Explore the vast range of titles available.

Unstructured clinical narratives are continuously being recorded as part of delivery of care in electronic health records, and dedicated tagging staff spend considerable effort manually assigning clinical codes for billing purposes. Despite these efforts, however, label availability and accuracy are both suboptimal. In this retrospective study, we aimed to automate the assignment of top-level International Classification of Diseases version 9 (ICD-9) codes to clinical records from human and veterinary data stores using minimal manual labor and feature curation. Automating top-level annotations could in turn enable rapid cohort identification, especially in a veterinary setting. To this end, we trained long short-term memory (LSTM) recurrent neural networks (RNNs) on 52,722 human and 89,591 veterinary records. We investigated the accuracy of both separate-domain and combined-domain models and probed model portability. We established relevant baseline classification performances by training Decision Trees (DT) and Random Forests (RF). We also investigated whether transforming the data using MetaMap Lite, a clinical natural language processing tool, affected classification performance. We showed that the LSTM-RNNs accurately classify veterinary and human text narratives into top-level categories with an average weighted macro F1 score of 0.74 and 0.68 respectively. In the \"neoplasia\" category, the model trained on veterinary data had a high validation accuracy in veterinary data and moderate accuracy in human data, with F1 scores of 0.91 and 0.70 respectively. Our LSTM method scored slightly higher than that of the DT and RF models. The use of LSTM-RNN models represents a scalable structure that could prove useful in cohort identification for comparative oncology studies. Digitization of human and veterinary health information will continue to be a reality, particularly in the form of unstructured narratives. Our approach is a step forward for these two domains to learn from and inform one another.

In the United States, gonadectomy is common and widely promoted as a component of responsible pet ownership. The recent publication of several studies examining the effect of gonadectomy on future health has challenged long-held assumptions and recommendations for gonadectomy in companion animals. The purpose of this study was to characterize the associations between gonadectomy and two outcomes: overweight/obesity and orthopedic injuries, in a large prospective study of Golden Retrievers. Age at gonadectomy was divided into four categories: intact (reference), ≤ 6 months, > 6 months ‒ ≤ 12 months, and > 12 months. Dogs with a Purina Body Condition Score of 7 or greater were classified as overweight or obese. Orthopedic injuries considered were the first instance of veterinary-reported cranial cruciate ligament injury and clinically evident osteoarthritis. We performed survival analyses on a cohort of Golden Retrievers to estimate the associations of interest using proportional hazards. We adjusted for age at study enrollment, owner-reported activity level, and dog's sex. Compared to intact dogs, all gonadectomy age categories showed increased risk for the development of overweight/obesity. (≤ 6 months, HR: 1.81, 95% CI: 1.36-2.40), p-value: <0.0001; 6 months to ≤ 12 months, HR: 2.21, 95% CI: 1.77-2.73, p-value: < 0.0001; > 12 months, HR: 1.56, 95% CI: 1.24-1.96, p-value: 0.0001). Compared to intact dogs, dogs who were ≤ 6 months at gonadectomy had increased risk for orthopedic injury (HR: 4.06, 95% CI: 2.15-7.67, p-value: <0.00001). This study presents prospectively acquired data demonstrating that gonadectomy is a risk factor for both overweight/obesity and chronic non-traumatic orthopedic injuries in a prospective cohort of Golden Retrievers. Our data suggest that gonadectomy at any age is a risk factor for overweight or obesity, but delaying gonadectomy until dogs are at least 6-12 months of age may help to decrease the risk for orthopedic injury.

The aim of this article is to provide a detailed description of the Golden Retriever Lifetime Study (GRLS), a prospective cohort study investigating nutritional, environmental, lifestyle, and genetic risk factors for cancer and other common diseases in dogs. Primary outcomes of interest include hemangiosarcoma, lymphoma, osteosarcoma, and high-grade mast cell tumors. Secondary outcomes of interest include other cancers, hypothyroidism, epilepsy, atopy, otitis externa, hip dysplasia, heart failure, and renal failure. A total of 3,044 United States Golden Retrievers aged 6 months to 2 years completed baseline enrollment from June 2012 to April 2015. As of May 31, 2021, 2,251 dogs remain engaged in the study, 352 have died, and 441 are lost to follow-up. Extensive annual questionnaires completed by owners and veterinarians gather information about lifestyle, environmental exposures, physical activity, reproductive history, behavior, diet, medications, and diagnoses. Dogs also have annual veterinary examinations and biospecimen collection (blood, serum, hair, nails, feces, urine) for biobanking. Additional reporting, including histology and tumor biobanking, is conducted for any malignancies or deaths. When an animal dies, full medical records are obtained, and necropsies are requested at owner discretion. Full or partial necropsies have been performed on 218 dogs. Questionnaire data are freely available to researchers with approved credentials who agree to a data use agreement. In addition, researchers can submit proposals to utilize biospecimens or obtain additional data.

Background The Golden Retriever Lifetime Study (GRLS) is one of the largest canine cohort studies undertaken in the United States to date. This study design allows for evaluation of multiple exposures and outcomes throughout the lifetime of each dog, but relies on participants to comply with study requirements over a long period of time. Failure to do so can lead to biased reporting of results. Objectives To examine factors associated with dog owner compliance for GRLS. Animals Golden Retrievers (n = 3044) whose owners elected to participate in GRLS. Methods Prospective, cohort study. A logistic regression model was constructed to examine associations between data collected at the time of initial enrollment in GRLS and the outcome of failure to fulfill all study obligations at the end of the first year after enrollment in GRLS. Results There were 192 (6.3%) owners who did not comply with study requirements 1 year after enrollment. Owners of dogs without a record of vaccination had nearly 4 times higher odds (adjusted OR: 3.7, 95% CI: 1.5, 9.2) of being noncompliant than owners of vaccinated dogs and owners of dogs that slept in the garage had nearly 6 times higher odds (adjusted OR: 5.7, 95% CI: 1.9, 17.0) of being noncompliant than owners of dogs that slept in their bedroom. Conclusions and Clinical Importance Survey questions about a dog's sleeping location at night and vaccination status are important indicators of an owner's odds of compliance in a prospective study. Use of similar questions during enrollment in cohort studies might help to predict owner compliance that can aid in subject selection.

Inbreeding depression has been demonstrated to impact vital rates, productivity, and performance in human populations, wild and endangered species, and in recent years, the domestic species. In all cases, standardized, high-quality phenotype data on all individuals are invaluable for longitudinal analyses such as those required to evaluate vital rates of a study cohort. Further, many investigators agree upon the preference for and utility of genomic measures of inbreeding in lieu of pedigree-based estimates of inbreeding. We evaluated the association of measures of reproductive fitness in 93 Golden Retrievers enrolled in the Golden Retriever Lifetime Study with a genomic measurement of inbreeding, F ROH . We demonstrate a statistically significant negative correlation between fecundity and F ROH . This work sets the stage for larger scale analyses to investigate genomic regions associated with fecundity and other measures of fitness.

Unlike human medical records, most of the veterinary records are free text without standard diagnosis coding. The lack of systematic coding is a major barrier to the growing interest in leveraging veterinary records for public health and translational research. Recent machine learning effort is limited to predicting 42 top-level diagnosis categories from veterinary notes. Here we develop a large-scale algorithm to automatically predict all 4577 standard veterinary diagnosis codes from free text. We train our algorithm on a curated dataset of over 100 K expert labeled veterinary notes and over one million unlabeled notes. Our algorithm is based on the adapted Transformer architecture and we demonstrate that large-scale language modeling on the unlabeled notes via pretraining and as an auxiliary objective during supervised learning greatly improves performance. We systematically evaluate the performance of the model and several baselines in challenging settings where algorithms trained on one hospital are evaluated in a different hospital with substantial domain shift. In addition, we show that hierarchical training can address severe data imbalances for fine-grained diagnosis with a few training cases, and we provide interpretation for what is learned by the deep network. Our algorithm addresses an important challenge in veterinary medicine, and our model and experiments add insights into the power of unsupervised learning for clinical natural language processing.

The Golden Retriever Lifetime Study (GRLS) is the first prospective longitudinal study attempted in veterinary medicine to identify the major dietary, genetic and environmental risk factors for cancer and other important diseases in dogs. The GRLS is an observational study that will follow a cohort of 3000 purebred Golden Retrievers throughout their lives via annual online questionnaires from the dog owner and annual physical examinations and collection of biological samples by the primary care veterinarian. The field of comparative medicine investigating naturally occurring disorders in pets is specifically relevant to the many diseases that have a genetic basis for disease in both animals and humans, including cancer, blindness, metabolic and behavioural disorders and some neurodegenerative disorders. The opportunity for the GRLS to provide high-quality data for translational comparative medical initiatives in several disease categories is great. In particular, the opportunity to develop a lifetime dataset of lifestyle and activity, environmental exposure and diet history combined with simultaneous annual biological sample sets and detailed health outcomes will provide disease incidence data for this cohort of geographically dispersed dogs and associations with a wide variety of potential risk factors. The GRLS will provide a lifetime historical context, repeated biological sample sets and outcomes necessary to interrogate complex associations between genes and environmental influences and cancer.

Large scale veterinary clinical records can become a powerful resource for patient care and research. However, clinicians lack the time and resource to annotate patient records with standard medical diagnostic codes and most veterinary visits are captured in free-text notes. The lack of standard coding makes it challenging to use the clinical data to improve patient care. It is also a major impediment to cross-species translational research, which relies on the ability to accurately identify patient cohorts with specific diagnostic criteria in humans and animals. In order to reduce the coding burden for veterinary clinical practice and aid translational research, we have developed a deep learning algorithm, DeepTag, which automatically infers diagnostic codes from veterinary free-text notes. DeepTag is trained on a newly curated dataset of 112,558 veterinary notes manually annotated by experts. DeepTag extends multitask LSTM with an improved hierarchical objective that captures the semantic structures between diseases. To foster human-machine collaboration, DeepTag also learns to abstain in examples when it is uncertain and defers them to human experts, resulting in improved performance. DeepTag accurately infers disease codes from free-text even in challenging cross-hospital settings where the text comes from different clinical settings than the ones used for training. It enables automated disease annotation across a broad range of clinical diagnoses with minimal preprocessing. The technical framework in this work can be applied in other medical domains that currently lack medical coding resources.

Purpose Calcitriol potentiates cisplatin-mediated activity in a variety of tumor models. We examine here, the effect of calcitriol and cisplatin pre-clinically and clinically in canine spontaneous tumors through in vitro studies on tumor cells and through a phase I study of calcitriol and cisplatin to identify the maximum-tolerated dosage (MTD) of this combination in dogs with cancer and to characterize the pharmacokinetic disposition of calcitriol in dogs. Methods Canine tumor cells were investigated for calcitriol/cisplatin interactions on proliferation using an MTT assay in a median-dose effect analysis; data were used to derive a combination index (CI). Cisplatin was given at a fixed dosage of 60 mg/m 2 . Calcitriol was given i.v. and the dosage was escalated in cohorts of three dogs until the MTD was defined. Serum calcitriol concentrations were quantified by radioimmunoassay. Results In vitro, CIs < 1.0 were obtained for all combinations of calcitriol/cisplatin examined. The MTD was 3.75 μg/kg calcitriol in combination with cisplatin, and hypercalcemia was the dose-limiting toxicosis. The relationship between calcitriol dosage and either C max or AUC was linear. Calcitriol dosages >1.5 μg/kg achieved C max  ≥ 9.8 ng/mL and dosages >1.0 μg/kg achieved AUC ≥ 45 h ng/mL. Conclusions Calcitriol and cisplatin have synergistic antiproliferative effects on multiple canine tumor cells and high-dosages of i.v. calcitriol in combination with cisplatin can be safely administered to dogs. C max and AUC at the MTD 3.75 μg/kg calcitriol exceed concentrations associated with antitumor activity in a murine model, indicating this combination might have significant clinical utility in dogs.

Abstract Background The Golden Retriever Lifetime Study (GRLS) is one of the largest canine cohort studies undertaken in the United States to date. This study design allows for evaluation of multiple exposures and outcomes throughout the lifetime of each dog, but relies on participants to comply with study requirements over a long period of time. Failure to do so can lead to biased reporting of results. Objectives To examine factors associated with dog owner compliance for GRLS. Animals Golden Retrievers (n = 3044) whose owners elected to participate in GRLS. Methods Prospective, cohort study. A logistic regression model was constructed to examine associations between data collected at the time of initial enrollment in GRLS and the outcome of failure to fulfill all study obligations at the end of the first year after enrollment in GRLS. Results There were 192 (6.3%) owners who did not comply with study requirements 1 year after enrollment. Owners of dogs without a record of vaccination had nearly 4 times higher odds (adjusted OR: 3.7, 95% CI: 1.5, 9.2) of being noncompliant than owners of vaccinated dogs and owners of dogs that slept in the garage had nearly 6 times higher odds (adjusted OR: 5.7, 95% CI: 1.9, 17.0) of being noncompliant than owners of dogs that slept in their bedroom. Conclusions and Clinical Importance Survey questions about a dog's sleeping location at night and vaccination status are important indicators of an owner's odds of compliance in a prospective study. Use of similar questions during enrollment in cohort studies might help to predict owner compliance that can aid in subject selection.