Medical infrared thermal imaging of canine appendicular bone neoplasia

Convolutional Neural Network to Classify Infrared Thermal Images of Fractured Wrists in Pediatrics

Convolutional neural network (CNN) models were devised and evaluated to classify infrared thermal (IRT) images of pediatric wrist fractures. The images were recorded from 19 participants with a wrist fracture and 21 without a fracture (sprain). The injury diagnosis was by X-ray radiography. For each participant, 299 IRT images of their wrists were recorded. These generated 11,960 images (40 participants × 299 images). For each image, the wrist region of interest (ROI) was selected and fast Fourier transformed (FFT) to obtain a magnitude frequency spectrum. The spectrum was resized to 100 × 100 pixels from its center as this region represented the main frequency components. Image augmentations of rotation, translation and shearing were applied to the 11,960 magnitude frequency spectra to assist with the CNN generalization during training. The CNN had 34 layers associated with convolution, batch normalization, rectified linear unit, maximum pooling and SoftMax and classification. The ratio of images for the training and test was 70:30, respectively. The effects of augmentation and dropout on CNN performance were explored. Wrist fracture identification sensitivity and accuracy of 88% and 76%, respectively, were achieved. The CNN model was able to identify wrist fractures; however, a larger sample size would improve accuracy.

Training and validation of a novel non-invasive imaging system for ruling out malignancy in canine subcutaneous and cutaneous masses using machine learning in 664 masses

Objective

To train and validate the use of a novel artificial intelligence-based thermal imaging system as a screening tool to rule out malignancy in cutaneous and subcutaneous masses in dogs.

Animals

Training study: 147 client-owned dogs with 233 masses. Validation Study: 299 client-owned dogs with 525 masses. Cytology was non-diagnostic in 94 masses, resulting in 431 masses from 248 dogs with diagnostic samples.

Procedures

The prospective studies were conducted between June 2020 and July 2022. During the scan, each mass and its adjacent healthy tissue was heated by a high-power Light-Emitting Diode. The tissue temperature was recorded by the device and consequently analyzed using a supervised machine learning algorithm to determine whether the mass required further investigation. The first study was performed to collect data to train the algorithm. The second study validated the algorithm, as the real-time device predictions were compared to the cytology and/or biopsy results.

Results

The results for the validation study were that the device correctly classified 45 out of 53 malignant masses and 253 out of 378 benign masses (sensitivity = 85% and specificity = 67%). The negative predictive value of the system (i.e., percent of benign masses identified as benign) was 97%.

Clinical relevance

The results demonstrate that this novel system could be used as a decision-support tool at the point of care, enabling clinicians to differentiate between benign lesions and those requiring further diagnostics.

Detection of canine obstructive nasal disease using infrared thermography: A pilot study

Infrared thermography detects variations in heat signature and is utilized in other species to non-invasively identify respiratory disease. This study aimed to determine if infrared thermography could be used to detect nasal disease in dogs. Eight dogs presenting for nasal disease (ND group) and ten healthy control dogs (C group) were enrolled. Dorsal and rostral images of the nose were acquired using a Fluke TiX580 60Hz thermal imaging camera. Images were analyzed using the accompanying software. Regions of interest were defined over the right and left nasal passages to determine the maximum (max), average (avg), and minimum (min) temperatures. Temperatures were compared between ND and C groups, and correlation to disease state (ND or C) was evaluated. Temperature differences and imaging patterns were subjectively compared with diagnosis based on computed tomography (CT) and histopathology. The ND group consisted of 5 spayed females and 3 neutered males. Clinical sings included unilateral epistaxis (n = 4); bilateral serous discharge and sneezing (n = 1); bilateral mucopurulent discharge, epistaxis, and sneezing (n = 1); unilateral mucoid discharge, epistaxis and sneezing (n = 1); and sneezing and unilateral epistaxis (n = 1). Temperatures were significantly different between ND and C groups on dorsal (max p = <0.001, avg p = 0.001, min p = <0.001) and rostral (max p = <0.001, avg p = <0.001, min p = 0.005) images. Temperature positively correlated to disease status (ND vs C group) in both dorsal and rostral images. Subjective analysis of images allowed correct identification of abnormal or normal 27/36 times. Obstructive nasal disease results in a local temperature increase in the affected nasal passage that can be non-invasively detected by infrared thermography.

Evaluation of the thermographic response of the lumbar region in dogs with bilateral hip osteoarthritis

Infrared thermography is suggested as a method of medical assistance for evaluating anatomical regions where there may be some inflammatory or painful condition that requires immediate medical attention. For this reason, this study aimed to characterize digital thermography of the lumbar region in police working dogs with hip osteoarthritis, 47 dogs with bilateral hip osteoarthritis were evaluated. A thermographic dorsoventral view of the dogs' backs were obtained, and mean and maximum temperatures were determined. In addition, results from the Canine Brief Pain Inventory, Liverpool Osteoarthritis in Dogs (LOAD), the Canine Orthopedic Index, and weight-bearing distribution evaluation were collected. Results were compared with an Independent Samples T-Test or ANOVA, followed by an LSD post hoc. Multiple regression was run to predict temperatures from age, sex, breed, body weight, and Orthopedic Foundation for Animals hip score and to predict clinical metrology scores and weight-bearing distribution results from temperature results. The sample included 28 males and 19 females, with a mean age of 6.3 ± 2.5 years and bodyweight of 26.9 ± 5.1 kg, with osteoarthritis hip scores of mild (70.2%), moderate (21.3%), and severe (8.5%). Significant differences were observed in mean temperatures between German Shepherd Dogs (GSD) and Labrador Retrievers (LR) (p < 0.01), and LR and Belgian Malinois (BM) (p=0.02). Mean temperature contributed to the prediction of LOAD and stiffness. LR showed significantly higher temperature values than GSD and BM. No differences were observed between osteoarthritis hip scores and mean temperature values contributed to LOAD scores' prediction. This relation needs to be addressed in further studies, involving a larger number of animals, and to determine changes in response to treatment.

Performance evaluation of E-nose and E-tongue combined with machine learning for qualitative and quantitative assessment of bear bile powder

Bear bile powder (BBP) is a valuable animal-derived product with a huge adulteration problem on market. It is a crucially important task to identify BBP and its counterfeit. Electronic sensory technologies are the inheritance and development of traditional empirical identification. Considering that each drug has its own specific odor and taste characteristics, electronic tongue (E-tongue), electronic nose (E-nose) and GC-MS were used to evaluate the aroma and taste of BBP and its common counterfeit. Two active components of BBP, namely tauroursodeoxycholic acid (TUDCA) and taurochenodeoxycholic acid (TCDCA) were measured and linked with the electronic sensory data. The results showed that bitterness was the main flavor of TUDCA in BBP, saltiness and umami were the main flavor of TCDCA. The volatiles detected by E-nose and GC-MS were mainly aldehydes, ketones, alcohols, hydrocarbons, carboxylic acids, heterocyclic, lipids, and amines, mainly earthy, musty, coffee, bitter almond, burnt, pungent odor descriptions. Four different machine learning algorithms (backpropagation neural network, support vector machine, K-nearest neighbor, and random forest) were used to identify BBP and its counterfeit, and the regression performance of these four algorithms was also evaluated. For qualitative identification, the algorithm of random forest has shown the best performance, with 100% accuracy, precision, recall and F1-score. Also, the random forest algorithm has the best R² and the lowest RMSE in terms of quantitative prediction.

Applicability of thermography for cancer diagnosis in small animals

Medical thermography is an imaging test used to monitor skin surface temperature. Although it is not a recent technique, significant advances have been made since the 2000s with the equipment modernization, leading to its popularization. In cancer diagnosis, the application of thermography is supported by the difference in thermal distribution between neoplastic processes and adjacent healthy tissue. The mechanisms involved in heat production by cancer cells include neoangiogenesis, increased metabolic rate, vasodilation, and the release of nitric oxide and pro-inflammatory substances. Currently, thermography has been widely studied in humans as a screening tool for skin and breast cancer, with positive results. In veterinary medicine, the technique has shown promise and has been described for skin and soft tissue tumors in felines, mammary gland tumors, osteosarcoma, mast cell tumors, and perianal tumors in dogs. This review discusses the fundamentals of the technique, monitoring conditions, and the role of thermography as a complementary diagnostic tool for cancer in veterinary medicine, as well as future perspectives for improvement.

A pilot study for a non-invasive system for detection of malignancy in canine subcutaneous and cutaneous masses using machine learning

Introduction

Early diagnosis of cancer enhances treatment planning and improves prognosis. Many masses presenting to veterinary clinics are difficult to diagnose without using invasive, time-consuming, and costly tests. Our objective was to perform a preliminary proof-of-concept for the HT Vista device, a novel artificial intelligence-based thermal imaging system, developed and designed to differentiate benign from malignant, cutaneous and subcutaneous masses in dogs.

Methods

Forty-five dogs with a total of 69 masses were recruited. Each mass was clipped and heated by the HT Vista device. The heat emitted by the mass and its adjacent healthy tissue was automatically recorded using a built-in thermal camera. The thermal data from both areas were subsequently analyzed using an Artificial Intelligence algorithm. Cytology and/or biopsy results were later compared to the results obtained from the HT Vista system and used to train the algorithm. Validation was done using a "Leave One Out" cross-validation to determine the algorithm's performance.

Results

The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the system were 90%, 93%, 88%, 83%, and 95%, respectively for all masses.

Conclusion

We propose that this novel system, with further development, could be used to provide a decision-support tool enabling clinicians to differentiate between benign lesions and those requiring additional diagnostics. Our study also provides a proof-of-concept for ongoing prospective trials for cancer diagnosis using advanced thermodynamics and machine learning procedures in companion dogs.

Thermal and Circulatory Changes in Diverse Body Regions in Dogs and Cats Evaluated by Infrared Thermography

Infrared thermography (IRT) has been proposed as a method for clinical research to detect local inflammatory processes, wounds, neoplasms, pain, and neuropathies. However, evidence of the effectiveness of the thermal windows used in dogs and cats is discrepant. This review aims to analyze and discuss the usefulness of IRT in diverse body regions in household animals (pets) related to recent scientific evidence on the use of the facial, body, and appendicular thermal windows. IRT is a diagnostic method that evaluates thermal and circulatory changes under different clinical conditions. For the face, structures such as the lacrimal caruncle, ocular area, and pinna are sensitive to assessments of stress degrees, but only the ocular window has been validated in felines. The usefulness of body and appendicular thermal windows has not been conclusively demonstrated because evidence indicates that biological and environmental factors may strongly influence thermal responses in those body regions. The above has led to proposals to evaluate specific muscles that receive high circulation, such as the bicepsfemoris and gracilis. The neck area, perivulvar, and perianal regions may also prove to be useful thermal windows, but their degree of statistical reliability must be established. In conclusion, IRT is a non-invasive technique that can be used to diagnose inflammatory and neoplastic conditions early. However, additional research is required to establish the sensitivity and specificity of these thermal windows and validate their clinical use in dogs and cats.

Evaluation of digital thermography imaging to assess and monitor treatment of police working dogs with naturally occurring hip osteoarthritis

BACKGROUND

In dogs, thermal imaging has been documented only recently, but a growing interest in this modality has led to studies using thermography to assess pathologies in the canine hip, stifle, elbow, intervertebral disc, and bone neoplasia. This study aimed to evaluate the use of digital thermography in assessing and evaluating treatment response in dogs with hip osteoarthritis (OA) and comparing its results with an objective measure and two clinical metrology instruments. In an experimental, randomized, double-blinded study, one hundred hip joints of fifty police working dogs with bilateral hip OA were evaluated. A dorsoventral and lateral thermographic image were obtained on days 0, 8, 15, 30, 90, and 180. Mean and maximal temperatures were determined. Additionally, the animal's weight-bearing distribution and radiographic examination of the hip joint (extended legs ventrodorsal view) were performed. Copies of the Canine Brief Pain Inventory (CBPI) and Canine Orthopaedic Index (COI) were obtained. Results were analyzed by ANOVA, followed by an LSD post-hoc test, and correlations were assessed with Spearman correlation coefficient, with p < 0.05.

RESULTS

Values recorded on the lateral view were higher than those on the dorsoventral view. No differences or correlations were found between Orthopedic Foundation for Animals hip grades and temperature. Digital thermographic images showed a weak significant correlation with weight-bearing evaluations (r = 0.13, p < 0.01) and different clinical metrology instruments scores (r = - 0.25, p < 0.01 for pain severity score, and r = - 0.21, p = 0.04 for gait). It also correlated with radiographic findings, specifically the circumferential femoral head osteophyte and caudolateral curvilinear osteophyte.

CONCLUSION

To our knowledge, this is the first study presenting the digital thermography assessment of Police working dogs submitted to treatment for hip OA. Digital thermography, mainly based on a lateral view evaluation, showed a weak significant correlation with stance analysis and clinical metrology instruments scores.

Detection of microRNA expression levels based on microarray analysis for classification of idiopathic pulmonary fibrosis

The etiology and pathophysiological mechanisms of idiopathic pulmonary fibrosis (IPF) are yet to be fully elucidated; however, mining of disease-related microRNAs (miRNAs/miRs) has improved the understanding of the progression of IPF. The aim of the current study was to screen miRNAs associated with IPF using three mathematical algorithms: One-way ANOVA, least absolute shrinkage and selector operation (LASSO) and support vector machine-recursive feature elimination (SVM-RFE). Using ANOVA, three miRNAs and two miRNAs were selected with opposite expression patterns in moderate and severe IPF, respectively. In total, two algorithms, LASSO and SVM-RFE, were used to perform feature selection of miRNAs. miRNAs from patients were also extracted from formalin-fixed paraffin-embedded tissues and detected using reverse transcription-quantitative PCR (RT-qPCR). The intersection of the three algorithms (ANOVA, LASSO and SVM-RFE) was taken as the final result of the miRNA candidates. Three miRNA candidates, including miR-124, hsa-miR-524-5p and hsa-miR-194 were therefore used as biomarkers. The receiver operating characteristic model demonstrated favorable discrimination between IPF and control groups, with an area under the curve of 78.5%. Moreover, RT-qPCR results indicated that miR-124, hsa-miR-524-5p, hsa-miR-194 and hsa-miR-133a were differentially expressed between patients with IPF and age-matched men without fibrotic lung disease. The target genes of these miRNAs were further predicted and Kyoto Encyclopedia of Genes and Genomes enrichment analysis was performed. Collectively, the present results suggested that the identified miRNAs associated with IPF may be useful biomarkers for the diagnosis of this disease.