A multicenter retrospective study compared ultrasonographic (US) and computed tomographic (CT) characteristics of parathyroid glands in 20 dogs with histopathological confirmation of primary hyperparathyroidism. Twenty-four glands were diseased (17 adenomas and 7 hyperplastic glands). Two blinded radiologists assessed US and CT studies in two readings. A similar total number of glands was visualized (US 59 and CT 63). Intermodality agreement for identification of the same individual glands was slight for presumed normal glands but perfect for diseased glands, with 23 of 24 diseased glands visualized on both modalities. Glands were more often poorly marginated on CT compared to US, whereas other descriptive features were similar across modalities. The only shared feature of diseased glands on both modalities was significantly greater size. Intermodality agreement for size measurements was poor for presumed normal glands, with CT measuring larger, but moderate to good for diseased glands. Intermodality agreement for classifying glands as enlarged (largest dimension ≥4 mm) was substantial. Enlargement-based localization sensitivity for diseased glands was similar between modalities (US 18/24, 75.0%; CT 19/24, 79.2%). Additional enlarged glands without histopathological confirmation of disease were identified in three instances on US and seven on CT. Interobserver and intraobserver agreement for quantitative parameters was similar between modalities; however, agreement for most descriptive parameters was superior on US. Overall, US likely remains the first-line imaging modality for localizing parathyroid disease; however, these findings support selective incorporation of CT to aid therapeutic decision-making in canine primary hyperparathyroidism.
Ultrasonographic kidney size assessment requires normalization to body size; the kidney length-to-aorta diameter (KL/AoD) ratio improves diagnostic accuracy, yet reference ranges vary across breeds. This study aimed to establish KL/AoD reference values in French Bulldogs and assess the effects of sex, age, body weight, kidney side, and aortic scanning plane. Fifty healthy French Bulldogs were examined. KLs were measured in the sagittal plane, and aortic diameters were recorded just caudal to the left renal artery in longitudinal (AoDL) and transverse (AoDT) planes. Each measurement was taken thrice by a single operator. Intraclass correlation coefficients (ICCs) assessed reproducibility; statistical analyses evaluated potential influencing factors (p < 0.05). ICC values ranged from 0.88 to 0.96, indicating excellent reliability. Mean LKL was 5.06 ± 0.44 cm, and RKL was 5.05 ± 0.44 cm. Mean AoDL was 0.70 ± 0.08 cm, significantly different from AoDT (0.75 ± 0.10 cm). No significant side difference was observed (p > 0.61); data from both kidneys were pooled. Sex significantly influenced all parameters (p < 0.001). Reference intervals, established using the American Society for Veterinary Clinical Pathology (ASVCP) robust method, were KL/AoDL 6.30-8.33 (males: 6.43-8.38; females: 5.97-7.91) and KL/AoDT 5.59-7.92 (males: 5.75-7.97; females: 4.96-7.65). Breed-specific reference intervals were established using the ASVCP robust method, with sex-specific values for KL/AoDL (males: 6.43-8.38; females: 5.97-7.91) and KL/AoDT (males: 5.75-7.97; females: 4.96-7.65), emphasizing the need to account for breed, sex, and aortic scanning plane in ultrasonographic assessment.
A 12-year-old male castrated Shih Tzu presented with a suspected intracranial granular cell tumor (GCT) in the left cerebrum. The clinical signs had resolved post stereotactic radiotherapy (SRT) of 3 consecutive days of 8 Gy, and serial magnetic resonance imaging (MRI) revealed maximum 56.8% tumor volume reduction 9 months post SRT. Serial MRI showed progressive cerebral microbleeds. The patient neurologically declined 18 months after SRT and was euthanized. A full necropsy, cytologic, and histologic analysis confirmed the diagnosis of GCT. This is the first reported canine case with cytologically and histologically confirmed intracranial GCT treated with SRT with serial MRI at regular intervals following treatment.
Ultrasound assessment of the pancreas in dogs is limited by operator subjectivity and technical variability, particularly when classifying echotexture and echogenicity. This prospective study evaluated quantitative ultrasonographic features of the pancreas in dogs and examined their associations with subjective ultrasonographic features, age, body composition, and technical factors. Seventy-two ultrasounds from client-owned dogs undergoing abdominal imaging for diverse clinical indications between 2016 and 2019 were included. Subjective echogenicity and echotexture were assessed, pancreatic and abdominal adipose thicknesses measured, and quantitative intensity and texture features extracted from regions of interest using image-analysis software. A subset of cases was evaluated by 104 veterinary radiologists in a parallel study to generate consensus classifications used for validation. Multivariable models were applied to assess associations between pancreatic features and age, body weight, adiposity, fat distribution, and technical variables. Mean pancreatic thickness was 10.8 mm (SD 3.1) and was positively associated with body weight. Pancreatic hyperechogenicity was uncommon (8.3%). Quantitative pancreatic intensity was associated with age. Heterogeneous echotexture was common (43%) and, on subjective assessment, was associated with older age and lower transducer frequency. Quantitative texture features were associated with age, pancreatic thickness, pancreas orientation, and transducer frequency. These findings suggest that pancreatic heterogeneity is frequent and age-associated, and that both subjective and quantitative assessments are influenced by technical factors.
Advancements in radiation therapy allow for targeting smaller tumor volumes with exceedingly conformal precision. Target motion during the breathing cycle needs to be anticipated. To date, no information exists that characterizes the motion of canine thoracic tumors during the respiratory cycle. The primary purpose of this study is to assess the pattern of motion of thoracic tumors in three-dimensional (3D) space between inspiratory and expiratory breath holds as a surrogate measurement of the breathing cycle. A secondary aim is to evaluate whether tumor position within the thorax relates to the degree of motion. Fourteen dogs were prospectively enrolled and received thoracic computed tomography (CT) scans with pre- and post-contrast inspiratory and expiratory phases achieved via manual breath holds. Gross tumor volume (GTV) and normal structures were delineated on inspiratory and expiratory image sets, and shifts were measured and analyzed. Smaller tumors had more motion in the dorsoventral direction compared to larger tumors (p = 0.047; rs = -0.44). The same observation was appreciated in the craniocaudal direction (p = 0.008; rs = -0.56). Dogs with multiple tumors had more motion in the dorsoventral direction and the craniocaudal direction compared to dogs with solitary tumors (p = 0.035; p = 0.022). Caudal tumors had significantly more 3D motion (p < 0.001) and craniocaudal motion (p < 0.001) compared to cranial tumors, whereas there was significantly more lateral motion (p < 0.001) between peripheral and central tumors, but no significant difference in 3D motion (p = 0.24). The variability of tumor motion between canine patients supports individualized assessments of each case to ensure best practice.
Pneumonia remains a leading cause of morbidity and mortality in companion animals, with delayed diagnosis significantly impacting treatment outcomes. Current veterinary radiographic analysis primarily relies on manual interpretation, which suffers from substantial inter-rater variability and limited sensitivity for early-stage pathologies. Accordingly, we introduced AI-driven approaches into pet health monitoring, notably outlining a step-by-step diagnostic pipeline to address those above challenges. Key components of the diagnostic pipeline include six parts as follows: data acquisition, data preprocessing, semantic segmentation, classification, risk estimation, and explainability by Gradient-weighted Class Activation Mapping (GradCAM). Utilizing PaddleSeg for semantic segmentation, we accurately localized and extracted the chest area from pet x-ray images, achieving a mean intersection over union (mIoU) of 0.8999. Following this, classification was conducted via the Mamba framework. The model demonstrated robust performance across both frontal (VD/DV) and lateral projections. For frontal (VD/DV) projections, it achieved an accuracy of 90.35%, with a macro-averaged F1-score of 0.9034 (negative: 90.35%; pneumonia: 90.33%) and an area under the ROC curve (AUC) of 0.9438. Performance on lateral projections was equally robust, yielding an accuracy of 89.8%, a macro-averaged F1-score of 0.8965 (negative: 89.64%; pneumonia: 89.66%), and an AUC of 0.9361. Finally, we proposed a novel risk-aware AI module tailored for pet radiographic disease classification, integrating multidimensional uncertainty and interpretability cues into a unified AI Risk Index (ARI), which represents a significant step toward deploying interpretable, safe AI systems in veterinary medicine and sets a precedent for other safety-critical domains.
Asian elephants possess a unique pulmonary anatomy that presents challenges for diagnosing respiratory diseases, particularly when reared under human care where environmental stressors, poor hygiene and limited physical activity predispose them to respiratory pathologies. Conventional diagnostic tools such as radiography, computed tomography (CT), and magnetic resonance imaging (MRI) are impractical for elephants due to their large body size. Diagnosis of lung pathologies is often delayed until postmortem. This study evaluated the diagnostic utility of lung ultrasound (LUS) as a noninvasive, real-time imaging modality in assessing lung health in Asian elephants. A total of 68 noncardiac transthoracic ultrasound scans were performed on 28 elephants to develop a systematic approach for pulmonary imaging. LUS demonstrated high sensitivity in detecting lung lesions, including interstitial syndrome, pulmonary edema, consolidation, pneumothorax, and pleural effusion, even in apparently healthy, asymptomatic individuals. Specific thoracic zones serving as acoustic windows were successfully mapped to improve consistency in lesion localization. Additionally, a semi-quantitative LUS score was developed for the objective assessment of lung aeration across 24 thoracic regions. Despite species-specific challenges such as narrow intercostal spaces and large lung volume, key sonographic features, including A-lines, B-lines, lung sliding, and bronchograms, were visualized. This study establishes LUS as a practical, field-adaptable tool for antemortem detection and monitoring of respiratory disease in elephants. Its ability to detect subclinical pathology offers significant benefits for early intervention, improved clinical outcomes, and support for conservation efforts. LUS has the potential to transform pulmonary diagnostics in elephant healthcare, bridging a critical gap in veterinary medicine for this endangered species.
Abdominal ultrasonography (AUS) is considered an important diagnostic tool for assessing pancreatic disease in dogs. However, AUS interpretation is highly subjective and operator-dependent, with no prior studies evaluating observer variability in assessing the canine pancreas. This survey-based, observer agreement study aimed to assess inter- and intra-observer variability in the ultrasonographic evaluation of canine pancreases. Fifteen ultrasound images of dog pancreases were independently reviewed by 104 veterinary radiologists and radiology residents. Twenty-four veterinary radiologists repeated the review approximately 2.5 months later. Assessments included relative echogenicity, echotexture, margination, and clinical interpretation. Agreement was quantified using Krippendorff's alpha (Kα), prevalence-adjusted bias-adjusted kappa (PABAK), Cohen's kappa (κ), and intraclass correlation coefficient (ICC). Average inter-observer agreement was poor (Kα = 0.34), whereas intra-observer agreement was moderate (PABAK = 0.49). Global pancreatic echogenicity and echogenicity relative to mesenteric fat had the highest inter-observer agreement (Kα = 0.58 and 0.53, respectively) as did echotexture assessed by a semi-objective scale (Kα = 0.50). However, veterinary radiologists demonstrated poor inter-observer agreement on the clinical significance (Kα = 0.35) and prioritized cause (Kα = 0.14) of the pancreatic heterogeneity. Observer variables (age, training level, experience, primary practice, and accrediting organization) had minimal effect on inter-observer agreement, apart from experience, where early-career radiologists (<10 years' experience) demonstrated the highest inter-observer agreement among all groups, with moderate agreement for global echogenicity (Kα = 0.71). These findings highlight substantial variability in the ultrasonographic interpretation of the canine pancreas, particularly in clinical decision-making. Standardization of assessment criteria is necessary to improve consistency, clinical validity, and reduce diagnostic uncertainty.
A 1.5-year-old English Springer Spaniel presented with a firm, non-painful, dorsal nasal swelling of gradual increasing size. Rostral to the swelling, a small round pit at the mucocutaneous junction of the caudodorsal midline nasal planum was noted. Computed tomography (CT) identified a focal, midline to left-sided swelling with a central cavity and a thin, linear, midline tube running in the subcutaneous space to the small pit at the mucocutaneous junction. The swelling was positioned directly dorsal to and protruding into a nasal suture widening. Surgical excision was performed. Histopathology confirmed a nasal dermoid sinus. Surgical excision was curative.
In dogs with chylothorax that are refractory to medical management, surgical intervention is often recommended. Computed tomographic lymphangiography (CTL) is the most commonly used preoperative imaging study, but it has the disadvantages of being time-consuming and requiring contrast medium administration. In human medicine, magnetic resonance imaging (MRI) allows noninvasive visualization of lymphatic ducts. The aim of this prospective study was to compare the use of a T2-W 3D fat-suppressed MRI sequence to traditional CTL for identifying the number and anatomic location of thoracic duct branches. MRI and CTL studies were performed in six healthy dogs. Images were anonymized and independently reviewed by a board-certified veterinary surgeon and a board-certified veterinary radiologist. For each study, the number of duct branches was recorded at four sites (T9, T11, T13, and L2) in addition to the location of the cisterna chyli. Additionally, the surgeon's proposed surgical site and duration of MRI and CTL procedures were noted. There was no significant difference in the number of branches identified between MRI and CTL (p = 0.751). Each dog had at least one location with only one observed thoracic duct branch on both CT and MRI, and this was chosen as the proposed surgical site for the dog. There was no statistical difference in study duration between CTL and MRI (p = 0.294). MRI may prove to be a beneficial and efficient alternative to CTL for surgical planning in dogs with chylothorax.
A 12-year-old male neutered domestic shorthair presented for a 3-month history of pollakiuria, stranguria, tenesmus, and dyschezia. Computed tomography (CT) identified an intrapelvic mass. Midline celiotomy revealed a cystic-like mass at the dorsal aspect of the intrapelvic urethra. Debulking and omentalization of the mass were performed. Histopathological examination was consistent with a phyllodes tumor of the prostate, a novel tumor in cats. At the 7-month follow-up, the patient was doing well, with improved pollakiuria and constipation. Unfortunately, local progression and possible pulmonary metastasis were identified on repeat CT.
In female animals, the functional status of the ovary is closely associated with ovarian hemodynamics. Contrast-enhanced ultrasonography (CEUS) is a noninvasive imaging technology that allows for the visualization of microvascular blood flow. However, there is a lack of research on the hemodynamics of rat ovaries. This study conducted CEUS dynamic monitoring of ovarian hemodynamics in rats during infancy and pregnancy, revealing significant variations in time-intensity curve (TIC) parameters between the two periods. The findings demonstrate the ability of CEUS to quantitatively assess the hemodynamic characteristics of rat ovaries, indirectly indicating their functional status. Moreover, these results offer a technical method for the noninvasive monitoring of ovarian functional diseases by CEUS.
A 10-year-old spayed female Korean shorthair cat was presented with hematemesis and lethargy. Marked volume of free peritoneal gas was identified on abdominal radiographs. Bile pigment was identified in peritoneal fluid obtained by abdominocentesis, consistent with biliary peritonitis. Ultrasound examination was limited by pneumoperitoneum and severe peritonitis, making it difficult to differentiate gastrointestinal perforation from biliary system rupture. Two defects in the proximal duodenum near the major duodenal papilla (MDP) were identified with a non-anesthetic computed tomography (CT) scan. Exploratory laparotomy confirmed two perforations and revealed severe biliary peritonitis. Despite surgical repair and intensive treatment, the cat died 3 days later. This case highlights the importance of considering duodenal perforation near the MDP as a differential diagnosis when pneumoperitoneum and biliary peritonitis are present.
This report describes a 25-year-old, intact male brown bear (Ursus arctos) presenting with weight loss, diarrhea, melena, apathy, and a 12% hematocrit. Ultrasound showed proximal duodenal wall thickening with submucosal cysts and a large, heterogeneous, highly vascularized mass in the descending duodenum, containing small gas pockets and compressing the lumen. The pancreas was visualized at the beginning of the descending duodenum, with normal echogenicity and slightly heterogeneous echotexture, containing a large cyst. Upper gastrointestinal endoscopy revealed bleeding and a partially obstructive duodenal mass with abundant fresh blood. A few days later, the animal died from hemorrhage as there were no possibilities for transfusion and consequently surgery. Histopathology revealed well-differentiated exocrine pancreatic adenocarcinoma infiltrating the duodenum. According to the scarce literature found, the ultrasonographic and endoscopic appearance of pancreatic adenocarcinoma in a brown bear has never been described.
A 7-month-old male neutered Labrador Retriever was referred for vomiting, inappetence, and lethargy. Baseline bloodwork revealed mild mature neutrophilia, elevated amylase and lipase, and mild hypochloremia. Abdominal ultrasound showed a cavitated left cranial abdominal mass with a "whirl sign," indicating a torsed organ. Surgical resection and histopathology confirmed pancreatic torsion, a rare disease in young dogs. The patient recovered uneventfully and was discharged the following day. This is the first report describing a "whirl sign" on ultrasound associated with pancreatic torsion in a dog.
This study assessed the feasibility of contrast-enhanced ultrasound (CEUS) for evaluating renal perfusion in horses and compared perfusion parameters between age groups. In a prospective study, eight healthy horses were divided into young (3-5 years, n = 4) and old (16-17 years, n = 4) groups. All animals underwent B-mode, Doppler ultrasonography, and CEUS of the right kidney after an intravenous bolus of SonoVue (0.02 mL/kg). Qualitative analysis assessed contrast kinetics, whereas quantitative parameters (peak intensity, time to peak [TP], area under the curve, and wash-out time) were obtained from time-intensity curves in the renal cortex and medulla. The technique was safely performed in all horses, and the 0.02 mL/kg dose proved optimal for achieving adequate renal enhancement and consistent quantitative measurements. Quantitative analysis revealed a significantly longer TP in the renal cortex of older horses compared to young ones (p = 0.03), suggesting an age-related reduction in perfusion velocity. No other parameters showed significant differences. CEUS proved to be a feasible and safe method for assessing renal perfusion in horses. The prolonged cortical TP in older animals indicates that CEUS may be a sensitive tool for detecting early, subclinical perfusion changes associated with aging, warranting further investigation in horses with renal disease.
The application of deep learning-based reconstruction (DLR) to magnetic resonance imaging (MRI) has been recently introduced in human and veterinary medicine to improve image quality without prolonging acquisition time. We hypothesized that cranial abdominal MRI with DLR would have superior image quality than conventional MRI. This prospective comparative pilot study aimed to compare cranial abdominal MR image quality with versus without DLR application in dogs. Transverse T2-weighted and T1-weighted images with and without contrast enhancement of the cranial abdomen region were performed in 10 clinically healthy dogs. The original and DLR images were generated from a single set of raw data using the DLR network embedded reconstruction pathway. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated for quantitative analysis. Qualitative analysis included evaluation of the edge sharpness of the liver and pancreas, the conspicuity of the adrenal glands, respiratory motion artifacts, coarseness, and overall image quality using a 4-point scale. Quantitative and qualitative values were compared between images with DLR and their counterparts. The mean SNR and CNR of all images with DLR were significantly higher than those of their counterparts (p < 0.05). Qualitative analysis parameters, except for respiratory motion artifacts, were significantly superior in images with DLR compared with counterparts for all sequences (p < 0.05). The present study demonstrated that cranial abdominal MRI with DLR is feasible to improve image quality without prolonging acquisition time.
A 4-year-old domestic shorthair cat from Switzerland presented with seizures and peripheral lymphadenomegaly. Thoracic radiographs demonstrated cranioventral pulmonary masses and enlargement of the cranial mediastinal and tracheobronchial lymph nodes. Abdominal ultrasonography revealed generalized lymphadenopathy characterized by rounded, homogeneously hypoechoic lymph nodes with an effaced hilus. Imaging differentials included lymphoma, feline coronavirus-associated granulomatous disease, and other fungal infections; cytology and culture confirmed mucormycosis. The pulmonary masses and lymphadenomegaly initially responded to azole therapy; however, they later progressed despite further treatment, and the cat was humanely euthanized 8 months after diagnosis. This case highlights an atypical presentation of mucormycosis in an immunocompetent adult cat, with an unusual lesion distribution localized within the lymphatic system, respiratory system, and suspected central nervous system involvement.
This retrospective study evaluated a novel hypofractionated radiation therapy (RT) protocol for canine urothelial carcinoma (UC). Medical records of dogs diagnosed with UC and treated were retrospectively reviewed. Inclusion criteria included diagnosis of UC, treatment with hypofractionated RT, absence of prostatic involvement or metastatic disease, and adequate follow-up. The RT protocol consisted of 25 Gy delivered to the planning target volume (PTV) in five fractions on an every-other-day schedule. Dynamic adaptive radiation therapy (DART) was used in select bladder tumors to minimize geographic miss. Data on adverse events, clinical response, adjuvant therapies, and prognostic factors were evaluated in relation to progression-free survival (PFS) and overall survival (OS). Eighteen dogs met the inclusion criteria. All received nonsteroidal anti-inflammatory drugs (NSAIDs), and 67% also received chemotherapy after RT. Tumor location included bladder only (55%), urethra only (28%), or both (17%). RT was well tolerated, with only low-grade gastrointestinal (17%) or urinary (6%) adverse events observed. Measurable tumor response was observed in 82% (14/17) with follow-up imaging. Median PFS was 394 days (95% CI: 169-619), and median OS was 565 days (95% CI: 497-632). Dogs with urethral involvement demonstrated a shorter PFS compared to those without (median = 182 vs. 457 days, respectively [p = 0.02]). Chemotherapy did not impact PFS or OS. The novel hypofractionated RT protocol was well tolerated alone and in combination with other treatments. Tumor responses and survival times were improved compared to historical data on dogs receiving NSAIDs and/or chemotherapy alone.
Radiographic and computed tomographic (CT) visibility of human medications has been widely studied; however, equivalent data for veterinary medications remain limited. This study evaluated the radiographic visibility, dissolution characteristics, and CT features of commonly used veterinary and some human medications using a standardized water model. Seventy-five medications were assessed for radiopacity prior to submersion, immediately after submersion, after 30 min, and after 1 h. All medications were visible without water. More than half (54.6%) remained identifiable immediately after submersion, decreasing to 13.3% after 30 min and to 6.6% after 1 h. The gastrointestinal complementary feed, vitamin B12, potassium chloride, and carprofen retained visibility throughout the experiment. Tablet area and height varied widely and demonstrated a weak positive correlation with visibility only after 60 min of submersion. Higher Hounsfield units (HU) values were moderately correlated with visibility at baseline, with the strength of this association decreasing over time as dissolution progressed. This study suggests radiography may be of limited use for the identification of ingested medication and provides images and HU values as a reference for clinicians.
