Computer Aided Diagnosis Based Hand Thermal Image Analysis: A Potential Tool for the Evaluation of Rheumatoid Arthritis

Medical imaging in rheumatoid arthritis: A review on deep learning approach

Arthritis is a musculoskeletal disorder. Millions of people have arthritis, making it one of the most common joint disorders. Osteoarthritis (OA) and rheumatoid arthritis (RA) are the most common types of arthritis among the many different types available. Pain, stiffness, and inflammation are among the early signs of arthritis, which can progress to severe immobility at a later stage if left untreated. Although arthritis cannot be cured at any point in time, it can be managed if diagnosed and treated correctly. Clinical diagnostic and medical imaging methods are currently used to evaluate OA and RA, both debilitating conditions. This review is focused on deep learning approaches used by taking medical imaging (X-rays and magnetic resonance imaging) as input for the detection of RA.

Application of machine learning algorithms in thermal images for an automatic classification of lumbar sympathetic blocks

PURPOSE

There are no previous studies developing machine learning algorithms in the classification of lumbar sympathetic blocks (LSBs) performance using infrared thermography data. The objective was to assess the performance of different machine learning algorithms to classify LSBs carried out in patients diagnosed with lower limbs Complex Regional Pain Syndrome as successful or failed based on the evaluation of thermal predictors.

METHODS

66 LSBs previously performed and classified by the medical team were evaluated in 24 patients. 11 regions of interest on each plantar foot were selected within the thermal images acquired in the clinical setting. From every region of interest, different thermal predictors were extracted and analysed in three different moments (minutes 4, 5, and 6) along with the baseline time (just after the injection of a local anaesthetic around the sympathetic ganglia). Among them, the thermal variation of the ipsilateral foot and the thermal asymmetry variation between feet at each minute assessed and the starting time for each region of interest, were fed into 4 different machine learning classifiers: an Artificial Neuronal Network, K-Nearest Neighbours, Random Forest, and a Support Vector Machine.

RESULTS

All classifiers presented an accuracy and specificity higher than 70%, sensitivity higher than 67%, and AUC higher than 0.73, and the Artificial Neuronal Network classifier performed the best with a maximum accuracy of 88%, sensitivity of 100%, specificity of 84% and AUC of 0.92, using 3 predictors.

CONCLUSION

These results suggest thermal data retrieved from plantar feet combined with a machine learning-based methodology can be an effective tool to automatically classify LSBs performance.

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.

An update on thermal imaging in rheumatoid arthritis

This review aims to summarise the recent literature concerning the usage of thermal imaging in the study of rheumatoid arthritis (RA). Most RA studies have applied thermal imaging as a static process alone although thermal imaging has been conducted with an additional dynamic/functional component. Algorithms to automate the analysis of thermal imaging in RA have also been described. Several RA thermal imaging studies have demonstrated differences in thermographic findings between RA patients and healthy controls and/or compared thermographic parameters with other clinical/functional/imaging parameters; while fewer studies have assessed the role of thermal imaging in discriminating disease severity in RA. Thermal imaging is a relatively low cost, non-invasive imaging technique offering an objective measurement of joint surface temperature in RA joint inflammation assessment. Although there has been an increasing literature build up on the use of thermography in RA, more validation work is still necessary to delineate the potential role(s) of its use among patients with RA. This timely review focusses on the recent literature concerning thermal imaging, and provides clinicians with an update on its recent development in RA.

Towards Exercise Radiomics: Deep Neural Network-Based Automatic Analysis of Thermal Images Captured During Exercise

Infrared thermography is increasingly applied in sports science due to promising observations regarding changes in skin's surface radiation temperature ( T_sr) before, during, and after exercise. The common manual thermogram analysis limits an objective and reproducible measurement of T_sr. Previous analysis approaches depend highly on expert knowledge and have not been applied during movement. We aimed to develop a deep neural network (DNN) capable of automatically and objectively segmenting body parts, recognizing blood vessel-associated T_sr distributions, and continuously measuring T_sr during exercise. We conducted 38 cardiopulmonary exercise tests on a treadmill. We developed two DNNs: body part network and vessel network, to perform semantic segmentation of 1 107 855 captured thermal images. Both DNNs were trained with 263 training and 75 validation images. Additionally, we compare the results of a common manual thermogram analysis with these of the DNNs. Performance analysis identified a mean IoU of 0.8 for body part network and 0.6 for vessel network. There is a high agreement between manual and automatic analysis (r = 0.999; p 0.001; T-test: p = 0.116), with a mean difference of 0.01 ^C (0.08). Non-parametric Bland Altman's analysis showed that the 95% agreement ranges between -0.086 ^C and 0.228 ^C. The developed DNNs enable automatic, objective, and continuous measurement of T_sr and recognition of blood vessel-associated T_sr distributions in resting and moving legs. Hence, the DNNs surpass previous algorithms by eliminating manual region of interest selection and form the currently needed foundation to extensively investigate T_sr distributions related to non-invasive diagnostics of (patho-)physiological traits in means of exercise radiomics.

Advances in Thermal Image Analysis for the Detection of Pregnancy in Horses Using Infrared Thermography

Infrared thermography (IRT) was applied as a potentially useful tool in the detection of pregnancy in equids, especially native or wildlife. IRT measures heat emission from the body surface, which increases with the progression of pregnancy as blood flow and metabolic activity in the uterine and fetal tissues increase. Conventional IRT imaging is promising; however, with specific limitations considered, this study aimed to develop novel digital processing methods for thermal images of pregnant mares to detect pregnancy earlier with higher accuracy. In the current study, 40 mares were divided into non-pregnant and pregnant groups and imaged using IRT. Thermal images were transformed into four color models (RGB, YUV, YIQ, HSB) and 10 color components were separated. From each color component, features of image texture were obtained using Histogram Statistics and Grey-Level Run-Length Matrix algorithms. The most informative color/feature combinations were selected for further investigation, and the accuracy of pregnancy detection was calculated. The image texture features in the RGB and YIQ color models reflecting increased heterogeneity of image texture seem to be applicable as potential indicators of pregnancy. Their application in IRT-based pregnancy detection in mares allows for earlier recognition of pregnant mares with higher accuracy than the conventional IRT imaging technique.

Infrared Thermography for the Evaluation of Inflammatory and Degenerative Joint Diseases: A Systematic Review

OBJECTIVE

Inflammation plays a central role in the pathophysiology of rheumatic diseases as well as in osteoarthritis. Temperature, which can be quantified using infrared thermography, provides information about the inflammatory component of joint diseases. This systematic review aims at assessing infrared thermography potential and limitations in these pathologies.

DESIGN

A systematic review was performed on 3 major databases: PubMed, Cochrane library, and Web of Science, on clinical reports of any level of evidence in English language, published from 1990 to May 2021, with infrared thermography used for diagnosis of osteoarthritis and rheumatic diseases, monitoring disease progression, or response to treatment. Relevant data were extracted, collected in a database, and analyzed for the purpose of this systematic review.

RESULTS

Of 718 screened articles 32 were found to be eligible for inclusion, for a total of 2094 patients. Nine studies reported the application to osteoarthritis, 21 to rheumatic diseases, 2 on both. The publication trend showed an increasing interest in the last decade. Seven studies investigated the correlation of temperature changes with osteoarthritis, 16 with rheumatic diseases, and 2 with both, whereas 2 focused on the pre-post evaluation to investigate treatment results in patients with osteoarthritis and 5 in patients with rheumatic diseases. A correlation was shown between thermal findings and disease presence and stage, as well as the clinical assessment of disease activity and response to treatment, supporting infrared thermography role in the study and management of rheumatic diseases and osteoarthritis.

CONCLUSIONS

The systematic literature review showed an increasing interest in this technology, with several applications in different joints affected by inflammatory and degenerative pathologies. Infrared thermography proved to be a simple, accurate, noninvasive, and radiation-free method, which could be used in addition to the currently available tools for screening, diagnosis, monitoring of disease progression, and response to medical treatment.

A computational method to differentiate rheumatoid arthritis patients using thermography data

BACKGROUND

The traditional rheumatoid arthritis (RA) diagnosis is very complicated because it uses many clinical and image data. Therefore, there is a need to develop a new method for diagnosing RA using a consolidated set of blood analysis and thermography data.

OBJECTIVE

The following issues related to RA are discussed: 1) Which clinical data are significant in the primary diagnosis of RA? 2) What parameters from thermograms should be used to differentiate patients with RA from the healthy? 3) Can artificial neural networks (ANN) differentiate patients with RA from the healthy?

METHODS

The dataset was composed of clinical and thermal data from 65 randomly selected patients with RA and 104 healthy subjects. Firstly, the univariate logistic regression model was proposed in order to find significant predictors. Next, the feedforward neural network model was used. The dataset was divided into the training set (75% of data) and the test set (25% of data). The Broyden-Fletcher-Goldfarb-Shanno (BFGS) and non-linear logistic function to transformation nodes in the output layer were used for training. Finally, the 10 fold Cross-Validation was used to assess the predictive performance of the ANN model and to judge how it performs.

RESULT

The training set consisted of the temperature of all fingers, patient age, BMI, erythrocyte sedimentation rate, C-reactive protein and White Blood Cells (10 parameters in total). High level of sensitivity and specificity was obtained at 81.25% and 100%, respectively. The accuracy was 92.86%.

CONCLUSIONS

This methodology suggests that the thermography data can be considered in addition to the currently available tools for screening, diagnosis, monitoring of disease progression.

Characteristic Features of Infrared Thermographic Imaging in Primary Raynaud's Phenomenon

Raynaud’s phenomenon (RP) is characterized by the episodic whitening of the fingers upon exposure to cold. Verification of the condition is crucial in vibration-exposed patients. The current verification method is outdated, but thermographic imaging seems promising as a diagnostic replacement. By investigating patients diagnosed with RP, the study aimed at developing a simple thermographic procedure that could be applied to future patients where verification of the diagnosis is needed. Twenty-two patients with primary RP and 58 healthy controls were examined using thermographic imaging after local cooling of the hands for 1 min in water of 10°C. A logistic regression model was fitted with the temperature curve characteristics to convey a predicted probability of having RP. The characteristics time to end temperature and baseline temperature were the most appropriate predictors of RP among those examined (p = 0.004 and p = 0.04, respectively). The area under the curve was 0.91. The cut-off level 0.46 yielded a sensitivity and specificity of 82% and 86%, respectively. The positive and negative predictive values were 69% and 93%, respectively. This newly developed thermographic method was able to distinguish between patients with RP and healthy controls and was easy to operate. Thus, the method showed great promise as a method for verification of RP in future patients. Trial registration: ClinicalTrials.gov NCT03094910.

Meta-Analysis and Systematic Review of the Application of Machine Learning Classifiers in Biomedical Applications of Infrared Thermography

Atypical body temperature values can be an indication of abnormal physiological processes associated with several health conditions. Infrared thermal (IRT) imaging is an innocuous imaging modality capable of capturing the natural thermal radiation emitted by the skin surface, which is connected to physiology-related pathological states. The implementation of artificial intelligence (AI) methods for interpretation of thermal data can be an interesting solution to supply a second opinion to physicians in a diagnostic/therapeutic assessment scenario. The aim of this work was to perform a systematic review and meta-analysis concerning different biomedical thermal applications in conjunction with machine learning strategies. The bibliographic search yielded 68 records for a qualitative synthesis and 34 for quantitative analysis. The results show potential for the implementation of IRT imaging with AI, but more work is needed to retrieve significant features and improve classification metrics.

Thermography and colour Doppler ultrasound: a potential complementary diagnostic tool in evaluation of rheumatoid arthritis in the knee region

The aim and objectives of this study were as follows: (i) to perform automated segmentation of knee thermal image using the regional isotherm-based segmentation (RIBS) algorithm and segmentation of ultrasound image using the image J software; (ii) to implement the RIBS algorithm using computer-aided diagnostic (CAD) tools for classification of rheumatoid arthritis (RA) patients and normal subjects based on feature extraction values; and (iii) to correlate the extracted thermal imaging features and colour Doppler ultrasound (CDUS) features in the knee region with the biochemical parameters in RA patients. Thermal image analysis based on skin temperature measurement and thermal image segmentation was performed using the RIBS algorithm in the knee region of RA patients and controls. There was an increase in the average skin temperature of 5.94% observed in RA patients compared to normal. CDUS parameters such as perfusion, effusion and colour fraction for the RA patients were found to be 1.2 ± 0.5, 1.8 ± 0.2 and 0.052 ± 0.002, respectively. CDUS measurements were performed and analysed using the image J software. Biochemical parameters such as erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) showed significant positive correlation with the thermal imaging parameters. The CDUS parameters such as effusion, perfusion and colour fraction correlated significantly with the clinical and functional assessment score. According to the results of this study, both infrared (IR) thermal imaging and CDUS offer better diagnostic potential in detecting early-stage RA. Therefore, the developed CAD model using thermal imaging could be used as a pre-screening tool to diagnose RA in the knee region.

Detection of inflammation from finger temperature profile in rheumatoid arthritis

Rheumatoid arthritis (RA) is a chronic inflammatory tissue disease that leads to cartilage, bone, and periarticular tissue damage. This study aimed to investigate whether the use of infrared thermography and measurement of temperature profiles along the hand fingers could detect the inflammation and improve the diagnostic accuracy of the cold provocation test (0 °C for 5 s) and rewarming test (23 °C for180 s) in RA patients. Thirty RA patients (mean age = 49.5 years, standard deviation = 13.0 years) and 22 controls (mean age = 49.8 years, standard deviation = 7.5 years) were studied. Outcomes were the minimal and maximal: baseline temperature (T1), the temperature post-cooling (T2), the temperature post-rewarming (T3), and the Tmax-Tmin along the axis of each finger. The statistical significance was observed for the thumb, index finger, middle finger, and ring finger post-cooling and post-rewarming. Receiver operating characteristics (ROC) analysis to distinguish between the two groups revealed that for the thumb, index finger, middle finger, and ring finger, the area under the ROC curve was statistically significantly (p < 0.05) post-cooling. The cold provocation test used in this study discriminates between RA patients and controls and detects an inflammation in RA patients by the measurement of temperature profiles along the fingers using an infrared camera. Graphical abstract.