Automatic classification and grading of canine tracheal collapse on thoracic radiographs by using deep learning.

Tracheal collapse is a chronic and progressively worsening disease; the severity of clinical symptoms experienced by affected individuals depends on the degree of airway collapse. Cutting-edge automated tools are necessary to modernize disease screening using radiographs across various veterinary settings, such as animal clinics and hospitals. This is primarily due to the inherent challenges associated with interpreting uncertainties among veterinarians. In this study, an artificial intelligence model was developed to screen canine tracheal collapse using archived lateral cervicothoracic radiographs. This model can differentiate between a normal and collapsed trachea, ranging from early to severe degrees. The you-only-look-once (YOLO) models, including YOLO v3, YOLO v4, and YOLO v4 tiny, were used to train and test data sets under the in-house XXX platform. The results showed that the YOLO v4 tiny-416 model had satisfactory performance in screening among the normal trachea, grade 1-2 tracheal collapse, and grade 3-4 tracheal collapse with 98.30% sensitivity, 99.20% specificity, and 98.90% accuracy. The area under the curve of the precision-recall curve was >0.8, which demonstrated high diagnostic accuracy. The intraobserver agreement between deep learning and radiologists was κ = 0.975 (P < .001), with all observers having excellent agreement (κ = 1.00, P < .001). The intraclass correlation coefficient between observers was >0.90, which represented excellent consistency. Therefore, the deep learning model can be a useful and reliable method for effective screening and classification of the degree of tracheal collapse based on routine lateral cervicothoracic radiographs.

Hematological and blood biochemistry parameters as prognostic indicators of survival in canine multicentric lymphoma treated with COP and L-COP protocols.

Background and Aim: Hematological and blood chemistry parameters are crucial for evaluating and monitoring canine multicentric lymphoma during chemotherapy. Pre-treatment hematological and blood chemistry parameters can be used as prognostic survival outcomes for this disease. Therefore, this study aimed to investigate the effect of hematological and blood chemistry parameters pre-treatment and 4 weeks post-treatment on the survival outcomes of dogs treated with either a combination of cyclophosphamide, vincristine, and prednisolone (COP) or a combination of COP with L-asparaginase (L-COP) protocols. Materials and Methods: We conducted a retrospective study. Medical records and hematological and blood chemistry parameters of 41 dogs with multicentric lymphoma treated with L-COP (n = 26) and the COP protocols (n = 15) were obtained from the hospital information system. Most cases were classified as high-grade lymphoma based on the Kiel cytological classification. The effects of hematological and blood chemistry parameters on survival outcomes were investigated using the Cox proportional hazard regression model. The median survival time (MST) for each hematological and blood chemistry parameter affecting survival outcome was established and compared using the Kaplan-Meier product limit method with the log-rank test. Results: Dogs with high-grade multicentric lymphoma that were treated with the COP protocol and had monocytosis at pre-treatment had a significantly shorter MST than dogs with normal monocyte counts (p = 0.033). In addition, dogs with azotemia, both pre-treatment and 4 weeks post-treatment, had a significantly shorter MST than dogs with normal serum creatinine levels (p = 0.012). Dogs with high-grade multicentric lymphoma treated with the L-COP protocol who had hypoalbuminemia (serum albumin concentration <2.5 mg/dL) at both pre-treatment and 4 weeks post-treatment had a significantly shorter MST than dogs with normal serum albumin levels (p < 0.001). Furthermore, dogs with leukocytosis at 4 weeks post-treatment had a significantly shorter MST than those with a normal total white blood cell count (p = 0.024). Conclusion: Serum albumin level can serve as a simple negative prognostic indicator of survival outcomes in dogs with high-grade multicentric lymphoma treated with the L-COP protocol. Dogs with hypoalbuminemia pre-treatment and 4 weeks post-treatment tended to have a shorter MST than those with normal serum albumin concentrations.

The Differences in Radiographic Vertebral Size in Dogs with Different Chest and Skull Types.

The objective of this study was to elucidate the differences in vertebral length, vertebral height, and vertebral length/height ratio of the fourth thoracic vertebra (T4), the second lumbar vertebra (L2), the fifth lumbar vertebra (L5) and the seventh lumbar vertebra (L7) based on radiographs in dogs with various body sizes, skull types, and thoracic conformations and to determine the relationships of these parameters with age and sex. A total of 258 dogs were included in this study and classified by three criteria-BW (Criterion 1), skull type (Criterion 2), and thoracic conformation (Criterion 3). Age had weak negative correlations with vertebral length and height. Sex did not affect the vertebral size parameters. BW had strong positive correlations with vertebral length and height, but there was no influence of BW on vertebral length/height ratio. Regarding the different body sizes and conformations, large breeds had vertebrae with significantly greater length and height than small and medium breeds (p < 0.001). In Criterion 2, the vertebrae of the mesocephalic dogs had significantly greater length and height than those of the brachycephalic and dolichocephalic dogs (p < 0.05). In Criterion 3, both deep-chest and round-chest dogs had vertebrae with significantly greater length and height than the barrel-chest dogs (p < 0.0001). Only vertebral length/height ratios of T4 were not influenced by age, sex, BW, skull type, and thoracic conformation. Age, differences in body size, skull type, and thoracic conformation could affect the vertebral size in dogs. Therefore, using breed-specific vertebral lengths and/or heights is a better approach for comparative radiographic analysis with vertebral measurements.

Computed tomographic diaphragmatic thickness: a promising method for the evaluation of diaphragmatic muscle in cardiopulmonary diseased cats.

Diaphragmatic dysfunction (DD) is defined as a weakening of the diaphragmatic muscle and can be an undetected cause of dyspnea. The objectives of this study were to explore the appropriate diaphragmatic location, measure diaphragmatic thickness (DT), evaluate the effect of intrinsic factors on DT, and compare DT between healthy and diseased cats, using 33 healthy cats and 15 diseased cats. A retrospective, analytical, case-control study using thoraco-abdominal feline computed tomography (CT) was performed. Two radiologists independently reviewed all images to verify inter- and intra-observer reliabilities and the best position for measuring DT. The effects of sex, age, and body weight were also studied, and cutoff values for detecting DT abnormalities were established. The results showed that the appropriate location for DT measurement was at the ventral border of the cranial endplate of the first lumbar vertebral body (L1) due to its highest intra- and inter-observer reliabilities. At this location, a significant difference in DT between the right and left hemidiaphragms (p = 0.01) was observed. Only sex had an impact on DT values. Interestingly, the DTs of cardiorespiratory-affected cats, both on the right and left sides, were significantly thinner than those of healthy cats. In conclusion, CT imaging is a reliable imaging method for determining diaphragmatic muscular atrophy. The ventral border of the cranial endplate of L1 is recommended for measuring the DT, and sex was the only factor affecting the DT measurement.