that qualitative and quantitative IRTI analyses play important roles. Through the combination of theory and experimental data, the IRTI analysis showed good results in identifying acute upper respiratory tract infection.

Conclusions

IRTI technology plays an important role in identifying the infrared expression location and characteristics of patients with acute upper respiratory tract infection as well as in the quantification of clear expression of body temperature, and it provides an objective and quantifiable reference basis for elucidating the pathogenesis of these patients.

Infrared Thermography as a Method of Verification in Raynaud's Phenomenon

Raynaud's phenomenon (RP) is characterized by the episodic whitening of the fingers upon exposure to cold. A recently described thermographic algorithm was proposed as a diagnostic replacement of the currently applied finger systolic pressure (FSP) test. The aim of the study was to evaluate the performance of the thermographic algorithm when applied in patients suspected of having RP. Forty-three patients were examined using thermographic imaging after local cooling of the hands in water of 10 °C for 1 min. The thermographic algorithm was applied to predict the probability of RP. The performance of the algorithm was evaluated with different cut-off levels. A new algorithm was proposed based on patients from the target population. The performance of the tested algorithm was noninferior to the FSP test, when a cut-off level of 0.05 was applied, yielding a sensitivity and specificity of 69% and 58%, respectively. The accuracy was 66%. The FSP test had a sensitivity and specificity of 77% and 37%, respectively, and the accuracy was 59%. The thermographic method proved useful for detecting RP and was able to replace the FSP test as a diagnostic test. The alternative algorithm revealed that other thermographic variables were more predictive of the target population, but this should be verified in future patients.

Convolutional Neural Networks for Differential Diagnosis of Raynaud’s Phenomenon Based on Hands Thermal Patterns

Raynaud’s phenomenon (RP) is a microvessels’ disorder resulting in transient ischemia. It can be either primary or secondary to connective tissue diseases, such as systemic sclerosis. The differentiation between primary and secondary to systemic sclerosis is of paramount importance to set the proper therapeutic strategy. Thus far, thermal infrared imaging has been employed to accomplish this task by monitoring the finger temperature response to a controlled cold challenge. A completely automated methodology based on deep convolutional neural network is here introduced with the purpose of being able to differentiate systemic sclerosis from primary RP patients by relying uniquely on thermal images of the hands acquired at rest. The classification performance of such a method was compared to that of a three-dimensional convolutional neural network model implemented to classify thermal images of the hands recorded during rewarming from a cold challenge. No significant differences were found between the two procedures, thus ensuring the possibility to avoid the cold challenge. Moreover, the convolutional neural network models were compared with standard feature-based approaches and showed higher performances, thus overcoming the limitations related to the feature extraction (e.g., biases introduced by the operator). Such automated procedures can constitute promising tools for large scale screening of primary RP and secondary to systemic sclerosis in clinical practice.