Clinical respiratory and radiographic abnormalities in dogs with spontaneous hypercortisolism.

The objective of this study was to characterize respiratory clinical signs, other than panting and respiratory distress, as well as thoracic radiographic abnormalities, in dogs with hypercortisolism. Although there have been case reports and studies evaluating the results of pulmonary scintigraphy, no studies have yet reviewed respiratory clinical signs and radiograph results in dogs with hypercortisolism. This study addresses this gap. A case series was evaluated and the dogs' clinical histories were obtained, including clinical signs and physical examination. Digital thoracic radiographs were analyzed to consider such parameters as the diameter of the main bronchi, lung patterns, and the size of the pulmonary trunk. The most common respiratory symptoms were snoring (61.9%), coughing (57.1%), and fatigue (52.4%). Physical examination revealed a high frequency of changes on lung auscultation (95.2%). The body condition score (BCS) was high in 95% of dogs and a significant correlation was observed between the presence of cyanosis and changes in lung auscultation, both of which present similar risk factors. Furthermore, body weight showed a moderate correlation with respiratory rate (RR = 0.571). Radiographic changes were evident in 47.5% of dogs, with the bronchial pattern being the most common (70%). Based on these results, it was observed that respiratory and radiographic abnormalities are frequent in dogs with spontaneous hypercortisolism and a high body condition score was relevant for exacerbating clinical respiratory signs, such as cyanosis and tachypnea.

L'objectif de cette étude était de caractériser les signes cliniques respiratoires, autres que le halètement et la détresse respiratoire, ainsi que les anomalies radiographiques thoraciques, chez les chiens atteints d'hypercortisolisme. Bien qu'il existe des rapports de cas et des études évaluant les résultats de la scintigraphie pulmonaire, aucune étude n'a encore examiné les signes cliniques respiratoires et les résultats radiographiques chez les chiens atteints d'hypercortisolisme. Cette étude comble cette lacune. Une série de cas a été évaluée et les antécédents cliniques des chiens ont été obtenus, incluant les signes cliniques et l'examen physique. Des radiographies thoraciques numériques ont été analysées pour prendre en compte des paramètres tels que le diamètre des bronches principales, les schémas pulmonaires et la taille du tronc pulmonaire. Les symptômes respiratoires les plus courants étaient le ronflement (61,9 %), la toux (57,1 %) et la fatigue (52,4 %). L'examen physique a révélé une fréquence élevée de changements à l'auscultation pulmonaire (95,2 %). Le score d'état corporel (BCS) était élevé chez 95 % des chiens et une corrélation significative a été observée entre la présence de cyanose et les modifications de l'auscultation pulmonaire, qui présentent toutes deux des facteurs de risque similaires. De plus, le poids corporel a montré une corrélation modérée avec la fréquence respiratoire (RR = 0,571). Des changements radiographiques étaient évidents chez 47,5 % des chiens, le schéma bronchique étant le plus courant (70 %). Sur la base de ces résultats, il a été observé que les anomalies respiratoires et radiographiques sont fréquentes chez les chiens atteints d'hypercortisolisme spontané, et qu'un score d'état corporel élevé était pertinent pour l'exacerbation des signes respiratoires cliniques, tels que la cyanose et la tachypnée.(Traduit par Docteur Serge Messier).

Development of an artificial intelligence-based algorithm for predicting the severity of myxomatous mitral valve disease from thoracic radiographs by using two grading systems.

A heart-convolutional neural network (heart-CNN) was designed and tested for the automatic classification of chest radiographs in dogs affected by myxomatous mitral valve disease (MMVD) at different stages of disease severity. A retrospective and multicenter study was conducted. Lateral radiographs of dogs with concomitant X-ray and echocardiographic examination were selected from the internal databases of two institutions. Dogs were classified as healthy, B1, B2, C and D, based on American College of Veterinary Internal Medicine (ACVIM) guidelines, and as healthy, mild, moderate, severe and late stage, based on Mitral INsufficiency Echocardiographic (MINE) score. Heart-CNN performance was evaluated using confusion matrices, receiver operating characteristic curves, and t-SNE and UMAP analysis. The area under the curve (AUC) was 0.88, 0.88, 0.79, 0.89 and 0.84 for healthy and ACVIM stage B1, B2, C and D, respectively. According to the MINE score, the AUC was 0.90, 0.86, 0.71, 0.82 and 0.82 for healthy, mild, moderate, severe and late stage, respectively. The developed algorithm showed good accuracy in predicting MMVD stages based on both classification systems, proving a potentially useful tool in the early diagnosis of canine MMVD.

The vertebral right heart index: A new radiographic method to assess right heart enlargement in dogs.

In veterinary medicine, the radiographic assessment of right heart enlargement (RHE) is essentially subjective. The aim of this study was to evaluate the vertebral right heart index (VRHi) as a new quantitative radiographic method to detect RHE in dogs. This was a multicenter, retrospective, observational study, including dogs with RHE and control dogs. All dogs had to have a thoracic radiographic study and a complete echocardiography on the same day. Right heart enlargement was defined as the presence of right atrial enlargement, right ventricular enlargement, and/or hypertrophy based on echocardiography. For the radiographic study, all the radiographic views available for each case were considered for measurement: right lateral (RL), left lateral (LL), ventrodorsal (VD), and dorsoventral (DV). The VRHi was measured using LL, RL, VD, and DV views. A total of 204 dogs were included: 91 dogs with RHE and 113 control dogs. The VRHi (RL), the VRHi (LL), and the VRHi (VD) were significantly greater in dogs with RHE compared with controls (P < .0001). The best diagnostic accuracy was observed for VRHi (LL) (area under the curve [AUC] 0.86, P < .0001; cutoff ≥ 3.5 vertebrae, sensitivity [Se] 71%, specificity [Sp] 89%), followed by VRHi (RL) (AUC 0.85, P < .0001; cutoff ≥ 3.5 vertebrae, Se 68%, Sp 86%) and VRHi (VD) (AUC 0.80, P = .0004; cutoff ≥ 3.0 vertebrae, Se 57%, Sp 95%). In conclusion, the lateral VRHi in LL and RL and the VD VRHi could represent useful radiological tools for the detection of RHE in dogs.

Classification of the quality of canine and feline ventrodorsal and dorsoventral thoracic radiographs through machine learning.

Thoracic radiographs are an essential diagnostic tool in companion animal medicine and are frequently used as a part of routine workups in patients presenting for coughing, respiratory distress, cardiovascular diseases, and for staging of neoplasia. Quality control is a critical aspect of radiology practice in preventing misdiagnosis and ensuring consistent, accurate, and reliable diagnostic imaging. Implementing an effective quality control procedure in radiology can impact patient outcomes, facilitate clinical decision-making, and decrease healthcare costs. In this study, a machine learning-based quality classification model is suggested for canine and feline thoracic radiographs captured in both ventrodorsal and dorsoventral positions. The problem of quality classification was divided into collimation, positioning, and exposure, and then an automatic classification method was proposed for each based on deep learning and machine learning. We utilized a dataset of 899 radiographs of dogs and cats. Evaluations using fivefold cross-validation resulted in an F1 score and AUC score of 91.33 (95% CI: 88.37-94.29) and 91.10 (95% CI: 88.16-94.03), respectively. Results indicated that the proposed automatic quality classification has the potential to be implemented in radiology clinics to improve radiograph quality and reduce nondiagnostic images.

Pulmonary interstitial emphysema and spontaneous pneumomediastinum are more prevalent in sighthounds than other dog breeds undergoing thoracic CT.

Spontaneous pneumomediastinum is an uncommon condition described in veterinary medicine with concurrent respiratory disease. It can be caused by the Macklin effect, which is when gas leaks from the alveoli into the surrounding interstitial lung tissue. Pulmonary interstitial emphysema (PIE) is the presence of gas within the pulmonary vascular sheaths and indicates the presence of the Macklin effect. In the authors' experience, spontaneous pneumomediastinum and PIE are more prevalent in sighthound dogs than in other breeds and are often considered incidental findings. This retrospective, observational, cross-sectional study compared the prevalence of PIE and subsequent pneumomediastinum in sighthound with other purebred dogs. It characterized the appearance of PIE in CT and analyzed a possible association with concomitant pulmonary pathologies or with the use of general anesthesia. Medical records and thoracic CTs of sighthounds and nonsighthound dogs from two institutions were reviewed. A total of 256 dogs, comprising 127 sighthounds and 129 other purebred dogs, were included. The prevalence of PIE and pneumomediastinum was statistically higher in sighthound (14.2%) compared with other nonsighthound dogs (2.2%). There was no statistical association between the presence of PIE and pneumomediastinum with different age and weight groups or with sex. There was no statistical difference in the prevalence of PIE and pneumomediastinum between dogs with and without pulmonary pathology or in dogs undergoing CT with sedation or general anesthesia. Spontaneous pneumomediastinum in sighthounds is more prevalent than in other breeds, and its prevalence is not associated with the presence of pulmonary pathology or the use of general anesthesia.

Vertebral heart size and vertebral left atrial size reference ranges in healthy Miniature Schnauzers.

OBJECTIVE: The inclusion of vertebral heart score (VHS) and, more recently, the inclusion of the vertebral left atrial size (VLAS) in radiographic evaluation have become important screening tools for identifying dogs with occult cardiac disease. Several recent papers have shown there are interbreed variations in the VHS reference range. Our hypothesis is that the Miniature Schnauzer would also have a higher reference range for its VHS. ANIMALS: The electronic medical records of IDEXX Telemedicine Consultants were searched for Miniature Schnauzers undergoing thoracic radiographs between March 1, 2022, and February 28, 2023. METHODS: Dogs were included if they had 3 view thoracic radiographs performed and no evidence of cardiopulmonary disease was detected. Dogs with incomplete radiographic studies or cardiac or extracardiac disease were excluded. The VHS and VLAS measurements were performed by 2 board-certified cardiologists independent of one another. RESULTS: A total of 1,000 radiographs were obtained of which 272 were included for the study. The overall range for the VHS in this cohort was 9.68 to 12.07 with a median of 10.9. For VLAS measurements, a range of 1.71 to 2.4 was documented with a median of 2.0. CLINICAL RELEVANCE: The VHS for Miniature Schnauzers without cardiac disease was confirmed to be higher than the canine reference range.

Radiographic and computed tomographic characteristics of intraluminal tracheal adenoma in a cat.

A 13-year-old spayed female Persian cat presented with dyspnea and nasal discharge. Thoracic radiography revealed a dome-shaped soft-tissue opacity in the carina. Computed tomography confirmed a soft tissue-attenuating mass in the carina and the left and right proximal main bronchi that appeared to arise from the tracheal wall. Tracheoscopy revealed an intraluminal broad-based mass with multilobulated borders at the same location. Histopathological evaluation revealed a benign neoplastic process of the glandular epithelial lineage, which was considered an adenoma. Tracheal adenomas should be included in the differential diagnosis of tracheal masses.

Physical examination and CT to assess thoracic injury in 137 cats presented to UK referral hospitals after trauma.

OBJECTIVES: The aim of the study was to describe clinical examination and thoracic CT (TCT) findings in cats after trauma, and to identify physical examination findings associated with both abnormalities on TCT and the need for therapeutic interventions. METHODS: A multicentre, retrospective, observational study was conducted. Cats admitted to the participating hospitals with a history of blunt trauma and that underwent TCT were eligible. Data were collected on signalment, history, physical examination, TCT findings and subsequent interventions. RESULTS: In total, 137 cats were included. Road traffic accidents (RTAs) were the most frequently reported cause of trauma (69%). Tachypnoea (32%), pale mucous membranes (22%) and dyspnoea (20%) were the most common abnormal findings on thoracic examination. The most frequently identified thoracic pathologies on TCT were atelectasis (34%), pulmonary contusions (33%), pneumothorax (29%) and pleural effusion (20%). Thoracocentesis was the most commonly performed intervention (12%), followed by chest drain placement (7%). A total of 45 (33%) cats had no physical examination abnormalities but did have abnormalities detected on TCT; six of these cats required interventions. Increasing numbers of thoracic abnormalities on clinical examination were associated with increasing likelihood of having abnormal findings on TCT (odds ratio [OR] 2.04, 95% confidence interval [CI] 1.21-3.44, P = 0.008) and of requiring an intervention (OR 1.82, 95% CI 1.32-2.51, P <0.001). CONCLUSIONS AND RELEVANCE: RTAs were the most common reported cause of blunt trauma. Atelectasis, pulmonary contusions and pneumothorax were the most common abnormalities identified on TCT, and thoracic drainage was the most utilised intervention. TCT may be useful in identifying cats with normal thoracic physical examination findings that have significant thoracic pathology, and a high number of abnormal findings on thoracic examination should raise suspicion for both minor and major thoracic pathology. The results of this study can be used to assist in selecting appropriate cases for TCT after blunt trauma.

An AI-based algorithm for the automatic evaluation of image quality in canine thoracic radiographs.

The aim of this study was to develop and test an artificial intelligence (AI)-based algorithm for detecting common technical errors in canine thoracic radiography. The algorithm was trained using a database of thoracic radiographs from three veterinary clinics in Italy, which were evaluated for image quality by three experienced veterinary diagnostic imagers. The algorithm was designed to classify the images as correct or having one or more of the following errors: rotation, underexposure, overexposure, incorrect limb positioning, incorrect neck positioning, blurriness, cut-off, or the presence of foreign objects, or medical devices. The algorithm was able to correctly identify errors in thoracic radiographs with an overall accuracy of 81.5% in latero-lateral and 75.7% in sagittal images. The most accurately identified errors were limb mispositioning and underexposure both in latero-lateral and sagittal images. The accuracy of the developed model in the classification of technically correct radiographs was fair in latero-lateral and good in sagittal images. The authors conclude that their AI-based algorithm is a promising tool for improving the accuracy of radiographic interpretation by identifying technical errors in canine thoracic radiographs.

Performance of a commercially available artificial intelligence software for the detection of confirmed pulmonary nodules and masses in canine thoracic radiography.

Advancements in the field of artificial intelligence (AI) are modest in veterinary medicine relative to their substantial growth in human medicine. However, interest in this field is increasing, and commercially available veterinary AI products are already on the market. In this retrospective, diagnostic accuracy study, the accuracy of a commercially available convolutional neural network AI product (Vetology AI®) is assessed on 56 thoracic radiographic studies of pulmonary nodules and masses, as well as 32 control cases. Positive cases were confirmed to have pulmonary pathology consistent with a nodule/mass either by CT, cytology, or histopathology. The AI software detected pulmonary nodules/masses in 31 of 56 confirmed cases and correctly classified 30 of 32 control cases. The AI model accuracy is 69.3%, balanced accuracy 74.6%, F1-score 0.7, sensitivity 55.4%, and specificity 93.75%. Building on these results, both the current clinical relevance of AI and how veterinarians can be expected to use available commercial products are discussed.

Machine learning can appropriately classify the collimation of ventrodorsal and dorsoventral thoracic radiographic images of dogs and cats.

OBJECTIVES: To determine the feasibility of machine learning algorithms for the classification of appropriate collimation of the cranial and caudal borders in ventrodorsal and dorsoventral thoracic radiographs. SAMPLES: 900 ventrodorsal and dorsoventral canine and feline thoracic radiographs were retrospectively acquired from the Picture Archiving and Communication system (PACs) system of the Ontario Veterinary College. PROCEDURES: Radiographs acquired from April 2020 to May 2021 were labeled by 1 radiologist in Summer of 2022 as either appropriately or inappropriately collimated for the cranial and caudal borders. A machine learning model was trained to identify the appropriate inclusion of the entire lung field at both the cranial and caudal borders. Both individual models and a combined overall inclusion model were assessed based on the combined results of both the cranial and caudal border assessments. RESULTS: The combined overall inclusion model showed a precision of 91.21% (95% CI [91, 91.4]), accuracy of 83.17% (95% CI [83, 83.4]), and F1 score of 87% (95% CI [86.8, 87.2]) for classification when compared with the radiologist's quality assessment. The model took on average 6 ± 1 second to run. CLINICAL RELEVANCE: Deep learning-based methods can classify small animal thoracic radiographs as appropriately or inappropriately collimated. These methods could be deployed in a clinical setting to improve the diagnostic quality of thoracic radiographs in small animal practice.

Providing comparison normal examples alongside pathologic thoracic radiographic cases can improve veterinary students' ability to identify abnormal findings or diagnose disease.

Learning by comparison is a frequently employed education strategy used across many disciplines and levels. Interpreting radiographs requires both skills of perception and pattern recognition, which makes comparison techniques particularly useful in this field. In this randomized, prospective, parallel-group study, students enrolled in second and third-year radiology veterinary courses were given a case-based thoracic radiographic interpretation assignment. A cohort of the participants was given cases with side-by-side comparison normal images while the other cohort only had access to the cases. Twelve cases in total were presented to the students, with 10 cases depicting examples of common thoracic pathologies, while 2 cases were examples of normal. Radiographs of both feline and canine species were represented. Correctness of response to multiple choice questions was tracked, as was year and group (group 1: non compare, Control; group 2: compare, Intervention). Students assigned to group 1 had a lower percentage of correct answers than students assigned to group 2 (45% Control vs. 52% Intervention; P = 0.01). This indicates that side-by-side comparison to a normal example is helpful in identifying disease. No statistical significance was noted for the correctness of responses according to the year of training (P = 0.90). The overall poor performance on the assignment, regardless of group or year, shows that students in the early years of undergraduate veterinary radiology training struggle with the interpretation of common pathologies, likely a result of a lack of exposure to a multitude of cases and normal variants.

The effect of different body positions on the cardiothoracic ratios obtained by chest radiography in Japanese macaques (Macaca fuscata) and rhesus macaques (Macaca mulatta).

BACKGROUND: Although some studies have reported cardiac diseases in macaques, an adequate screening method for cardiac enlargement has not yet been established. This study aimed to evaluate the positioning of macaques for radiographs and establish reference intervals for the cardiothoracic ratio (CTR). MATERIALS AND METHODS: We developed a device for chest radiography in the sitting position and performed chest radiography in 50 Japanese and 48 rhesus macaques to evaluate the CTR and chest cavity size. RESULTS: In Japanese and rhesus macaques, the thorax height was significantly larger, the heart width was significantly smaller, and the mean CTR was significantly smaller in the sitting position than in the prone position. The reference intervals for CTR in the sitting position were 51.6 ± 4.6% and 52.2 ± 5.1% in Japanese and rhesus macaques, respectively. CONCLUSION: Thoracic radiographic images obtained in a sitting position resulted in a smaller CTR and a larger thorax height, which could be useful for detecting pulmonary and cardiac abnormalities.

Accuracy of artificial intelligence software for the detection of confirmed pleural effusion in thoracic radiographs in dogs.

The use of artificial intelligence (AI) algorithms in diagnostic radiology is a developing area in veterinary medicine and may provide substantial benefit in many clinical settings. These range from timely image interpretation in the emergency setting when no boarded radiologist is available to allowing boarded radiologists to focus on more challenging cases that require complex medical decision making. Testing the performance of artificial intelligence (AI) software in veterinary medicine is at its early stages, and only a scant number of reports of validation of AI software have been published. The purpose of this study was to investigate the performance of an AI algorithm (Vetology AI® ) in the detection of pleural effusion in thoracic radiographs of dogs. In this retrospective, diagnostic case-controlled study, 62 canine patients were recruited. A control group of 21 dogs with normal thoracic radiographs and a sample group of 41 dogs with confirmed pleural effusion were selected from the electronic medical records at the Cummings School of Veterinary Medicine. The images were cropped to include only the area of interest (i.e., thorax). The software then classified images into those with pleural effusion and those without. The AI algorithm was able to determine the presence of pleural effusion with 88.7% accuracy (P < 0.05). The sensitivity and specificity were 90.2% and 81.8%, respectively (positive predictive value, 92.5%; negative predictive value, 81.8%). The application of this technology in the diagnostic interpretation of thoracic radiographs in veterinary medicine appears to be of value and warrants further investigation and testing.

Septic pericardial disease and mediastinal abscessation in a cat with an intrathoracic needle foreign body.

OBJECTIVE: To describe the successful management of a cat with an intrathoracic sewing needle foreign body and septic pericardial effusion. CASE SUMMARY: A 10-year-old neutered female domestic longhair cat was referred for an intrathoracic metallic foreign body identified via thoracic radiography. Two weeks prior, the cat may have ingested a sewing needle. She was presented hemodynamically unstable; point-of-care thoracic ultrasound identified pericardial effusion with right atrial tamponade. Pericardiocentesis stabilized hemodynamic parameters. The effusion was grossly purulent, and bacterial culture grew an Actinomyces sp. The cat underwent a median sternotomy to remove the foreign body, debride associated mediastinal abscesses, and perform a partial pericardiectomy. Over the next 10 days, the cat was managed in hospital with a left unilateral thoracostomy tube and intermittent lavage of the pleural cavity. The cat was discharged, and follow-up at 7 days showed no evidence of pericardial or pleural effusion. NEW OR UNIQUE INFORMATION PROVIDED: Contrary to previous reports, this case shows that extra-gastrointestinal, specifically intrathoracic, sewing needle foreign bodies can cause significant morbidity in cats. To the authors' knowledge, this report is the first to describe septic pericardial disease resulting from documented foreign body ingestion in the cat. It is also the first case report of successful surgical management of mediastinal abscessation in the cat.

Bilateral thoracic radiographs increase lesion detection in horses with pneumonia or pulmonary neoplasia but do not bring any additional benefit for inflammatory or diffuse pulmonary disease.

Published studies describing the effects of bilateral radiographic projections on the detection of equine pulmonary lesions are currently lacking. The objectives of this retrospective, single center, observational study were to compare unilateral and bilateral thoracic radiographic projections for the detection of pulmonary lesions in a group of horses. Based on their clinical diagnosis, 167 adults and foals with bilateral thoracic radiographs were classified as having pneumonia (n = 88), inflammatory or diffuse pulmonary disease (n = 72), and pulmonary masses (n = 7). After an initial interrater repeatability test, right-to-left and left-to-right projections were anonymized and independently interpreted by a radiologist blinded to the clinical diagnosis. Scores were attributed for each pattern/lesion (alveolar, interstitial, bronchial, nodules/masses, cavitary lesions) and each quadrant. Agreement between scores from each projection was evaluated with Bland-Altman plots. Lesions identified on one side but not on the contralateral projection were considered discordant. There was no preferential lateralization of pulmonary lesions. The prevalence of discordance was 14.4%, 9.0%, and 4.2% for alveolar pattern, nodules/masses, and cavitary lesions, respectively. Up to nine horses (10.2%) with pneumonia could have been misdiagnosed. A pulmonary mass would have been missed in one case. For inflammatory or diffuse disease, discordance was slight, and the addition of contralateral projections had no impact on radiographic interpretation. In conclusion, in horses with pneumonia or neoplasia, bilateral projections, or adding at least one contralateral caudoventral view, increased the probability of identifying pulmonary lesions. In horses with inflammatory or diffuse disease, bilateral thoracic radiography provided no additional benefit.

Comparison of error rates between four pretrained DenseNet convolutional neural network models and 13 board-certified veterinary radiologists when evaluating 15 labels of canine thoracic radiographs.

Convolutional neural networks (CNNs) are commonly used as artificial intelligence (AI) tools for evaluating radiographs, but published studies testing their performance in veterinary patients are currently lacking. The purpose of this retrospective, secondary analysis, diagnostic accuracy study was to compare the error rates of four CNNs to the error rates of 13 veterinary radiologists for evaluating canine thoracic radiographs using an independent gold standard. Radiographs acquired at a referral institution were used to evaluate the four CNNs sharing a common architecture. Fifty radiographic studies were selected at random. The studies were evaluated independently by three board-certified veterinary radiologists for the presence or absence of 15 thoracic labels, thus creating the gold standard through the majority rule. The labels included "cardiovascular," "pulmonary," "pleural," "airway," and "other categories." The error rates for each of the CNNs and for 13 additional board-certified veterinary radiologists were calculated on those same studies. There was no statistical difference in the error rates among the four CNNs for the majority of the labels. However, the CNN's training method impacted the overall error rate for three of 15 labels. The veterinary radiologists had a statistically lower error rate than all four CNNs overall and for five labels (33%). There was only one label ("esophageal dilation") for which two CNNs were superior to the veterinary radiologists. Findings from the current study raise numerous questions that need to be addressed to further develop and standardize AI in the veterinary radiology environment and to optimize patient care.