Search

Reports of machine learning implementations in veterinary imaging are infrequent but changes in machine learning architecture and access to increased computing power will likely prompt increased interest. This diagnostic accuracy study describes a particular form of machine learning, a deep learning convolution neural network (ConvNet) for hip joint detection and classification of hip dysplasia from ventro-dorsal (VD) pelvis radiographs submitted for hip dysplasia screening. 11,759 pelvis images were available together with their Fédération Cynologique Internationale (FCI) scores. The dataset was dicotomized into images showing no signs of hip dysplasia (FCI grades "A" and "B", the "A-B" group) and hips showing signs of dysplasia (FCI grades "C", "D," and "E", the "C-E" group). In a transfer learning approach, an existing pretrained ConvNet was fine-tuned to provide models to recognize hip joints in VD pelvis images and to classify them according to their FCI score grouping. The results yielded two models. The first was successful in detecting hip joints in the VD pelvis images (intersection over union of 85%). The second yielded a sensitivity of 0.53, a specificity of 0.92, a positive predictive value of 0.91, and a negative predictive value of 0.81 for the classification of detected hip joints as being in the "C-E" group. ConvNets and transfer learning are applicable to veterinary imaging. The models obtained have potential to be a tool to aid in hip screening protocols if hip dysplasia classification performance was improved through access to more data and possibly by model optimization.

The liquid-based Carbo-gel technology enables real-time multimodal image guidance for improved therapeutic precision., Diagnostic imaging often outperforms the surgeon’s ability to identify small structures during therapeutic procedures. Smart soft tissue markers that translate the sensitivity of diagnostic imaging into optimal therapeutic intervention are therefore highly warranted. This paper presents a unique adaptable liquid soft tissue marker system based on functionalized carbohydrates (Carbo-gel). The liquid state of these markers allows for high-precision placement under image guidance using thin needles. Based on step-by-step modifications, the image features and mechanical properties of markers can be optimized to bridge diagnostic imaging and specific therapeutic interventions. The performance of Carbo-gel is demonstrated for markers that (i) have radiographic, magnetic resonance, and ultrasound visibility; (ii) are palpable and visible; and (iii) are localizable by near-infrared fluorescence and radio guidance. The study demonstrates encouraging proof of concept for the liquid marker system as a well-tolerated multimodal imaging marker that can improve image-guided radiotherapy and surgical interventions, including robotic surgery.

Objectives The aim of this study was to evaluate stifle joint rotational stability in response to body position and repositioning in dogs undergoing computed tomography (CT). Materials and Methods Nine dogs (18 stifles) with no history or clinical signs of stifle injury undergoing CT examinations for other reasons were included in the study. Dogs were positioned in alternating dorsal and sternal recumbency with the pelvic limbs extended caudally and unrestrained, for a total of four examinations. Scans included the entire tibia and distal femur. Using defined landmarks on the tibia and femur, stifle joint rotational angles were measured from multiplanar reconstructions made by two independent observers, and repositioning effects and intra- and interobserver agreement evaluated. Results Repositioning produced repeatable stifle joint rotational angles (pooled within-subject standard deviations of 0.9° and 1.4°) and intraobserver repeatability was good (within-subject standard deviations 1.4°), but interobserver agreement was poor. Dorsal and sternal positioning produced equivalent results. Clinical Significance The results indicate that little rotation occurs in the healthy extended canine stifle joint during positioning for CT and support CT as a method for assessing limb alignment around the stifle joint, provided that the limb distal to the femorotibial joint is unrestrained. Clinically, it is likely that measured changes in rotational angles across (and distant to) the stifle joint are real, rather than being due to rotation artefact during imaging.

Audio fingerprinting involves extraction of quantitative frequency descriptors that can be used for indexing, search and retrieval of audio signals in sound recognition software. We propose a similar approach with medical ultrasonographic Doppler audio signals. Power Doppler periodograms were generated from 84 ultrasonographic Doppler signals from the common carotid arteries in 22 dogs. Frequency features were extracted from each periodogram and included in a principal component analysis (PCA). From this 10 audio samples were pairwise classified as being either similar or dissimilar. These pairings were compared to a similar classification based on standard quantitative parameters used in medical ultrasound and to classification performed by a panel of listeners. The ranking of sound files according to degree of similarity differed between the frequency and conventional classification methods. The panel of listeners had an 88% agreement with the classification based on quantitative frequency features. These findings were significantly different from the score expected by chance (p

Objective: We aimed to assess the efficacy and benefit-risk profile of pregabalin (PGN) to reduce the clinical signs of central neuropathic pain (CNeP) as reflected by scratching episodes in dogs with symptomatic syringomyelia (SM). Study design: Randomized, double-blind, placebo-controlled crossover study. Animals: A total of 12 client-owned Cavalier King Charles Spaniels (age, 1.1–7.4 years, bodyweight, 8.2–10.8 kg) with magnetic resonance imaging-confirmed SM and clinical signs of CNeP. Methods: Dogs were randomized to either PGN 150 mg or placebo for 25 days, followed by 48 hour washout period before crossover to the alternate phase of 25 days. The primary outcome was defined as number of scratching events during 10 minutes of video-recorded physical activity. Treatment effect was estimated using a generalized estimation equation model. Benefit-risk and quality of life assessments were obtained through owner interviews focusing on potential adverse events. Results: The treatment effect estimate was an 84% (95% confidence interval = 75–89%) reduction in mean number of scratching events relative to baseline compared with placebo (p < 0.0001). Owner-assessed satisfactory quality of life was status quo and rated as ‘good’ or ‘could not be better’ in six/11 dogs and improved in four/11 dogs. The most prevalent adverse events were increased appetite in nine/12 dogs and transient ataxia in nine/12 dogs. There was one dog withdrawn by the owner 7 days after crossover to PGN owing to persistent ataxia. No dogs needed rescue analgesia during the trial. Conclusions and clinical relevance: PGN is superior to placebo in the reduction of clinical signs of SM-related CNeP in dogs. At a dose range of 13–19 mg kg–1 orally twice daily, the encountered adverse events were acceptable to all but one owner.

Fractal analysis of canine pulmonary vessels could allow quantification of their space-filling properties. Aims of this prospective, analytical, cross-sectional study were to describe methods for reconstructing three dimensional pulmonary arterial vascular trees from computed tomographic pulmonary angiogram, applying fractal analyses of these vascular trees in dogs with and without diseases that are known to predispose to thromboembolism, and testing the hypothesis that diseased dogs would have a different fractal dimension than healthy dogs. A total of 34 dogs were sampled. Based on computed tomographic pulmonary angiograms findings, dogs were divided in three groups: diseased with pulmonary thromboembolism (n = 7), diseased but without pulmonary thromboembolism (n = 21), and healthy (n = 6). An observer who was aware of group status created three-dimensional pulmonary artery vascular trees for each dog using a semiautomated segmentation technique. Vascular three-dimensional reconstructions were then evaluated using fractal analysis. Fractal dimensions were analyzed, by group, using analysis of variance and principal component analysis. Fractal dimensions were significantly different among the three groups taken together (P = 0.001), but not between the diseased dogs alone (P = 0.203). The principal component analysis showed a tendency of separation between healthy control and diseased groups, but not between groups of dogs with and without pulmonary thromboembolism. Findings indicated that computed tomographic pulmonary angiogram images can be used to reconstruct three-dimensional pulmonary arterial vascular trees in dogs and that fractal analysis of these three-dimensional vascular trees is a feasible method for quantifying the spatial relationships of pulmonary arteries. These methods could be applied in further research studies on pulmonary and vascular diseases in dogs.

Digital radiography is widely seen to be forgiving of poor exposure technique and to provide consistent high quality diagnostic images. Optimal quality images are however not universal; sub-optimal images are encountered. Evaluators on hip dysplasia schemes encounter images from multiple practices produced on equipment from multiple manufacturers. For images submitted to the Danish Kennel Club for hip dysplasia screening, a range of quality is seen and the evaluators are of the impression that variations in image quality area associated with particular equipment. This study was undertaken to test the hypothesis that there is an association between image quality in digital radiography and the manufacturer of the detector equipment, and to demonstrate the applicability of visual grading analysis (VGA) for image quality evaluation in veterinary practice. Data from 16,360 digital images submitted to the Danish Kennel Club were used to generate the hypothesis that there is an association between detector manufacturer and image quality and to create groups for VGA. Image quality in a subset of 90 images randomly chosen from 6 manufacturers to represent high and low quality images, was characterized using VGA and the results used to test for an association between image quality and system manufacturer. The range of possible scores in the VGA was −2 to +2 (higher scores are better). The range of the VGA scores for the images in the low image quality group (n = 45) was −1.73 to +0.67, (median −1.2). Images in the high image quality group (n = 44) ranged from −1.52 to +0.53, (median −0.53). This difference was statistically significant (p < 0.001). The study shows an association between VGA scores of image quality and detector manufacturer. Possible causes may be that imaging hardware and/or software are not equal in terms of quality, that the level of support sought and given differs between systems, or a combination of the two. Clinicians purchasing equipment should be mindful that image quality can differ across systems. VGA is practical for veterinarians to compare image quality between systems or within a system over time.

Objectives: To investigate dogs with acute onset of intracranial signs suspected of stroke by primary veterinary clinicians, and establish possible differential diagnoses and long-term outcome. In addition, serum C-reactive protein and plasma cytokines were investigated as potential biomarkers of disease. Materials and Methods: All cases were evaluated by neurologic examination, routine haematology and biochemistry and measurement of serum C-reactive protein, plasma cytokine concentrations (interleukin-2, -6, -8, -10, tumour necrosis factor) and low-field MRI. Results: Primary veterinarians contacted the investigators with 85 suspected stroke cases. Only 20 met the inclusion criteria. Of these, two were diagnosed with ischaemic stroke. Other causes were idiopathic vestibular syndrome (n=6), brain tumour (n=5) and inflammatory brain disease (n=2); in five cases a precise diagnosis could not be determined. Median survival times were: brain tumour, 3 days, idiopathic vestibular syndrome, 315 days, ischaemic stroke, 365 days and inflammatory central nervous system (CNS) disease, 468 days. The median plasma concentrations of interleukin-2, -6, -8, -10 or tumour necrosis factor were not significantly increased in any of the diagnosis groups compared to healthy controls. Serum C-reactive protein was higher in dogs with brain tumours and inflammatory brain disease but not above the upper bound of the reference interval. Clinical Significance: Dogs that present with acute onset intracranial disease may have ischaemic stroke but are more likely to have other causes. Many dogs with such acute onset of neurological dysfunction (brain tumours excluded) may recover within a couple of weeks despite their initial severe clinical appearance.

OBJECTIVE To examine ultrasonographic predictors of ovarian development in European eels (Anguilla anguilla) undergoing hormonal treatment for assisted reproduction. ANIMALS 83 female European eels. PROCEDURES Eels received weekly IM injections of salmon pituitary extract (first injection = week 1). Ultrasonography of the ovaries was performed twice during hormonal treatment (weeks 7 and 11). Eels were identified on the basis of body weight as having an adequate response by weeks 14 to 20 or an inadequate response after injections for 21 weeks. Eels were euthanized at the end of the experiment and classified by use of ovarian histologic examination. Ovarian cross-sectional area and size of eel (ie, length3) were used to classify eels (fast responder, slow responder, or nonresponder) and to calculate an ultrasonographic-derived gonadosomatic index. Gray-level co-occurrence matrices were calculated from ovarian images, and 22 texture features were calculated from these matrices. RESULTS The ultrasonographic-derived gonadosomatic index differed significantly between fast responders and slow responders or nonresponders at both weeks 7 and 11. Principal component analysis revealed a pattern of separation between the groups, and partial least squares discriminant analysis revealed signals in the ovarian texture that discriminated females that responded to treatment from those that did not. CONCLUSIONS AND CLINICAL RELEVANCE Ovarian texture information in addition to morphometric variables can enhance ultrasonographic applications for assisted reproduction of eels and potentially other fish species. This was a novel, nonlethal method for classifying reproductive response of eels and the first objective texture analysis performed on ultrasonographic images of the gonads of fish.

Muscle hemorrhages are regularly observed in especially the tip of the tenderloin muscles of slaughter pigs. In order to characterize the hemorrhages, a macro- and microscopic examination of tenderloins with (n=5) and without (n=4) hemorrhages and the associated vertebral column was carried out. Furthermore, all columns were radiographed and two were CT scanned. Histologically, the muscle hemorrhages contained cells from bone marrow and growth line cartilage. Ventral epiphysiolysis in either the cranial or caudal epiphysis of the first lumbar vertebrae (L1) in 8 out of the 9 vertebral columns was present. In the 5 cases with tenderloin hemorrhage, similar hemorrhage with growth line cartilage was found within the fracture of the epiphysis. The hemorrhages develop secondarily to epiphysiolysis in the lumbar vertebrae, where the tenderloin attaches to the spine. The lesions probably develop around the time of stunning.

Volume measurements of the brain are of interest in the diagnosis of brain pathology. This is particularly so in the investigation hydrocephalus and canine cognitive dysfunction (CCD), both of which result in thinning of the cerebral cortex and enlarged ventricles. Volume assessment can be made using computed tomography or more usually magnetic resonance imaging (MRI). There is, however, some uncertainty in the interpretation of such volume data due to the great variation in skull size and shape seen in dog. In this retrospective study, we examined normal MRI images from 63 dogs < 6 years of age. We used a continuous variable, the cranial index (CrI) to indicate skull shape and compared it with MRI volume measurements derived using Cavalieri's principle. We found a negative correlation between CrI and the ratio of cortical to ventricular volume. Breeds with a high CrI (large laterolateral compared to rostrocaudal cranial cavity dimension) had a smaller ratio of cortical to ventricular volume (low C:V ratio) than breeds with lower CrI skull types. It is important to consider this effect of skull shape on the relative volume estimates of the cerebral cortex and ventricles when trying to establish if pathology is present.

Background Diagnosis of pulmonary thromboembolism (PTE) in dogs relies on computed tomography pulmonary angiography (CTPA), but detailed interpretation of CTPA images is demanding for the radiologist and only large vessels may be evaluated. New approaches for better detection of smaller thrombi include dual energy computed tomography (DECT) as well as computer assisted diagnosis (CAD) techniques. The purpose of this study was to investigate the performance of quantitative texture analysis for detecting dogs with PTE using grey-level co-occurrence matrices (GLCM) and multivariate statistical classification analyses. CT images from healthy (n = 6) and diseased (n = 29) dogs with and without PTE confirmed on CTPA were segmented so that only tissue with CT numbers between −1024 and −250 Houndsfield Units (HU) was preserved. GLCM analysis and subsequent multivariate classification analyses were performed on texture parameters extracted from these images. Results Leave-one-dog-out cross validation and receiver operator characteristic (ROC) showed that the models generated from the texture analysis were able to predict healthy dogs with optimal levels of performance. Partial Least Square Discriminant Analysis (PLS-DA) obtained a sensitivity of 94% and a specificity of 96%, while Support Vector Machines (SVM) yielded a sensitivity of 99% and a specificity of 100%. The models, however, performed worse in classifying the type of disease in the diseased dog group: In diseased dogs with PTE sensitivities were 30% (PLS-DA) and 38% (SVM), and specificities were 80% (PLS-DA) and 89% (SVM). In diseased dogs without PTE the sensitivities of the models were 59% (PLS-DA) and 79% (SVM) and specificities were 79% (PLS-DA) and 82% (SVM). Conclusion The results indicate that texture analysis of CTPA images using GLCM is an effective tool for distinguishing healthy from abnormal lung. Furthermore the texture of pulmonary parenchyma in dogs with PTE is altered, when compared to the texture of pulmonary parenchyma of healthy dogs. The models’ poorer performance in classifying dogs within the diseased group, may be related to the low number of dogs compared to texture variables, a lack of balanced number of dogs within each group or a real lack of difference in the texture features among the diseased dogs. Electronic supplementary material The online version of this article (doi:10.1186/s12917-017-1117-1) contains supplementary material, which is available to authorized users.

Background Surveillance for bovine genetic diseases in Denmark identified a hitherto unreported congenital syndrome occurring among progeny of a Holstein sire used for artificial breeding. A genetic aetiology due to a dominant inheritance with incomplete penetrance or a mosaic germline mutation was suspected as all recorded cases were progeny of the same sire. Detailed investigations were performed to characterize the syndrome and to reveal its cause. Results Seven malformed calves were submitted examination. All cases shared a common morphology with the most striking lesions being severe facial dysplasia and complete prolapse of the eyes. Consequently the syndrome was named facial dysplasia syndrome (FDS). Furthermore, extensive brain malformations, including microencephaly, hydrocephalus, lobation of the cerebral hemispheres and compression of the brain were present. Subsequent data analysis of progeny of the sire revealed that around 0.5% of his offspring suffered from FDS. High density single nucleotide polymorphism (SNP) genotyping data of the seven cases and their parents were used to map the defect in the bovine genome. Significant genetic linkage was obtained for three regions, including chromosome 26 where whole genome sequencing of a case-parent trio revealed two de novo variants perfectly associated with the disease: an intronic SNP in the DMBT1 gene and a single non-synonymous variant in the FGFR2 gene. This FGFR2 missense variant (c.927G>T) affects a gene encoding a member of the fibroblast growth factor receptor family, where amino acid sequence is highly conserved between members and across species. It is predicted to change an evolutionary conserved tryptophan into a cysteine residue (p.Trp309Cys). Both variant alleles were proven to result from de novo mutation events in the germline of the sire. Conclusions FDS is a novel genetic disorder of Holstein cattle. Mutations in the human FGFR2 gene are associated with various dominant inherited craniofacial dysostosis syndromes. Given the phenotypic similarities in FDS affected calves, the genetic mapping and absence of further high impact variants in the critical genome regions, it is highly likely that the missense mutation in the FGFR2 gene caused the FDS phenotype in a dominant mode of inheritance. Electronic supplementary material The online version of this article (doi:10.1186/s12863-017-0541-3) contains supplementary material, which is available to authorized users.

Fractal analysis of canine pulmonary vessels could allow quantification of their space-filling properties. Aims of this prospective, analytical, cross-sectional study were to describe methods for reconstructing three dimensional pulmonary arterial vascular trees from computed tomographic pulmonary angiogram, applying fractal analyses of these vascular trees in dogs with and without diseases that are known to predispose to thromboembolism, and testing the hypothesis that diseased dogs would have a different fractal dimension than healthy dogs. A total of 34 dogs were sampled. Based on computed tomographic pulmonary angiograms findings, dogs were divided in three groups: diseased with pulmonary thromboembolism (n = 7), diseased but without pulmonary thromboembolism (n = 21), and healthy (n = 6). An observer who was aware of group status created three-dimensional pulmonary artery vascular trees for each dog using a semiautomated segmentation technique. Vascular three-dimensional reconstructions were then evaluated using fractal analysis. Fractal dimensions were analyzed, by group, using analysis of variance and principal component analysis. Fractal dimensions were significantly different among the three groups taken together (P = 0.001), but not between the diseased dogs alone (P = 0.203). The principal component analysis showed a tendency of separation between healthy control and diseased groups, but not between groups of dogs with and without pulmonary thromboembolism. Findings indicated that computed tomographic pulmonary angiogram images can be used to reconstruct three-dimensional pulmonary arterial vascular trees in dogs and that fractal analysis of these three-dimensional vascular trees is a feasible method for quantifying the spatial relationships of pulmonary arteries. These methods could be applied in further research studies on pulmonary and vascular diseases in dogs.

As the number of images per study increases in the field of veterinary radiology, there is a growing need for computer-assisted diagnosis techniques. The purpose of this study was to evaluate two machine learning statistical models for automatically identifying image regions that contain the canine hip joint on ventrodorsal pelvis radiographs. A training set of images (120 of the hip and 80 from other regions) was used to train a linear partial least squares discriminant analysis (PLS-DA) model and a nonlinear artificial neural network (ANN) model to classify hip images. Performance of the models was assessed using a separate test image set (36 containing hips and 20 from other areas). Partial least squares discriminant analysis model achieved a classification error, sensitivity, and specificity of 6.7%, 100%, and 89%, respectively. The corresponding values for the ANN model were 8.9%, 86%, and 100%. Findings indicated that statistical classification of veterinary images is feasible and has the potential for grouping and classifying images or image features, especially when a large number of well-classified images are available for model training.

Knowledge of congenital malformations and their causes in horses is generally sparse. Such conditions require more scientific attention to improve their diagnostics and inform prevention strategies. Here, a unique syndrome of bilateral oblique facial clefts (meloschisis), rudimentary eyes and hydrocephalus is reported in an equine foetus spontaneously aborted at gestation day 224. The cause of abortion was considered to be intrauterine death caused by umbilical cord torsions and subsequent compromised blood flow, but the aetiology of the malformation could not be determined. A detailed history, which includes exposure to a range of pharmaceutical compounds during the early stages of pregnancy, is provided and emphasizes the need for accurate recording of treatments in pregnant animals.

Communicating radiological reports to peers has pedagogical value. Students may be uneasy with the process due to a lack of communication and peer review skills or to their failure to see value in the process. We describe a communication exercise with peer review in an undergraduate veterinary radiology course. The computer code used to manage the course and deliver images online is reported, and we provide links to the executable files. We tested to see if undergraduate peer review of radiological reports has validity and describe student impressions of the learning process. Peer review scores for student-generated radiological reports were compared to scores obtained in the summative multiple choice (MCQ) examination for the course. Student satisfaction was measured using a bespoke questionnaire. There was a weak positive correlation (Pearson correlation coefficient = 0.32, p

There is no evidence-based diagnostic approach for diagnosis of pulmonary thromboembolism (PTE) in dogs. Many dogs with diseases that predispose to thrombosis are hypercoagulable when assessed with thromboelastography (TEG), but no direct link has been established. The aims of this study were: (1) to investigate if diseased dogs with PTE, diagnosed by computed tomography pulmonary angiography (CTPA), had evidence of hypercoagulability by TEG; (2) to characterise haemostatic and inflammatory changes in dogs with PTE; (3) to construct models for prediction of PTE based on combinations of haemostatic and inflammatory variables; and (4) to evaluate the performance of D-dimer measurement for prediction of PTE. Twenty-five dogs were included in this prospective observational study (PTE: n=6; non-PTE: n=19). Clot strength G values did not differ between the PTE and non-PTE groups in tissue factor (TF) or kaolin-activated TEG analyses. Haemostatic and inflammatory variables did not differ between the two groups. Linear discriminant analysis generated a model for prediction of PTE with a sensitivity and specificity of 100% when TF results were used as TEG data, and a model with sensitivity of 83% and specificity of 100% when kaolin results were used as TEG data. Receiver operating characteristic analysis of D-dimer levels showed that a value of >0.3mg/L yielded a sensitivity of 100% and a specificity of 71.4%. In conclusion, the study supports CTPA as method for diagnosing canine PTE, but shows that TEG alone cannot identify dogs with PTE. Models for prediction of PTE were generated, but require further validation.

Background Lethal chondrodysplasia (bulldog syndrome) is a well-known congenital syndrome in cattle and occurs sporadically in many breeds. In 2015, it was noticed that about 12 % of the offspring of the phenotypically normal Danish Holstein sire VH Cadiz Captivo showed chondrodysplasia resembling previously reported bulldog calves. Pedigree analysis of affected calves did not display obvious inbreeding to a common ancestor, suggesting the causative allele was not a rare recessive. The normal phenotype of the sire suggested a dominant inheritance with incomplete penetrance or a mosaic mutation. Results Three malformed calves were examined by necropsy, histopathology, radiology, and computed tomography scanning. These calves were morphologically similar and displayed severe disproportionate dwarfism and reduced body weight. The syndrome was characterized by shortening and compression of the body due to reduced length of the spine and the long bones of the limbs. The vicerocranium had severe dysplasia and palatoschisis. The bones had small irregular diaphyses and enlarged epiphyses consisting only of chondroid tissue. The sire and a total of four affected half-sib offspring and their dams were genotyped with the BovineHD SNP array to map the defect in the genome. Significant genetic linkage was obtained for several regions of the bovine genome including chromosome 5 where whole genome sequencing of an affected calf revealed a COL2A1 point mutation (g.32473300 G > A). This private sequence variant was predicted to affect splicing as it altered the conserved splice donor sequence GT at the 5’-end of COL2A1 intron 36, which was changed to AT. All five available cases carried the mutant allele in heterozygous state and all five dams were homozygous wild type. The sire VH Cadiz Captivo was shown to be a gonadal and somatic mosaic as assessed by the presence of the mutant allele at levels of about 5 % in peripheral blood and 15 % in semen. Conclusions The phenotypic and genetic findings are comparable to a previously reported COL2A1 missense mutation underlying lethal chondrodysplasia in the offspring of a mosaic French Holstein sire (Igale Masc). The identified independent spontaneous splice site variant in COL2A1 most likely caused chondrodysplasia and must have occurred during the early foetal development of the sire. This study provides a first example of a dominant COL2A1 splice site variant as candidate causal mutation of a severe lethal chondrodysplasia phenotype. Germline mosaicism is a relatively frequent mechanism in the origin of genetic disorders and explains the prevalence of a certain fraction of affected offspring. Paternal dominant de novo mutations are a risk in cattle breeding, especially because the ratio of defective offspring may be very high and be associated with significant animal welfare problems. Electronic supplementary material The online version of this article (doi:10.1186/s12917-016-0739-z) contains supplementary material, which is available to authorized users.

Imaging for suspected spinal disease not related to intervertebral disc disease is appropriate in cases where: clinical signs are referable to the spine or there is a breed predisposition for congenital abnormalities; there is a history of severe trauma, e.g. road traffic accident; there is a suspicion of spinal disease from other imaging studies. This chapter explores indications; normal survey or contrast studies; abnormal imaging findings.

Radiography is an excellent tool to generate diagnostic information in cases where either a skeletal lesion or systemic disease with skeletal manifestation is suspected. The fact that radiographic equipment is available in almost all veterinary practices makes it the imaging method most often used initially to detect and characterize lesions of bones and joints. This chapter covers radiography techniques and the basics of musculoskeletal radiography.

Purpose To compare the distribution and uptake of the hypoxia tracer 64 Cu-diacetyl-bis( N 4 -methylthiosemicarbazone) ( 64 Cu-ATSM) PET/CT, FDG PET/CT and dynamic contrast enhanced perfusion CT (DCE-pCT) in spontaneous canine tumors. In addition 64 Cu-ATSM distribution over time was evaluated. Methods and materials Nine spontaneous cancer-bearing dogs were prospectively enrolled. FDG (1h pi.) and 64 Cu-ATSM (3 and 24h pi.) PET/CT were performed over three consecutive days. DCE-pCT was performed on day 2. Tumor uptake of FDG and 64 Cu-ATSM was assessed semi-quantitatively and the distribution of FDG, 64 Cu-ATSM and CT perfusion parameters correlated. Results 64 Cu-ATSM distribution on scans performed 24h apart displayed moderate to strong correlation; however, temporal changes were observed. The spatial distribution pattern of 64 Cu-ATSM between scans was moderately to strongly positively correlated to FDG, whereas the correlation of CT perfusion parameters to FDG and to 64 Cu-ATSM yielded more varying results. Conclusions 64 Cu-ATSM uptake was positively correlated to FDG. 64 Cu-ATSM was found to be relatively stable between PET scans performed at different time points, important temporal changes were however observed in hypo-perfused regions. These findings potentially indicate that prolonged uptake periods for 64 Cu-ATSM imaging may be needed. Although a moderate to strong correlation between 64 Cu-ATSM and FDG PET/CT is observed, the two tracers provide different biological information with an overlapping spatial distribution.

The aim of the study was to construct a screening programme for disseminated histiocytic sarcoma (DHS) in Bernese Mountain dogs using diagnostic imaging and blood analysis and evaluate blood borne biomarkers as early disease detection biomarkers. Healthy Bernese Mountain dogs were screened on four occasions in an attempt to detect early disease. Eleven blood borne biomarkers were examined for their worth as early tumour biomarkers. During 2.5 years, five dogs with early DHS were identified; four of these by diagnostic imaging. No dogs developed symptomatic DHS without being detected within 6 months of the screening programme. Only serum ferritin showed potential as a blood borne marker of the disease. Median survival times for the dogs with early DHS were 226 days. Screening programmes every 6 months for Bernese Mountain dogs over 4 years of age including diagnostic imaging and ferritin measurements may identify early DHS.

Dual energy X-ray absorptiometry (DEXA) is a reference method for assessing body composition but is seldom `accessible in veterinary settings. Computed tomography (CT) can provide similar body composition estimates and we propose that it can be used in body composition studies in animals. We compared CT and DEXA data from 73 healthy adult neutered domestic cats. Three approaches for measuring adipose tissue percentage from full-body CT scans were explored. By examining the frequency distribution of voxels by Hounsfield unit (HU) value, it is possible to calculate a fat index (Fat%) that is in close agreement with the fat percentages obtained from DEXA scans. Fat% values obtained by the best of the methods had a mean difference of 0.96% (95% confidence interval 0.33-1.59%) from the DEXA results. Fat% obtained by the other two methods were characterized by good correlation but poor agreement and in one of the methods, the difference between the values from the two modalities was proportional to their mean. By using CT, it is possible to obtain body composition estimates that are in close agreement with those available using DEXA. While the significance of individual Fat% measurements obtained from CT can be difficult to interpret and to compare between centers, CT can contribute to research studies concerned either with nutrition or with obesity-related disorders.

2-Deoxy-2-[¹⁸F]fluoro-D-glucose positron emission tomography/computed tomography (FDG PET/CT) is becoming increasingly available as an imaging modality in veterinary medicine. The purpose of this study was to report semiquantitative standard uptake values (SUV) of malignant and nonmalignant tissues and organs in canine cancer patients. FDG PET/CT was performed in 14 dogs including, nine mesenchymal tumors, four carcinomas, and one incompletely excised mast cell tumor. A generally higher FDG uptake was observed in carcinomas relative to sarcomas. Maximum SUV of carcinomas ranged from 7.6 to 27.0, and for sarcomas from 2.0 to 10.6. The FDG SUV of several organs and tissues, including regional brain uptake is reported, to serve as a reference for future FDG PET studies in canine cancer patients. Several potential pitfalls have been recognized in interpretation of FDG PET images of human patients, a number of these were also observed in this study.

The surgical treatment of impingement of dorsal spinous processes (DSPs) at 2 equine veterinary hospitals following a similar diagnostic and surgical protocol is described with special reference to the diagnosis and case selection. The diagnosis was made from clinical examination and results of radiography and diagnostic analgesia of the affected sites. Scintigraphy was also used as an aid to diagnosis at one hospital. Surgical treatment was generally recommended when conservative treatment, which included 3-9 months rest, intralesional corticosteroids and physiotherapy, had failed. The DSPs were approached through a midline incision in the skin and supraspinous ligament. They were resected using an oscillating saw. Postoperatively, horses were returned to lungeing work after 2 weeks and riding commenced at 3 months, in most cases, and by 6 months in all successful cases. From one to 6 DSPs between thoracic vertebra (T) 11 and lumbar vertebra (L) 3 were resected in individual horses. T15, 16 and 17 were the spines most commonly treated. Long-term follow-up information was obtained in 209 horses. Seventy-two percent of these horses returned to full work and a further 9% improved sufficiently to be used for some athletic work.

OBJECTIVE. Our objective was to create an electronic learning tool that incorporates the pedagogic advantages of peer assessment and self-assessment into report-generating skills.CONCLUSION. The tool was created using Web programming software. It was tested with 12 veterinary students and provided peer–peer and tutor–student interactions that supported learning. The tool is suited for training any professional concerned with the interpretation of images, particularly resident radiologists.

Objective assessment of pulmonary disease from computed tomography (CT) examinations is desirable but difficult. When such assessments can be made, it is important that they are related to some part of the pathophysiologic process present. Herein we propose that automated volume histogram analysis can yield data that allow differentiation of normal from abnormal lung, and that the magnitude of disease will have an association with objective CT indices. Data from pulmonary CT images from 34 foxes (six uninfected controls and 28 infected with Angiostrongylus vasorum, subdivided by age and infective dose) were available. Lung tissue was segmented from surrounding tissue using an automated segmentation method. A volume histogram showing voxel frequency for each CT number in the range -1024 to -250 HU was created from the entire image stack from each fox. Using these data, the inter-quartile range and the CT number at the 95th percentile were determined. The results showed that segmentation could be readily achieved but that areas of severely diseased lung were excluded. Based on two-way analysis of variance for both the inter-quartile range and the CT number at the 95th percentile, both quantities were significantly affected by the infection status of the animal and were related to worm burden (P < 0.001). The study shows that this form of analysis is readily achieved and provides quantitative data that can be used to assess disease severity, progression, and response to treatment.

The location of adipose tissue depots is important in determining their significance. Research into the physical and chemical differences between these depots is therefore of interest. Using image analysis, this paper examines the influence of location on the linear attenuation coefficient of adipose tissue for X-rays, in computed tomography (as indicated by CT number) at three time points. Nine pigs were CT scanned on three separate occasions approximately 1 month apart. The mean CT number was -78, -100, and -104 for visceral adipose tissue (VAT) from the first to the final scan, respectively. The corresponding CT numbers for subcutaneous adipose tissue (SAT) were -80, -101, and -106. There was a significant difference between the CT numbers at each location at each scan (P values from 0.025 to

Changes in the Hounsfield Unit value of adipose tissue and of non-adipose soft tissue during growth are poorly documented. This study examines the HU of these tissues in growing pigs. Computer tomography scans were made in nine growing pigs on three occasions, approximately four weeks apart. Average body weight was 51, 94, and 121 kg for each successive scan. Images from the level of the diaphragm to the hips were analyzed. The mean HU and its standard deviation, for adipose tissue and for non-adipose soft tissues was determined for each pig. The mean adipose tissue HU for all pigs was -90, -98 and -101 on the first, middle and last scans, respectively. Corresponding HU values for non-adipose soft tissues were, 52, 51 and 49. There was a significant difference (p0.01) between HU at each scan time for each set of tissues.

An estimate of the thickness of subcutaneous adipose tissue at differing positions around the body was required in a study examining body composition. To eliminate human error associated with the manual placement of markers for measurements and to facilitate the collection of data from a large number of animals and image slices, automation of the process was desirable. The open-source and free image analysis program ImageJ was used. A macro procedure was created that provided the required functionality. The macro performs a number of basic image processing procedures. These include an initial process designed to remove the scanning table from the image and to center the animal in the image. This is followed by placement of a vertical line segment from the mid point of the upper border of the image to the image center. Measurements are made between automatically detected outer and inner boundaries of subcutaneous adipose tissue along this line segment. This process was repeated as the image was rotated (with the line position remaining unchanged) so that measurements around the complete circumference were obtained. Additionally, an image was created showing all detected boundary points so that sites of measurement could be checked by visual inspection. This paper provides a detailed account of this macro, illustrates the effect of its components on the image, and the applicability of standard image processing techniques to veterinary imaging.

Veterinary radiology is a long established subject discipline in veterinary science. It is fair to say that every veterinary graduate alive today will have received formal training in this subject irrespective of their date or place of graduation. Despite or perhaps because of this long ancestry, it is worthwhile examining the extent of the subject’s boundaries and its place in veterinary medicine. One can ask what imaging modalities fall under the remit of the subject and why they do so. Veterinary activities have always been concerned with the diagnosis and treatment of disease, with clinical and experimental animal research, and with agriculture, to select but a few areas of interest. How great a contribution can veterinary imaging make to these areas? The limits of what radiologists can and cannot do becomes unclear as the breakdown of traditional barriers between disciplines, considered essential to progress in medical education and scientific progress, continues (1). It has always been the case that the collection and use of image-related data are not the sole preserve of veterinary radiologists; uncertainties can exist as to who should be involved. The radiology community, both users and suppliers alike, has to ask what value imaging brings to collaborative work and how this value is best realized (2). If uncertainties are accepted, they can be seen to arise from a number of recognizable factors. Some are longstanding while others are quite new. Important factors relate to the wide range of veterinary involvement in biological sciences and the roles that veterinarians take on in society. As the profession’s sphere of involvement extends, so too do the frontiers that veterinary imaging reach. Also long a factor, but certainly an accelerating trend, is the arrival of new technologies, such as multi-slice imaging that generates data in quantities, orders of magnitude greater than their predecessors. A newer phenomenon is the rise of widely available Patient Archive and Communications Systems (PACS). These can separate the radiologist from the clinic, or to put it more positively, can allow non-centralized clinicians access to radiologists at central remote sites. Having to deal with new technologies has always been a feature of radiology. This is seen from its far past with the transition from saving images on glass plates to the use of emulsions mounted on celluloid, from manual to automated film processing and more recently, from two-dimensional imaging to three-dimensional multislice technologies that allow volumetric acquisitions. The last of these, which allows sub-millimeter image slices, have the same dimensions in three orthogonal planes. These isovolume voxels represent a massive amount of data and allow superior planar reconstructions which in turn demand new approaches to reading studies (3). Each transition represents an order of magnitude change and in each case the radiology imaging community has risen to the challenge. This article will trace some of these issues in the hope of seeing how they might be extrapolated into the future. We are all radiologists some of our working time and a very few of us are radiologists all of the time. It will benefit each individual along this spectrum of involvement, to have an awareness of each other’s presence. By looking at past changes and future prospects, we can better appreciate the subject that is veterinary imaging in all its manifestations.

Glycolysis, hypoxia, and proliferation are important factors in the tumor microenvironment contributing to treatment-resistant aggressiveness. Imaging these factors using combined functional positron emission tomography and computed tomography can potentially guide diagnosis and management of cancer patients. A dog with fibrosarcoma was imaged using 18F-FDG, 64Cu-ATSM, and 18F-FLT before, during, and after 10 fractions of 4.5 Gy radiotherapy. Uptake of all tracers decreased during treatment. Fluctuations in 18F-FDG and 18F-FLT PET uptakes and a heterogeneous spatial distribution of the three tracers were seen. Tracer distributions partially overlapped. It appears that each tracer provides distinct information about tumor heterogeneity and treatment response.

The objective of this study was to characterize the ultrasonographic patterns of normal superficial lymph nodes and to evaluate whether ultrasonography can help discriminate between different lymphadenopathies (reactive, lymphoma, and metastases) in dogs. Three hundred and eighteen superficial lymph nodes in 142 dogs were studied by B-mode, color flow mapping, power, and spectral Doppler ultrasonography. Size, echogenicity, nodal border definition, presence of a nodal hilus, acoustic enhancement and distribution of vascular flow, as well as perfusion indices were measured. Multivariate statistics using discriminant analysis was used to determine which parameters can be used to predict the diagnosis of the lymph node. The size of the lymph node, distribution of vascular flow within the lymph node, and pulsatility index (PI) in combination gave a classification error of 23% for the four groups of lymph nodes. This was improved to 11% if the nodes were divided into two groups: benign and malignant. There was a significant difference in resistive index (RI) and PI between benign and malignant nodes. Cut-off values were determined using receiver operator curves, 0.68 RI and 1.49 PI.

As part of a prospective study in bone mineralisation in adult pigs it was necessary to establish guidelines and to define sites for bone mineral measurements. Particular requirements were that, the protocol should be suitable for a mass screening programme in both postmortem specimens and in live animals, and should deliver results of known reliability. Estimates of bone mineral content (BMC) and bone mineral density (BMD) in areas within the 4th metacarpal bone yielded coefficients of variation (CV) in the order of 7% for both regions and estimates in regions which included the entire metacarpal-phalangeal area yielded CV values in the order of 0.7% and 0.6% for BMC and BMD, respectively. A region of interest taken from the coccygeal vertebrae yielded coefficient of variation values of 3% and 2% for BMC and BMD, respectively. Accuracy of dual-energy X-ray absorptiometry (DXA) was estimated using a standard curve derived from BMC determined by ashing. There was a high correlation between mineral content determined by DXA and by ashing (R(2)=0.99, p

Forty-three guinea pigs presented as part of a screening programme were divided into six categories on the basis of age and reproductive history. Three age groupings were used: less than one year, one to two years, and over two years. Each age group was further divided into two categories: animals which had had single or multiple pregnancies, and animals which had never bred. Each animal received a clinical examination and was scanned using B-mode ultrasound for the presence of ovarian cysts. Possible relationships between reproductive history and the prevalence of cysts were investigated using Mantel-Haenzel analysis, and between prevalence, cyst size and age using linear regression analysis. Two out of the 43 animals (4·7 per cent) showed symmetrical alopecia. No statistically significant correlation between reproductive history and the prevalence of cysts was detected at the 95 per cent confidence level. A statistically significant relationship was found, however, between cyst size and age (P< 0 ·01) and between cyst prevalence and age (P < 0·02).

An experimental venous clot model using the lateral saphenous vein of sheep is described. Eight experimental Suffolk crossbred sheep were used. A mixture of human fibrinogen, in some cases labelled with (125)I, bovine thrombin and homologous whole blood was placed via a catheter into a surgically isolated segment of the lateral saphenous vein. The resulting heterologous clot was imaged daily for 6 days using venography, or monitored using an external gamma ray detector. Clots were radiographically detectable for the 6 days of the study. They were totally occlusive for a mean of 4.2 days (SD 2.2) and were occlusive in the immediate 24 hour period after surgery. The fibrin component of the clot was persistent (85 per cent of the initial fibrin[ogen] present after 6 days). Radiographically the clots were seen as filling defects within partially filled vessels, or their presence was inferred from the absence of filling. A collateral blood supply was apparent immediately on vessel occlusion. No adverse effects, evidence of infection or limb oedema were seen. The model provided a reproducible blood clot within the lateral saphenous vein of the sheep. Clot imaging using venography was effective and readily achieved. It is suggested that the model is useful when a stable, intravenous deposit of heterologous (e.g. human) fibrin is required in vivo at a site suitable for venography and radionucleid monitoring.