Computed tomographic evaluation of abdominal fat in minipigs

CT image segmentation of meat sheep Loin based on deep learning

There are no clear boundaries between internal tissues in sheep Computerized Tomography images, and it is difficult for traditional methods to meet the requirements of image segmentation in application. Deep learning has shown excellent performance in image analysis. In this context, we investigated the Loin CT image segmentation of sheep based on deep learning models. The Fully Convolutional Neural Network (FCN) and 5 different UNet models were applied in image segmentation on the data set of 1471 CT images including the Loin part from 25 Australian White rams and Dolper rams using the method of 5-fold cross validation. After 10 independent runs, different evaluation metrics were applied to assess the performances of the models. All models showed excellent results in terms evaluation metrics. There were slight differences among the results from the six models, and Attention-UNet outperformed others methods with 0.998±0.009 in accuracy, 4.391±0.338 in AVER_HD, 0.90±0.012 in MIOU and 0.95±0.007 in DICE, respectively, while the optimal value of LOSS was 0.029±0.018 from Channel-UNet, and the running time of ResNet34-UNet is the shortest.

Mesenteric fat cryolipolysis attenuates insulin resistance in the Ossabaw swine model of the metabolic syndrome

BACKGROUND

The rising prevalence of insulin resistance (IR), metabolic syndrome, and type 2 diabetes are associated with increases in abdominal mesenteric fat. Adipocytes are sensitive to low temperatures, making cryolipolysis of mesenteric fat an attractive treatment modality to potentially reduce IR.

OBJECTIVES

We aimed to determine whether (1) cryolipolysis is safe in reducing the volume of the mesenteric fat and (2) reduction in mesenteric fat volume reduces indices of IR and glycemic dysfunction.

SETTING

Indiana University School of Medicine.

METHODS

A novel cooling device and method delivered cryolipolysis in a controlled manner to avoid tissue ablative temperatures. Ossabaw pigs (n = 8) were fed a high-fat diet for 9 months to develop visceral obesity, IR, and metabolic syndrome. Following laparotomy, mesenteric fat cryolipolysis (MFC) was performed in 5 pigs, while 3 served as sham surgery controls. The volume of the mesenteric fat was measured by computed tomography and compared with indices of glucose intolerance before and at 3 and 6 months postprocedure.

RESULTS

MFC safely reduced mesenteric fat volume by ∼30% at 3 months, which was maintained at 6 months. Body weight did not change in either the MFC or sham surgery control groups. Measure of glycemic control, insulin sensitivity, and blood pressure significantly improved after MFC compared with sham controls.

CONCLUSION

MFC reduces the volume of mesenteric fat and improves glycemic control in obese, IR Ossabaw pigs, without adverse effects.

CT imaging features of fat stranding in cats and dogs with abdominal disorder

BACKGROUND

Fat stranding is a non-specific finding of an increased fat attenuation on computed tomography (CT) images. Fat stranding is used for detecting the underlying lesion in humans.

OBJECTIVES

To assess the clinical significance of fat stranding on CT images for identifying the underlying cause in dogs and cats.

METHODS

In this retrospective study, the incidence, location, extent, distribution, and pattern of fat stranding were assessed on CT studies obtained from 134 cases.

RESULTS

Fat stranding was found in 38% (51/134) of all cases and in 35% (37/107) of tumors, which was significantly higher in malignant tumors (44%) than benign tumors (12%). Moreover, fat stranding was found in more than two areas in malignant tumors (16/33) and in a single area in benign tumors (4/4). In inflammation, fat stranding was demonstrated in 54% (7/13) in a single area (7/7) as a focal distribution (6/7). In trauma, fat stranding was revealed in 50% (7/14) and most were in multiple areas (6/7). Regardless of the etiologies, fat stranding was always around the underlying lesion and a reticular pattern was the most common presentation. Logistic regression analysis revealed that multiple areas (p = 0.040) of fat stranding and a reticulonodular pattern (p = 0.022) are the significant predictors of malignant tumor.

CONCLUSIONS

These findings indicated that CT fat stranding can be used as a clue for identifying the underlying lesion and can be useful for narrowing the differential list based on the extent and pattern.

Non-invasive methods for the determination of body and carcass composition in livestock: dual-energy X-ray absorptiometry, computed tomography, magnetic resonance imaging and ultrasound: invited review

The ability to accurately measure body or carcass composition is important for performance testing, grading and finally selection or payment of meat-producing animals. Advances especially in non-invasive techniques are mainly based on the development of electronic and computer-driven methods in order to provide objective phenotypic data. The preference for a specific technique depends on the target animal species or carcass, combined with technical and practical aspects such as accuracy, reliability, cost, portability, speed, ease of use, safety and for in vivo measurements the need for fixation or sedation. The techniques rely on specific device-driven signals, which interact with tissues in the body or carcass at the atomic or molecular level, resulting in secondary or attenuated signals detected by the instruments and analyzed quantitatively. The electromagnetic signal produced by the instrument may originate from mechanical energy such as sound waves (ultrasound - US), 'photon' radiation (X-ray-computed tomography - CT, dual-energy X-ray absorptiometry - DXA) or radio frequency waves (magnetic resonance imaging - MRI). The signals detected by the corresponding instruments are processed to measure, for example, tissue depths, areas, volumes or distributions of fat, muscle (water, protein) and partly bone or bone mineral. Among the above techniques, CT is the most accurate one followed by MRI and DXA, whereas US can be used for all sizes of farm animal species even under field conditions. CT, MRI and US can provide volume data, whereas only DXA delivers immediate whole-body composition results without (2D) image manipulation. A combination of simple US and more expensive CT, MRI or DXA might be applied for farm animal selection programs in a stepwise approach.

Feasibility of MRI and selection of adequate region of interest for longitudinal studies of growth and fatness in swine models of obesity

PURPOSES

To determine the feasibility of MRI for imaging subcutaneous and visceral adiposity in longitudinal studies in obese swine models (Iberian pig). To establish the anatomical regions of interest (ROIs) and measurement points (MPs) adequate for their evaluation through analyses on the inter-individual variability and over-time reproducibility and through the assessment of their reliability and validity by comparison with in vivo and ex vivo zoometric data.

MATERIAL AND METHODS

Five male and five female pigs were used from four (live weight around 48 kg and back-fat depth around 18 mm) to eight months old (live weight 134 kg and back-fat depth around 48 mm). MRI was carried out with a Panorama 0.23T scanner (Philips Medical Systems, Best, Netherlands), using a body/spine XL coil.

RESULTS

The ROIs of election for visualization of subcutaneous data are located from the cranial margin of left diaphragmatic crura to the lumbar vertebrae L3. Visceral adiposity may be equally evaluated from the vertebrae L1 to L3.

CONCLUSIONS

MRI allows the evaluation of subcutaneous and visceral fatness in a single acquisition, which improves animal welfare and time- and cost-efficiency and provides an accurate, consistent and repeatable procedure for sequential studies of adiposity in obese swine.

Maternal malnutrition and offspring sex determine juvenile obesity and metabolic disorders in a swine model of leptin resistance

The present study aimed to determine, in a swine model of leptin resistance, the effects of type and timing of maternal malnutrition on growth patterns, adiposity and metabolic features of the progeny when exposed to an obesogenic diet during their juvenile development and possible concomitant effects of the offspring sex. Thus, four groups were considered. A CONTROL group involved pigs born from sows fed with a diet fulfilling their daily maintenance requirements for pregnancy. The treated groups involved the progeny of females fed with the same diet but fulfilling either 160% or 50% of pregnancy requirements during the entire gestation (OVERFED and UNDERFED, respectively) or 100% of requirements until Day 35 of pregnancy and 50% of such amount from Day 36 onwards (LATE-UNDERFED). OVERFED and UNDERFED offspring were more prone to higher corpulence and fat deposition from early postnatal stages, during breast-feeding; adiposity increased significantly when exposed to obesogenic diets, especially in females. The effects of sex were even more remarkable in LATE-UNDERFED offspring, which had similar corpulence to CONTROL piglets; however, females showed a clear predisposition to obesity. Furthermore, the three groups of pigs with maternal malnutrition showed evidences of metabolic syndrome and, in the case of individuals born from OVERFED sows, even of insulin resistance and the prodrome of type-2 diabetes. These findings support the main role of early nutritional programming in the current rise of obesity and associated diseases in ethnics with leptin resistance.

Whole Body Computed Tomography with Advanced Imaging Techniques: A Research Tool for Measuring Body Composition in Dogs

The use of computed tomography (CT) to evaluate obesity in canines is limited. Traditional CT image analysis is cumbersome and uses prediction equations that require manual calculations. In order to overcome this, our study investigated the use of advanced image analysis software programs to determine body composition in dogs with an application to canine obesity research. Beagles and greyhounds were chosen for their differences in morphology and propensity to obesity. Whole body CT scans with regular intervals were performed on six beagles and six greyhounds that were subjected to a 28-day weight-gain protocol. The CT images obtained at days 0 and 28 were analyzed using software programs OsiriX, ImageJ, and AutoCAT. The CT scanning technique was able to differentiate bone, lean, and fat tissue in dogs and proved sensitive enough to detect increases in both lean and fat during weight gain over a short period. A significant difference in lean : fat ratio was observed between the two breeds on both days 0 and 28 (P < 0.01). Therefore, CT and advanced image analysis proved useful in the current study for the estimation of body composition in dogs and has the potential to be used in canine obesity research.