This retrospective study describes the MRI features of extradural spinal neoplasia in 60 dogs and seven cats to identify potential distinguishing features between tumor classes and individual tumor types within each class. In dogs, mesenchymal tumors were most common (48%), with undifferentiated sarcomas being the predominant tumor type. Round cell neoplasms were second most common (35%), with lymphoma and multiple myeloma/plasma cell tumor comprising the majority of cases. Only two benign tumors were identified. In cats, lymphoma was most common (5/7), with one case of mesenchymal neoplasia and one case of metastatic carcinoma. Despite some overlap, certain imaging features were able to help prioritize differential diagnoses. The combined features that predicted round cell neoplasia (84%) included the preservation of vertebral shape, homogeneous contrast enhancement, and lesion centering on bone. The combined features that predicted mesenchymal neoplasia (73%) included altered vertebral shape, heterogeneous contrast enhancement, and lesion centering on paraspinal soft tissues. Round cell neoplasms were more likely to have cortical sparing, preservation of overall shape, lesion centering on bone, small soft tissue tumor size, and homogeneous contrast enhancement. Both epithelial and mesenchymal neoplasms were more likely to have cortical lysis, a cavitary component to the soft tissue mass and medium to large soft tissue mass size. The findings of this study can aid in prioritizing differential diagnoses in cases of extradural spinal neoplasia in cats and dogs, which can impact case management, but tissue sampling remains the gold standard for definitive diagnosis.

Nasal mucosal contact and caudal aberrant nasal turbinates were significantly more prevalent in brachycephalic dogs than in normocephalic dogs in our study. Computed tomography can be a valuable aid in obtaining data on nasal mucosal contact, caudal aberrant nasal turbinates and septal deviations. Combination of computed tomography with endoscopy and functional airway testing would be useful to further evaluate the correlation between intranasal features and symptoms of brachycephalic airway syndrome.

Ultrasonography and radiography are standard diagnostic tests for cats with suspected splenic disease, however published information on outside sources of variation are currently lacking. The purpose of this prospective, randomized, crossover group study was to evaluate effects of common sedative drugs on the sonographic and radiographic characteristics of the spleen in healthy cats. Fifteen healthy adult research cats were randomly assigned into one of three groups corresponding to different sequences of administration of five sedative drugs/drug combinations (acepromazine; butorphanol; dexmedetomidine; midazolam and butorphanol (MB); and dexmedetomidine, butorphanol, and ketamine (DBK)), administered at 1‐week intervals. At each visit, three‐view abdominal radiographic and ultrasonographic examinations were performed prior to sedation and repeated 15‐30 min and 2‐3 h post sedation. Two board‐certified radiologists (one ACVR and one ACVR/ECVDI) evaluated the anonymized and randomized images. Acepromazine resulted in significantly increased sonographic and radiographic splenic measurements from baseline, which remained significantly increased 2‐3 h post sedation. The mean magnitude of this change ranged from 0.9 mm (tail height, SD 1.4 mm) to 1.8 mm (body height, SD 1.7 mm) for ultrasound, and was 2.2 mm (ventrodorsal width, SD 2.3 mm) for radiographs. With butorphanol, there was no significant change in splenic size. For dexmedetomidine, MB, and DBK, there was a trend toward increased splenic size from baseline to the first post‐sedation timepoint, which was statistically significant for radiographic measurements, although not for ultrasound. Findings indicated that acepromazine should be avoided prior to imaging while butorphanol may be used when sedation is needed in cats presenting for potential splenic disease.

CASE DESCRIPTION A 4-year-old spayed female Beagle was evaluated because of a 2-month history of intermittent pollakiuria, stranguria, dysuria, and abdominal pain. A diagnosis of bacterial cystitis was initially made, but clinical signs persisted despite appropriate antimicrobial treatment, so the dog was referred for further evaluation and treatment. CLINICAL FINDINGS Abdominal ultrasonography revealed a large, thin-walled, cystic structure in the urinary bladder at the level of the expected right ureterovesicular junction that communicated with the uniformly dilated right ureter. Severe right-sided pyelectasia was also detected. A presumptive diagnosis was made of a right-sided orthotopic ureterocele with secondary hydroureter and pyelectasia. TREATMENT AND OUTCOME Cystoscopy revealed a large cystic structure in the region of the right ureterovesicular junction without obvious communication between the ureter and urinary bladder. Portable C-arm fluoroscopy was used to confirm the presence of an intramural orthotopic tract and to measure the diameter of the ureter and renal pelvis via retrograde contrast ureteropyelography. Complete laser ablation of the ureterocele was performed by incising it circumferentially near its base. Clinical signs resolved immediately following the procedure. Six weeks later, the dog remained free of clinical signs and abdominal ultrasonography revealed resolution of hydroureter with persistence of mild right-sided pyelectasia. CLINICAL RELEVANCE Cystoscopic-guided laser ablation of an orthotopic ureterocele secondary to ureterovesicular stenosis was a safe and effective minimally invasive treatment for the dog of this report, resulting in immediate and continued improvement of clinical signs and ultrasonographic changes. Laser ablation should be considered as an alternative to surgery for the treatment of orthotopic ureteroceles in dogs.

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