Detection theory for accurate and non-invasive skin cancer diagnosis using dynamic thermal imaging

Feasibility Study on the Use of Infrared Cameras for Skin Cancer Detection under a Proposed Data Degradation Model

Infrared thermography is considered a useful technique for diagnosing several skin pathologies but it has not been widely adopted mainly due to its high cost. Here, we investigate the feasibility of using low-cost infrared cameras with microbolometer technology for detecting skin cancer. For this purpose, we collected infrared data from volunteer subjects using a high-cost/high-quality infrared camera. We propose a degradation model to assess the use of lower-cost imagers in such a task. The degradation model was validated by mimicking video acquisition with the low-cost cameras, using data originally captured with a medium-cost camera. The outcome of the proposed model was then compared with the infrared video obtained with actual cameras, achieving an average Pearson correlation coefficient of more than 0.9271. Therefore, the model successfully transfers the behavior of cameras with poorer characteristics to videos acquired with higher-quality cameras. Using the proposed model, we simulated the acquisition of patient data with three different lower-cost cameras, namely, Xenics Gobi-640, Opgal Therm-App, and Seek Thermal CompactPRO. The degraded data were used to evaluate the performance of a skin cancer detection algorithm. The Xenics and Opgal cameras achieved accuracies of 84.33% and 84.20%, respectively, and sensitivities of 83.03% and 83.23%, respectively. These values closely matched those from the non-degraded data, indicating that employing these lower-cost cameras is appropriate for skin cancer detection. The Seek camera achieved an accuracy of 82.13% and a sensitivity of 79.77%. Based on these results, we conclude that this camera is appropriate for less critical applications.

The Role of Artificial Intelligence in Early Diagnosis and Molecular Classification of Head and Neck Skin Cancers: A Multidisciplinary Approach

Cancer remains a significant global health concern, with increasing genetic and metabolic irregularities linked to its onset. Among various forms of cancer, skin cancer, including squamous cell carcinoma, basal cell carcinoma, and melanoma, is on the rise worldwide, often triggered by ultraviolet (UV) radiation. The propensity of skin cancer to metastasize highlights the importance of early detection for successful treatment. This narrative review explores the evolving role of artificial intelligence (AI) in diagnosing head and neck skin cancers from both radiological and pathological perspectives. In the past two decades, AI has made remarkable progress in skin cancer research, driven by advances in computational capabilities, digitalization of medical images, and radiomics data. AI has shown significant promise in image-based diagnosis across various medical domains. In dermatology, AI has played a pivotal role in refining diagnostic and treatment strategies, including genomic risk assessment. This technology offers substantial potential to aid primary clinicians in improving patient outcomes. Studies have demonstrated AI's effectiveness in identifying skin lesions, categorizing them, and assessing their malignancy, contributing to earlier interventions and better prognosis. The rising incidence and mortality rates of skin cancer, coupled with the high cost of treatment, emphasize the need for early diagnosis. Further research and integration of AI into clinical practice are warranted to maximize its benefits in skin cancer diagnosis and treatment.

Is thermography an effective screening tool for differentiating benign and malignant skin lesions in the head and neck? A systematic review

The aim of this study was to assess, through a systematic review, the status of infrared thermography (IRT) as a diagnostic tool for skin neoplasms of the head and neck region and in order to validate its effectiveness in differentiating benign and malignant lesions. A search was carried out in the LILACS, PubMed/MEDLINE, SCOPUS, Web of Science and EMBASE databases including studies published between 2004 and 2024, written in the Latin-Roman alphabet. Accuracy studies with patients aged 18 years or over presenting benign and malignant lesions in the head and neck region that evaluated the performance of IRT in differentiating these lesions were included. Lesions of mesenchymal origin and studies that did not mention histopathological diagnosis were excluded. The systematic review protocol was registered in the PROSPERO database (CRD42023416079). Reviewers independently analyzed titles, abstracts, and full-texts. After extracting data, the risk of bias of the selected studies was assessed using the QUADAS - 2 tool. Results were narratively synthesized and the certainty of evidence was measured using the GRADE approach. The search resulted in 1,587 records and three studies were included. Only one of the assessed studies used static IRT, while the other two studies used cold thermal stress. All studies had an uncertain risk of bias. In general, studies have shown wide variation in the accuracy of IRT for differentiating between malignant and benign lesions, with a low level of certainty in the evidence for both specificity and sensitivity.

Continuous-Spectrum-Polarization Recombinant Optical Encryption with a Dielectric Metasurface

The Jones matrix, with eight degrees of freedom (DoFs), provides a general mathematical framework for the multifunctional design of metasurfaces. Theoretically, the maximum eight DoFs can be further extended in the spectrum dimension to endow unique encryption capabilities. However, the topology and intrinsic spectral responses of meta-atoms constrains the continuous engineering of polarization evolution over wavelength dimension. In this work, a forward evolution strategy to quickly establish the mapping relationships between the solutions of the dispersion Jones matrix and the spectral responses of meta-atoms is reported. Based on the eigenvector transformation method, arbitrary conjugate polarization channels over the continuous-spectrum dimension are successfully reconstructed. As a proof-of-concept, a silicon metadevice is demonstrated for optical information encryption transmission. Remarkably, the arbitrary combination forms of polarization and wavelength dimension increase the information capacity (210 ), and the measured polarization contrasts of the conjugate polarization conversion are >94% in the entire wavelength range (3-4 µm). It is believed that the proposed approach will benefit secure optical and quantum information technologies.

Automated segmentation and classification of hand thermal images in rheumatoid arthritis using machine learning algorithms: A comparison with quantum machine learning technique

The aims and objectives of the study were to i) perform image segmentation using a color-based k-means clustering algorithm and feature extraction using binary robust invariant scalable key points (BRISK), maximum stable extremal regions (MSER), features from accelerated segment test (FAST), Harris, and orientated FAST and rotated BRIEF (ORB); ii) compare the performance of classical machine learning techniques such as LogitBoost, Bagging, and SVM with a quantum machine learning technique. For the proposed study, 240 hand thermal images were acquired in the dorsal view and ventral view of both the right and left-hand regions of RA and normal subjects. The hot spot regions from the thermograms were segmented using a color-based k-means clustering technique. The features from the segmented hot spot region were extracted using different feature extraction methods. Finally, normal and RA groups were categorized using LogitBoost, Bagging, and support vector machine (SVM) classifiers. The proposed study used two testing methods, such as 10-fold cross-validation and a percentage split of 80-20%. The LogitBoost classifier outperformed with an accuracy of 93.75% using the 10-fold cross-validation technique compared to other classifiers. Also, the quantum support vector machine (QSVM) obtained a prediction accuracy of 92.7%. Furthermore, the QSVM model reduces the computational cost and training time of the model to classify the RA and normal subjects. Thus, thermograms with classical machine learning and quantum machine learning algorithms could be considered a feasible technique for classifying normal and RA groups.

Three-dimensional point spread function estimation method for mid-wave infrared microscope imaging

A three-dimensional point spread function experimental estimation method based on the system's focal plane array spatial local impulse response of a mid-wave infrared microscope is presented. The method uses several out-of-focus two-dimensional point spread function planes to achieve a single three-dimensional point spread function of the whole microscope's optical spreading, expanding the limits of infrared optical technology by one dimension. This technique includes stages of image acquisition, nonuniformity correction, filtering, and multi-planar reconstruction steps, and its effectiveness is demonstrated on biological sample image restoration by means of a multi-planar refocusing application.

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.

Skin Cancer Detection Using Infrared Thermography: Measurement Setup, Procedure and Equipment

Infrared thermography technology has improved dramatically in recent years and is gaining renewed interest in the medical community for applications in skin tissue identification applications. However, there is still a need for an optimized measurement setup and protocol to obtain the most appropriate images for decision making and further processing. Nowadays, various cooling methods, measurement setups and cameras are used, but a general optimized cooling and measurement protocol has not been defined yet. In this literature review, an overview of different measurement setups, thermal excitation techniques and infrared camera equipment is given. It is possible to improve thermal images of skin lesions by choosing an appropriate cooling method, infrared camera and optimized measurement setup.

Adjunctive Diagnostic Methods for Skin Cancer Detection: A Review of Electrical Impedance-Based Techniques

Skin cancer is among the fastest-growing cancers with an excellent prognosis, if detected early. However, the current method of diagnosis by visual inspection has several disadvantages such as overlapping tumor characteristics, subjectivity, low sensitivity, and specificity. Hence, several adjunctive diagnostic techniques such as thermal imaging, optical imaging, ultrasonography, tape stripping methods, and electrical impedance imaging are employed along with visual inspection to improve the diagnosis. Electrical impedance-based skin cancer detection depends upon the variations in electrical impedance characteristics of the transformed cells. The information provided by this technique is fundamentally different from other adjunctive techniques and thus has good prospects. Depending on the stage, type, and location of skin cancer, various impedance-based devices have been developed. These devices when used as an adjunct to visual methods have increased the sensitivity and specificity of skin cancer detection up to 100% and 87%, respectively, thus demonstrating their potential to minimize unnecessary biopsies. In this review, the authors track the advancements and progress made in this technique for the detection of skin cancer, focusing mainly on the advantages and limitations in the clinical setting. © 2022 Bioelectromagnetics Society.

Dynamic thermal imaging for pigmented basal cell carcinoma and seborrheic keratosis

BACKGROUND

Clinical differentiation between pigmented basal cell carcinoma (BCC) and seborrheic keratosis (SK) can sometimes be difficult. Noninvasive diagnostic technologies, such as thermal imaging, can be helpful in these situations. This study explored the use of dynamic thermal imaging (DTI), which records thermal images after the application of external thermal stimuli (heat or cold) for the differential diagnosis of pigmented BCC and SK.

MATERIALS AND METHODS

Twenty-two patients with pigmented BCC and 15 patients with SK participated in this study. Dynamic thermal images of lesions (pigmented BCC or SK) and control sites (contralateral normal skin) were recorded after the heat and cold stimuli. Temperature changes in the region of interest (ROI) are plotted as a thermal response graph. After fitting an exponential equation to each thermal response graph, the rate constants were compared between groups (pigmented BCC versus control, SK versus control).

RESULTS

The thermal response graphs revealed that the average temperature of pigmented BCC showed faster thermal recovery to baseline than the control site. There was a significant difference in the rate constants of the fitted exponential equations between the pigmented BCCs and the control sites (p<.001). However, we did not find a significantly different thermal recovery pattern between SK lesions and control sites.

CONCLUSIONS

DTI can be used as a diagnostic tool for distinguishing pigmented BCC from SK by comparing thermal recovery patterns between target lesions (pigmented BCC or SK) and the control site.

Comparative Analysis of Diagnostic Techniques for Melanoma Detection: A Systematic Review of Diagnostic Test Accuracy Studies and Meta-Analysis

Melanoma has the highest mortality rate among skin cancers, and early-diagnosis is essential to maximize survival rate. The current procedure for melanoma diagnosis is based on dermoscopy, i.e., a qualitative visual inspection of lesions with intrinsic limited diagnostic reliability and reproducibility. Other non-invasive diagnostic techniques may represent valuable solutions to retrieve additional objective information of a lesion. This review aims to compare the diagnostic performance of non-invasive techniques, alternative to dermoscopy, for melanoma detection in clinical settings. A systematic review of the available literature was performed using PubMed, Scopus and Google scholar databases (2010-September 2020). All human, in-vivo, non-invasive studies using techniques, alternative to dermoscopy, for melanoma diagnosis were included with no restriction on the recruited population. The reference standard was histology but dermoscopy was accepted only in case of benign lesions. Attributes of the analyzed studies were compared, and the quality was evaluated using CASP Checklist. For studies in which the investigated technique was implemented as a diagnostic tool (DTA studies), the QUADAS-2 tool was applied. For DTA studies that implemented a melanoma vs. other skin lesions classification task, a meta-analysis was performed reporting the SROC curves. Sixty-two references were included in the review, of which thirty-eight were analyzed using QUADAS-2. Study designs were: clinical trials (13), retrospective studies (10), prospective studies (8), pilot studies (10), multitiered study (1); the remain studies were proof of concept or had undefined study type. Studies were divided in categories based on the physical principle employed by each diagnostic technique. Twenty-nine out of thirty-eight DTA studies were included in the meta-analysis. Heterogeneity of studies' types, testing strategy, and diagnostic task limited the systematic comparison of the techniques. Based on the SROC curves, spectroscopy achieved the best performance in terms of sensitivity (93%, 95% CI 92.8-93.2%) and specificity (85.2%, 95%CI 84.9-85.5%), even though there was high concern regarding robustness of metrics. Reflectance-confocal-microscopy, instead, demonstrated higher robustness and a good diagnostic performance (sensitivity 88.2%, 80.3-93.1%; specificity 65.2%, 55-74.2%). Best practice recommendations were proposed to reduce bias in future DTA studies. Particular attention should be dedicated to widen the use of alternative techniques to conventional dermoscopy.

Using thermal imaging to measure changes in breast cancer-related lymphoedema during reflexology

Reflexology lymph drainage (RLD) for breast cancer-related lymphoedema (BCRL) may have a positive impact on arm swelling and pain. Thermal imaging is a means of tracking temperature change by visual images. This study aimed to explore the use of thermal imaging in treatment for BCRL. The swollen arms of two participants with BCRL were photographed using a thermal imaging camera during a single RLD treatment. Limb Volume Circumferential Measurement (LVCM) of both arms was taken before, after and the next day. The images were examined for visual changes, and temperature data were extracted. Images showed differences in temperature within the affected hand and arm over 45 minutes. LVCM data indicated a loss of limb volume in the affected arm in both cases, which continued to decrease over 24 hours. Thus, thermal imaging may be useful in tracking temperature change during treatment for BCRL.

Focal dynamic thermal imaging for label-free high-resolution characterization of materials and tissue heterogeneity

Evolution from static to dynamic label-free thermal imaging has improved bulk tissue characterization, but fails to capture subtle thermal properties in heterogeneous systems. Here, we report a label-free, high speed, and high-resolution platform technology, focal dynamic thermal imaging (FDTI), for delineating material patterns and tissue heterogeneity. Stimulation of focal regions of thermally responsive systems with a narrow beam, low power, and low cost 405 nm laser perturbs the thermal equilibrium. Capturing the dynamic response of 3D printed phantoms, ex vivo biological tissue, and in vivo mouse and rat models of cancer with a thermal camera reveals material heterogeneity and delineates diseased from healthy tissue. The intuitive and non-contact FDTI method allows for rapid interrogation of suspicious lesions and longitudinal changes in tissue heterogeneity with high-resolution and large field of view. Portable FDTI holds promise as a clinical tool for capturing subtle differences in heterogeneity between malignant, benign, and inflamed tissue.

A Method to Determine Human Skin Heat Capacity Using a Non-Invasive Calorimetric Sensor

A calorimetric sensor has been designed to measure the heat flow dissipated by a 2 x 2 cm² skin surface. In this work, a non-invasive method is proposed to determine the heat capacity and thermal conductance of the area of skin where the measurement is made. The method consists of programming a linear variation of the temperature of the sensor thermostat during its application to the skin. The sensor is modelled as a two-inputs and two-outputs system. The inputs are 1) the power dissipated by the skin and transmitted by conduction to the sensor, and 2) the power dissipated in the sensor thermostat to maintain the programmed temperature. The outputs are 1) the calorimetric signal and 2) the thermostat temperature. The proposed method consists of a sensor modelling that allows the heat capacity of the element where dissipation takes place (the skin) to be identified, and the transfer functions (TF) that link the inputs and outputs are constructed from its value. These TFs allow the determination of the heat flow dissipated by the surface of the human body as a function of the temperature of the sensor thermostat. Furthermore, as this variation in heat flow is linear, we define and determine an equivalent thermal resistance of the skin in the measured area. The method is validated with a simulation and with experimental measurements on the surface of the human body.

Numerical modelling of skin tumour tissue with temperature-dependent properties for dynamic thermography

Dynamic thermography has been clinically proven to be a valuable diagnostic technique for skin tumour detection as well as for other medical applications, and shows many advantages over static thermography. Numerical modelling of heat transfer phenomena in biological tissue during dynamic thermography can aid the technique by improving process parameters or by estimating unknown tissue parameters based on measurement data. This paper presents a new non-linear numerical model of multilayer skin tissue containing a skin tumour together with thermoregulation response of the tissue during the cooling-rewarming process of dynamic thermography. The thermoregulation response is modelled by temperature-dependent blood perfusion rate and metabolic heat generation. The aim is to describe bioheat transfer more realistically. The model is based on the Pennes bioheat equation and solved numerically using a subdomain BEM approach treating the problem as axisymmetrical. The paper includes computational tests for Clark II and Clark IV tumours, comparing the models using constant and temperature-dependent properties which showed noticeable differences and highlighted the importance of using a local thermoregulation model. Results also show the advantage of using dynamic thermography for skin tumour screening and detection at an early stage. One of the contributions of this paper is a complete sensitivity analysis of 56 model parameters based on the gradient of the surface temperature difference between tumour and healthy skin. The analysis shows that size of the tumour, blood perfusion rate, thermoregulation coefficient of the tumour, body core temperature and density and specific heat of the skin layers in which the tumour is embedded are important for modelling the problem, and so have to be determined more accurately to reflect realistic skin response of the investigated tissue, while metabolic heat generation and its thermoregulation are not.

On-the-fly extrinsic calibration of multimodal sensing system for fast 3D thermographic scanning

The fusion of three-dimensional (3D) geometrical and two-dimensional (2D) thermal information provides a promising method for characterizing temperature distribution of 3D objects, extending infrared imaging from 2D to 3D to support various thermal inspection applications. In this paper, we present an effective on-the-fly calibration approach for accurate alignment of depth and thermal data to facilitate dynamic and fast-speed 3D thermal scanning tasks. For each pair of depth and thermal frames, we estimate their relative pose by minimizing the objective function that measures the temperature consistency between a 2D infrared image and the reference 3D thermographic model. Our proposed frame-to-model mapping scheme can be seamlessly integrated into a generic 3D thermographic reconstruction framework. Through graphics-processing-unit-based acceleration, our method requires less than 10 ms to generate a pair of well-aligned depth and thermal images without hardware synchronization and improves the robustness of the system against significant camera motion.

Measurement of human body surface heat flux using a calorimetric sensor

We have developed a calorimetric sensor that can perform local measurements of the heat flux transmitted by conduction between a human body and thermostat located inside the sensor. The sensor has a detection area of 2 × 2 cm² and, in its current configuration, facilitates measurement with a resolution of 10 mW. In this paper, measurements of two healthy male subjects of different ages (24 and 60 years) are presented. We study the variation in the power dissipated by the human body surface as a function of time for a thermostat temperature of 28 °C. We also study the same power with thermostat temperatures varying from 24° to 36°C. Measurements are performed on three different surface areas of the human body: the sternum, abdomen, and hand. The ambient room temperature during all measurements was in the range of 22-24 °C, and the subjects were seated and resting. The results show that the heat flux in the trunk is much more stable than that in the hand and that the heat flux in the sternum is greater than that in other areas. Additionally, this flux is higher in the younger subject (42 mW/cm²) than in the older subject (35 mW/cm²). We also defined a thermal parameter that represents the thermal resistance between the sensor thermostat and the skin. The mean value of this parameter varies between 51 and 71 K/W depending on the subject and measurement area.

Normative surface skin temperature changes due to blood redistribution: A prospective study

The continuing development and manufacture of infrared devices, together with improvements in thermal body mapping techniques have simplified surface skin thermography which is being used more extensively than ever before. Normative thermography data, however, remains incomplete. A normative blood redistribution range of skin temperatures was established for use as a reference for laboratory infrared thermography (IT), thermal body mapping, and mass fever screenings. 500 healthy volunteers participated in this prospective study. To determine the maximum range of the skin temperature changes due to the posture-related physiological blood redistribution, the volunteers were asked to keep one extremity up and another extremity down whilst lying, sitting, and standing. We obtained 6000 hand and 400 foot temperature readings. The normal temperature was 29.1 ± 0.6 °C for the middle fingers and 27.8 ± 0.7 °C for the toes. The physiological temperature change during body position changes ranged from 4 to 6 °C (fingers: 27-31 °C; toes: 26-32 °C). At normal room temperature, the surface skin temperature may vary within this range due to blood redistribution. These changes reflect the individual variability of vasomotor activity. This physiological range of temperatures should be taken into account during IT and other thermography-involved investigations.

An embedded system for image segmentation and multimodal registration in noninvasive skin cancer screening

We present a heterogeneous architecture for image registration and multimodal segmentation on an embedded system for noninvasive skin cancer screening. The architecture combines Otsu thresholding and the random walker algorithm to perform image segmentation, and features a hardware implementation of the Harris corner detection algorithm to perform region-of-interest detection and image registration. Running on a Xilinx XC7Z020 reconfigurable system-on-a-chip, our prototype computes the initial segmentation of a 400×400-pixel region of interest in the visible spectrum in 12.1 seconds, and registers infrared images against this region at 540 frames per second, while consuming 1.9W.