Skip to main content

Gabi Seiler

Assoc Professor of Radiology

Diplomate, European College of Veterinary Diagnostic Imaging

Diplomate, American College of Veterinary Radiology

Terry Companion Vet Med Center 2577


Dr. Gabriela Seiler received her Doctor of Veterinary Medicine degree from the University of Bern, Switzerland and worked in small animal practice in Switzerland for 2 years while performing a doctoral thesis before entering a radiology residency program at the University of Bern.  She became board certified in radiology by the European College of Veterinary Diagnostic Imaging in 2001. After two years as a Lecturer in Radiology she became an Assistant Professor of Radiology at the University of Pennsylvania in 2004 and completed board certification by the American College of Veterinary Radiology in 2006. She joined the faculty at North Carolina State University in 2009 as Associate Professor (tenure track) with promotion to Professor in 2017.


Certificate in Clinical Research

Area(s) of Expertise

The main research focus is in the use of advanced ultrasound imaging methods
in veterinary and translational medicine with a special interest in tumor perfusion. Information about blood supply of a tumor is important for characterization of tumor type and selection of optimal treatment methods and response to treatment. Cancer research in dogs and cats is relevant to human patients as well as the spontaneous nature and the size of the patient closely resembles cancer in people. One method to investigate tumor perfusion non-invasively is contrast-enhanced ultrasound. Ultrasound contrast media are stabilized gas-filled microbubbles that are small enough to pass through capillaries. Their highly elastic shells are compressed or resonated when activated by an ultrasound beam, and they consequently produce a strong signal which can be registered by the ultrasound transducer. Vasculature and tissue perfusion of a tumor can thus be assessed repeatedly and non-invasively. Contrast-enhanced ultrasound has already been established as a safe imaging method in dogs and cats for assessment of various organs. It has been shown to accurately distinguish benign from malignant liver nodules in dogs. Furthermore, ultrasound contrast media can be targeted to specific vascular receptors or used as a carrier for substances used for chemotherapy or gene therapy.

Characterization of Early and Late Stage Feline CKD Using Contrast-Enhanced Ultrasound and Ultrasound Elastography

Ultrasound Molecular Imaging to Assess Therapeutic Response


View all publications 


Date: 10/01/19 - 9/30/23
Amount: $220,333.00
Funding Agencies: American Kennel Club Canine Health Foundation, Inc.

Gallbladder mucocele formation is a disorder of the gallbladder where excess secretion of abnormally thick mucus results in obstruction to the normal flow of bile or rupture of the gallbladder contents into the abdominal cavity. The disease has a strong breed predisposition for Shetland sheepdogs, Border terriers, Cocker spaniels, Schnauzers, Pomeranians, Bichon frises, Chihuahuas, Pugs, and Beagles and uncommonly affects non-purebred dogs. The only curative treatment for the disease is a costly surgery to remove the gallbladder and an average of 27% of dog’s die of post-operative complications. Our long-term goal is to identify the underlying cause of mucocele formation in dogs and to formulate a rational treatment strategy to prevent or medically resolve them. The overall objective of this proposal is to establish if the course of mucocele formation can be reversed by correcting noteworthy metabolic disturbances that we have painstakingly documented in these dogs. Our strong published data, generated by support from the AKC Canine Health Foundation (Grant #01986), demonstrated that dogs with mucocele formation are relatively deficient in essential dietary compounds that play key roles in normal metabolism and/or whose biological activity may help to promote normal secretory function to the gallbladder epithelium. These studies propose to determine if supplementation with these dietary factors will arrest and reverse the course of mucocele formation in dogs. Success of these studies will have an important positive impact by providing an immediate benefit to dogs with this disease even if we have yet to identify the underlying cause.

Date: 06/01/18 - 5/31/23
Amount: $235,822.00
Funding Agencies: National Institutes of Health (NIH)

Cancer affects 39.6% of Americans at some point during their lifetime. Solid tumor microenvironments are characterized by a disorganized, leaky vasculature that promotes regions of low oxygenation. In fact, tumor hypoxia is a key predictor of poor treatment outcome for all radiotherapy, chemotherapy and surgery procedures, as well as a hallmark of metastatic potential. In particular, tumor cell resistance to radiotherapy is 3 fold increased in anoxic cells and even very small tumors comprise 10-30% of hypoxic regions in the form of chronic and/or transient hypoxia fluctuating over course of seconds to days. Recently, lipid-stabilized oxygen microbubbles (OMBs) have been used in vivo to relieve tumor hypoxia in sonodynamic therapy when injected directly in the tumors, as well as shown to sustain asphyxiated animals for over two hours when injected intra-peritoneally. Our preliminary data supports our hypothesis that oxygen microbubbles could also be used to relieve tumor hypoxia during radiotherapy and significantly improve treatment outcome. In addition, there has been no systemic OMB delivery demonstration to date for tumor hypoxia modulation with OMBs, due in part to the difficulty of measuring hypoxia in vivo reliably in combination with these administrations. We hypothesize that we can guide specially formulated OMBs via ultrasound imaging, and preferentially release oxygen, in the tumor for radiosensitization to significantly improve the radiotherapy therapeutic ratio. In order to test our hypothesis, will optimize microbubble formulations and administration parameters, evaluate biological mechanisms and kinetics, and validate our hypothesis both in a rodent model and then in a translational large animal model across two of the leading veterinary schools in the country. To achieve these goals, we approach this project with a collaborative team of leading experts in the fields of microbubbles, oxygen transport, radiation oncology, and tumor biology.

Date: 12/01/16 - 5/31/21
Amount: $73,870.00
Funding Agencies: Morris Animal Foundation

Gallbladder mucoceles were rare as recently as 10 years ago. They are now recognized as one of the most common, poorly understood, and deadliest biliary diseases of dogs. A mucocele develops when the gallbladder secretes abnormal mucus that eventually obstructs or ruptures the gallbladder. By the time the diagnosis is made, emergency surgery to remove the gallbladder is often required and up to 45% of these dogs do not survive to discharge. As veterinarians we don’t know why dogs develop mucoceles nor how to predict, prevent, or reverse their formation. There is therefore a critical need to determine the cause of mucocele formation so that we can put an end to the epidemic rise in cost of healthcare and lost lives of these dogs. Hypothesis/Objectives: The long-range goal of this research is to pinpoint the cause of gallbladder mucocele formation and to identify evidence-based strategies for prediction, prevention, or reversal of mucocele formation in dogs. The objective in this application is to determine the role of gallbladder epithelial secretion in the pathogenesis and treatment of mucocele formation. Our working theory, based on key insights from our prior MAF study (D12CA-044), is that mucoceles are caused by an acquired inhibition of Cl- secretion by the cystic fibrosis transmembrane regulatory channel (CFTR) located in the gallbladder epithelium. Moreover, that this “defect” results from biliary excretion of a xenobiotic that blocks CFTR function only in genetically susceptible dogs. In Specific Aim I we will determine the identity and functional activity of the Cl- secretory channels that reside in the gallbladder epithelium of dogs that have formed mucoceles compared to normal dogs and dogs with non-mucocele gallbladder disease. These studies will determine if dogs that form mucoceles have a specific defect in CFTR function and whether pharmacological stimulation of non-CFTR Cl- channels can be used to rescue normal gallbladder secretory activity. Specific Aim II will determine if dogs that form mucoceles share a common genetic polymorphism in the CFTR gene or have a common impairment of CFTR expression or intracellular trafficking that can mechanistically explain or predict the predisposition of individual dogs to mucocele formation.

Date: 04/13/15 - 3/31/21
Amount: $746,761.00
Funding Agencies: National Institutes of Health (NIH)

Contrast enhanced ultrasound imaging is poised to make substantial advances in clinical diagnostics. New imaging systems are the size of laptop computers and even smart phones, and are priced only a fraction of what they were a decade ago. Ultrasound imaging requires no ionizing radiation, and the time required to perform an imaging study, as well as image acquisition time, is less than that of MRI and nuclear imaging technologies. Contrast enhanced ultrasound studies can be performed safely, at low cost and at the patient’s bedside as often as several times per day. The advent of safe, portable and inexpensive ultrasound methods to track the location and extent of angiogenesis in primary tumors and their metastases would be a seminal accomplishment. Thus, one of the most exciting new applications for contrast ultrasound is molecular imaging. Ultrasound molecular imaging (USMI) technology involves the use of microbubbles synthesized with a targeting ligand that binds to a specific vascular epitope. One of the most successful applications of USMI has been with the use of RGD-coated microbubbles to target αVβ3 integrin, a biomarker for tumor angiogenesis. Results in rodent models of cancer have shown that ultrasound molecular imaging provides an early indication of tumor response to therapy, earlier than the gold-standard RECIST approach based on tumor size, as well as blood flow assessment. While ultrasound molecular imaging of tumor angiogenesis has passed the pilot developmental stage and shows promise, it has not yet been significantly evaluated within the context of monitoring the tumor response to pharmacologic therapy in a clinical setting. Since there are many regulatory and production hurdles limiting immediate molecularly targeted contrast agent use in humans in the United States, a desirable alternative is to evaluate spontaneous tumors in canine patients, which can be performed immediately yet retains high clinical relevance. Therefore, one major goal of our work is to validate USMI in canine patients. Additionally, one must consider that the spatial and temporal heterogeneity in the angiogenic molecules associated with tumor development makes it difficult to use a single molecular imaging agent to assess the molecular events associated with tumor angiogenesis accurately.1 Fortunately, targeted microbubbles are rapidly removed from circulation (~5-10 min), allowing the assessment of multiple targets within a single imaging session. Thus, the first aim of this project will validate USMI for two targets (αVβ3 and VEGFR2) in conjunction with standard-of-care pharmacologic interventions.

Date: 01/01/14 - 6/30/20
Amount: $135,354.00
Funding Agencies: American Kennel Club Canine Health Foundation, Inc.

The gallbladder mucocele (GBM) is one of the most common, poorly understood and deadliest biliary diseases of dogs. A GBM develops when the gallbladder secretes abnormal mucus that eventually obstructs or ruptures the gallbladder. GBM formation afflicts all dogs, but especially Shetland sheepdogs and dogs with disorders of steroid hormone or lipid metabolism. By the time a diagnosis of GBM is made, emergency surgery to remove the gallbladder is often required. After surgery only 22-50% of dogs survive to be discharged from the hospital. There is a critical need to determine why dogs form GBM so we can prevent the high cost and lost lives of these dogs. Based on the breeds and diseases that predispose to GBM, we hypothesize these dogs have a unique disturbance in cholesterol or lipid metabolism that instigates GBM formation. In specific aim 1 we will determine the mechanistic origin of abnormal cholesterol and/or lipid metabolism by performing targeted mass spectrometry using serum of dogs with and without GBM formation. In specific aim 2 we will identify the metabolic pathways responsible for abnormal lipid transport and mucin secretion by performing microarray analysis of gene expression by the gallbladder epithelium of dogs with and without GBM. Upon completion of these studies we expect to have located the mechanistic origin of GBM formation. These results will then allow us to investigate specific metabolic end-product(s) of abnormal cholesterol or lipid metabolism as the instigating cause and/or predictive biomarker of GBM formation in dogs.

Date: 03/01/16 - 7/01/19
Amount: $5,484.00
Funding Agencies: American College of Veterinary Internal Medicine (ACVIM)

This pilot study proposes to directly compare the results of TTE using commercial echocardiographic contrast to a “gold standard” diagnosis made either by computed tomographic or selective angiographic imaging. Echocardiographic contrast agents are used to highlight blood flow and tissue perfusion. Commercial echocardiographic contrast agents consist of a gas core that is encapsulated in a lipid or polymer shell. The gas bubbles are small enough to pass intact through the pulmonary circulation (unlike the traditional “bubble study” contrast), and the gas core creates highly echogenic microbubbles that are easily detected echocardiographically. Contrast agents are commonly used in human patients in Europe and Asia,26,27 and have been used to assess coronary artery disease. Isolated human deaths that occurred soon after contrast administration some years ago raised safety concerns regarding the technique, but these deaths were not proven to be caused by contrast administration.28 Large multicenter studies have since shown that the use of commercial echo contrast during stress echocardiography is safe in people.29,30 Multiple smaller studies have shown that contrast-enhanced ultrasonography is a safe and clinically valuable diagnostic procedure in dogs and cats, and these studies have not been associated with adverse events.31

Date: 04/30/18 - 6/30/19
Amount: $7,500.00
Funding Agencies: American College of Veterinary Radiology

Ureteral jets have been described in B-mode ultrasonography as echogenic streams of urine from the ureterovesicular junctions into the urinary bladder. A visible ureteral jet confirms patency of the ureter and can also confirm a normal insertion of the ureterovesicular junction in the region of the trigone. It is therefore helpful to determine presence and location of ureteral jets in patients with suspected ureteral obstruction either due to intraluminal material or due to trigonal urinary bladder masses, and in patients with suspected ureteral ectopia. Unfortunately, ureteral jets are not reliably detectable in every patient, and absence of an identifiable ureteral jet in B-mode or color Doppler ultrasound imaging does not rule out patency or ectopia. Absence of an identifiable ureteral jet may be secondary to hemodynamic variables at time of imaging and degree of urine production. Doppler ultrasound (Color or Power) can be used to enhance the signal of the urine inflow into the bladder, but can be a time-consuming procedure if urine output is low or if the ureterovesicular papillae are not conspicuous. Doppler imaging is also susceptible to motion artifact. Administration of a diuretic (most commonly furosemide) is sometimes used in clinical practice to increase frequency of ureteral jetting, and as a result aid in confirmation of ureteral insertion and patency. Optimal route of administration of diuretics for detection of ureteral jets has not been described, but recent work has shown that subcutaneous administration of furosemide resulted in similar urine output to both intravenous and oral routes using indwelling urinary catheters. Relative specific gravity of urine within the upper (renal, ureteral) and lower (urinary bladder) urinary tract has been described as a possible mechanism for why ureteral jets are visible, but other factors such as turbulence or cavitation at the ureteral orifice have also been speculated. There is currently no veterinary literature that describes the underlying mechanism(s) by which ureteral jets are visible with ultrasound, presence and frequency of ureteral jets in canine patients, optimal time from administration of a diuretic for assessment of ureteral jets, or whether route of administration of the diuretic affects time to ureteral jet visibility. The purpose of this study is to determine in vitro if relative specific gravity of urine between the upper (renal, ureteral) and lower (urinary bladder) urinary tract affects conspicuity of ureteral jets on B-mode and Doppler sonography. We also aim to prospectively describe the time from administration of diuretic via different routes (intravenous and subcutaneous) to time to visible ureteral jets and the overall efficacy of the two administration routes in increasing the conspicuity of ureteral jets.

Date: 03/01/15 - 10/31/16
Amount: $3,888.00
Funding Agencies: Collie Health Foundation

Gastric volvulus in humans is usually associated with abnormal function of the muscle between the chest and abdomen (diaphragm) or the nerve supplying this muscle. Evaluation of the diaphragm in dogs with GDV or breeds predisposed to GDV has not been previously reported. This pilot study will determine if there obvious differences in the motion of this muscle when comparing dogs that have recovered from GDV to breed and age matched controls. If abnormalities are found using two non-invasive tests (video x-ray and ultrasound) then a larger follow-up study will be performed. Identification of abnormalities in diaphragmatic function may lead to the development of outpatient diagnostic tests to identify dogs at risk for developing GDV.

Date: 05/01/12 - 12/31/14
Amount: $7,494.00
Funding Agencies: American College of Veterinary Radiology

Liver nodules are observed commonly on CT images in dogs. Knowledge of the CT appearance of benign versus malignant liver nodules would be helpful to determine their nature non-invasively. This would facilitate treatment decisions and help determine if further diagnostics should be aggressively pursued. In people, liver nodules can often be characterized as benign or malignant based on their CT characteristics, without invasive testing. In veterinary medicine there is very little information on CT characteristics of malignant vs. benign liver nodules. This is due to the difficulty in sampling a specific liver nodule, as CT-guided sampling of the liver is more cumbersome and risky than ultrasound-guided tissue sampling, and nodules seen on ultrasound imaging can rarely be correlated to a specific CT nodule. Virtual navigation software allows fusion of CT and US images, so that both CT and US images will move simultaneously in 3-dimensions during the ultrasound examination. This allows accurate correlation of US and CT findings. The purpose of this study is to evaluate a virtual navigation software system for ultrasound-guided tissue sampling of specific liver nodules identified using CT. We hypothesize this will allow comparison of CT findings with histologic diagnosis. Six cadaver dogs will be used to characterize the accuracy of the virtual navigation system. Six 3.0mm barium impregnated polyethylene spheres (BIPS) will be inserted into the liver, for a total of 36 hepatic lesions. In these cadavers, registration accuracy will be evaluated using both fiducial skin markers and anatomic landmarks for image registration. We will also evaluate the effect of repositioning the cadaver between CT and US imaging on registration accuracy with an accepted error being <5mm. CT will be performed and volumetric CT data will be uploaded onto the virtual navigation system (Virtual Navigation System, Esaote SpA, Genoa, Italy). A 22 gauge needle will be guided to the liver lesion using virtual navigation, and distance between needle tip and pseudolesion will be determined with an additional CT scan. Using the most accurate registration and positioning method, ultrasound-guided sampling of liver nodules in clinical patients will be performed in 20 dogs using virtual navigation software. CT imaging findings during triple-phase contrast enhanced CT will be evaluated and compared with cytologic and histopathologic findings. Virtual navigation finally provides a means of combining CT and ultrasound imaging, and correlating imaging findings with cytologic or histopathologic diagnoses. With establishment of feasibility and accuracy of this virtual navigation software, liver nodules could be diagnosed noninvasively and a future application would include radiofrequency and microwave ablation of non-surgically resectable tumors.

Date: 09/01/11 - 7/31/14
Amount: $52,156.00
Funding Agencies: Morris Animal Foundation

Background ? Gallbladder mucoceles were remarkably rare as recently as 10 years ago. They are now recognized as one of the most common biliary diseases of dogs. A cholecystectomy is the most common treatment for dogs with a mucocele. The surgery is associated with a very high mortality rate - 22-50% of dogs with mucoceles do not survive to discharge. As veterinarians we don?t know why dogs get mucoceles, we don?t know what the congealed mass of ?mucus? is really composed of, and there is no human counterpart to this condition. The long-term goals of this research are to determine the cause of gallbladder mucocele formation in dogs, to identify a bile biomarker of risk for mucocele maturation, and to formulate a rational strategy for dietary or pharmacological prevention or dissolution of mucoceles in dogs. Hypothesis ? Our working theory is that mucoceles are caused by a change in the primary composition of bile and that this ?altered? bile incites hypersecretion and abnormal processing of mucins by the gallbladder. In support of this theory we hypothesize here that 1) the primary (common hepatic duct) bile of dogs with mucoceles will have a unique metabolomic composition compared to dogs without mucoceles & 2) that the gallbladder of dogs with mucoceles will significantly differ in expression and content of gel-forming apomucins. Specific Objectives ? In Specific Aim I the metabolomic composition of common hepatic duct bile from dogs with mucoceles and normal dogs will be analyzed by means of liquid chromatography and mass spectrometry (LC-MS/MS) to identify any elements that are unique to dogs with mucoceles. In Specific Aim II we will determine if there are any differences in expression of gallbladder apomucins between dogs with mucoceles compared to normal dogs and identify the composition and biostructure of gallbladder ?mucus? from dogs with mucoceles compared to normal dogs to determine why it behaves so abnormally. Experimental Approach ? In Specific Aim I we will collect bile from the common hepatic ducts of dogs with mucocele undergoing cholecystectomy and control dogs after euthanasia by animal control (n=10 each). Because this bile lies upstream of the gallbladder it lacks the confounding influence of modification by the gallbladder epithelium and best represents primary bile. Analysis by means of LC-MS/MS will enable us to detect a myriad of possible differences between bile from dogs with and without mucoceles. This ensures our success in detecting a difference in composition of bile if such a difference indeed exists and then enables us to determine the identity of the unique elements by using a targeted MS analysis. Our institution diagnosed 45 dogs with mucocele last year and 4 cases have already been collected for this study. In Specific Aim II we will use samples of gallbladder tissue and content from these same dogs (n=10 each) to quantify differences in expression of gallbladder apomucins between dogs with mucoceles compared to normal dogs using RT-qPCR. Laser micro-dissection will be used to capture, identify, and quantify apomucin mRNAs specifically expressed by the gallbladder epithelium. Canine-specific qPCR assays are already designed or in hand. A direct analysis of the gallbladder content, undertaken by a collaborator who is a mucin biologist, will determine the composition and biostructural abnormality of gallbladder ?mucus? using state-of-the-art approaches to separate and analyze mucins, globular proteins and their complexes in gallbladder content of normal dogs and dogs with mucoceles. Expected Results/Anticipated Outcomes ? This study has the potential to unite all underlying causes for mucocele formation in dogs into a single change in bile composition. This ?biomarker? will enhance our current understanding of why mucoceles form, how to prevent their development, and aid in the early screening of dogs for predisposition to mucocele formation. Given the association of mucocele formation in dogs with endocrinopathy or phopholipid transport defects (i.e. Shelties), we suspect a chan

View all grants