Veterinary informatics: forging the future between veterinary medicine, human medicine, and One Health initiatives-a joint paper by the Association for Veterinary Informatics (AVI) and the CTSA One Health Alliance (COHA)

Deep Learning-Based Classification of Canine Cataracts from Ocular B-Mode Ultrasound Images

Cataracts are a prevalent cause of vision loss in dogs, and timely diagnosis is essential for effective treatment. This study aimed to develop and evaluate deep learning models to automatically classify canine cataracts from ocular ultrasound images. A dataset of 3155 ultrasound images (comprising 1329 No cataract, 614 Cortical, 1033 Mature, and 179 Hypermature cases) was used to train and validate four widely used deep learning models (AlexNet, EfficientNetB3, ResNet50, and DenseNet161). Data augmentation and normalization techniques were applied to address category imbalance. DenseNet161 demonstrated the best performance, achieving a test accuracy of 92.03% and an F1-score of 0.8744. The confusion matrix revealed that the model attained the highest accuracy for the No cataract category (99.0%), followed by Cortical (90.3%) and Mature (86.5%) cataracts, while Hypermature cataracts were classified with lower accuracy (78.6%). Receiver Operating Characteristic (ROC) curve analysis confirmed strong discriminative ability, with an area under the curve (AUC) of 0.99. Visual interpretation using Gradient-weighted Class Activation Mapping indicated that the model effectively focused on clinically relevant regions. This deep learning-based classification framework shows significant potential for assisting veterinarians in diagnosing cataracts, thereby improving clinical decision-making in veterinary ophthalmology.

Applications and Considerations of Artificial Intelligence in Veterinary Sciences: A Narrative Review

In recent years, artificial intelligence (AI) has brought about a significant transformation in healthcare, streamlining manual tasks and allowing professionals to focus on critical responsibilities while AI handles complex procedures. This shift is not limited to human healthcare; it extends to veterinary medicine as well, where AI's predictive analytics and diagnostic abilities are improving standards of animal care. Consequently, healthcare systems stand to gain notable advantages, such as enhanced accessibility, treatment efficacy, and optimized resource allocation, owing to the seamless integration of AI. This article presents a comprehensive review of the manifold applications of AI within the domain of veterinary science, categorizing them into four domains: clinical practice, biomedical research, public health, and administration. It also examines the primary machine learning algorithms used in relevant studies, highlighting emerging trends in the field. The research serves as a valuable resource for scholars, offering insights into current trends and serving as a starting point for those new to the field.

The missing link: Electronic health record linkage across species offers opportunities for improving One Health

Objective -

Significant opportunities for understanding the co-occurrence of conditions across species in coincident households remain untapped. We determined the feasibility of creating a Companion Care Registry (CCR) for analysis of health data from the University of Colorado Health (UCHealth) patients and their companion animals who received veterinary care at the geographically-adjacent Colorado State University Veterinary Teaching Hospital (CSU-VTH).

Materials and Methods -

Using a hybrid deterministic and probabilistic record linkage method, non-medical Personally Identifiable Information was securely matched to determine the total number of UCHealth patients within the HIPAA-compliant Health Data Compass Research Data Warehouse (2015-2024) who took a companion animal to the CSU-VTH (2019-2024).

Results -

12,115 matches were identified, indicating 29% of CSU-VTH clients were UCHealth patients.

Discussion -

The overlap between CSU-VTH clients and UCHealth patients underscores the potential feasibility and utility of a CCR.

Conclusion -

This work provides a mechanism to evaluate environmental and inter-species influences on One Health.

Towards mHealth applications for pet animal owners: a comprehensive literature review of requirements

BACKGROUND

Veterinarians experience high workloads and stress levels in their daily work, of which they need to be relieved as much as possible. The general public is showing great interest in digital health services. At the same time, animal owners and veterinarians are seeing telehealth services as particularly positive for triage aspects in veterinary medicine. One approach to support veterinarians may be to enable pet owners to, for instance, make informed decisions on how urgent their animal needs to be examined by a veterinary professional through an mHealth application. For this, stakeholder requirements need to be gathered, which should provide as a starting point for the development of such a decision support system.

RESULTS

955 publications were screened, resulting in the extraction of 10 requirements to mHealth applications for animal owners from 13 publications. Most frequently mentioned aspects were: ensuring complete information input by the user (6 mentions) and displaying a disclaimer about application limitations prominently (5 mentions).

CONCLUSIONS

Most of the extracted requirements focus on the design of the human-computer interface, revealing this as a crucial point to such applications, especially in guiding animal owners through information and ensuring understanding, particularly of application limitations. However, the small number of included publications shows that primary research in this field, in general, and in this specific topic in particular, is needed in order to fully reflect the requirements for an mHealth application to help animal owners decide on their animal's need to be examined by a veterinary professional.

Review of applications of deep learning in veterinary diagnostics and animal health

Deep learning (DL), a subfield of artificial intelligence (AI), involves the development of algorithms and models that simulate the problem-solving capabilities of the human mind. Sophisticated AI technology has garnered significant attention in recent years in the domain of veterinary medicine. This review provides a comprehensive overview of the research dedicated to leveraging DL for diagnostic purposes within veterinary medicine. Our systematic review approach followed PRISMA guidelines, focusing on the intersection of DL and veterinary medicine, and identified 422 relevant research articles. After exporting titles and abstracts for screening, we narrowed our selection to 39 primary research articles directly applying DL to animal disease detection or management, excluding non-primary research, reviews, and unrelated AI studies. Key findings from the current body of research highlight an increase in the utilisation of DL models across various diagnostic areas from 2013 to 2024, including radiography (33% of the studies), cytology (33%), health record analysis (8%), MRI (8%), environmental data analysis (5%), photo/video imaging (5%), and ultrasound (5%). Over the past decade, radiographic imaging has emerged as most impactful. Various studies have demonstrated notable success in the classification of primary thoracic lesions and cardiac disease from radiographs using DL models compared to specialist veterinarian benchmarks. Moreover, the technology has proven adept at recognising, counting, and classifying cell types in microscope slide images, demonstrating its versatility across different veterinary diagnostic modality. While deep learning shows promise in veterinary diagnostics, several challenges remain. These challenges range from the need for large and diverse datasets, the potential for interpretability issues and the importance of consulting with experts throughout model development to ensure validity. A thorough understanding of these considerations for the design and implementation of DL in veterinary medicine is imperative for driving future research and development efforts in the field. In addition, the potential future impacts of DL on veterinary diagnostics are discussed to explore avenues for further refinement and expansion of DL applications in veterinary medicine, ultimately contributing to increased standards of care and improved health outcomes for animals as this technology continues to evolve.

Developing a one health data integration framework focused on real-time pathogen surveillance and applied genomic epidemiology

BACKGROUND

The One Health approach aims to balance and optimize the health of humans, animals, and ecosystems, recognizing that shared health outcomes are interdependent. A One Health approach to disease surveillance, control, and prevention requires infrastructure for coordinating, collecting, integrating, and analyzing data across sectors, incorporating human, animal, and environmental surveillance data, as well as pathogen genomic data. However, unlike data interoperability problems faced within a single organization or sector, data coordination and integration across One Health sectors requires engagement among partners to develop shared goals and capacity at the response level. Successful examples are rare; as such, we sought to develop a framework for local One Health practitioners to utilize in support of such efforts.

METHODS

We conducted a systematic scientific and gray literature review to inform development of a One Health data integration framework. We discussed a draft framework with 17 One Health and informatics experts during semi-structured interviews. Approaches to genomic data integration were identified.

RESULTS

In total, 57 records were included in the final study, representing 13 pre-defined frameworks for health systems, One Health, or data integration. These frameworks, included articles, and expert feedback were incorporated into a novel framework for One Health data integration. Two scenarios for genomic data integration were identified in the literature and outlined.

CONCLUSIONS

Frameworks currently exist for One Health data integration and separately for general informatics processes; however, their integration and application to real-time disease surveillance raises unique considerations. The framework developed herein considers common challenges of limited resource settings, including lack of informatics support during planning, and the need to move beyond scoping and planning to system development, production, and joint analyses. Several important considerations separate this One Health framework from more generalized informatics frameworks; these include complex partner identification, requirements for engagement and co-development of system scope, complex data governance, and a requirement for joint data analysis, reporting, and interpretation across sectors for success. This framework will support operationalization of data integration at the response level, providing early warning for impending One Health events, promoting identification of novel hypotheses and insights, and allowing for integrated One Health solutions.

Recent Issues in the Development and Application of Targeted Therapies with Respect to Individual Animal Variability

This literature review explores the impact of molecular, genetic, and environmental factors on the efficacy of targeted therapies in veterinary medicine. Relevant studies were identified through systematic searches of PubMed, Web of Science, Scopus, and ScienceDirect using keywords such as "species-specific treatment strategies", "signalling pathways", "epigenetic and paragenetic influences", "targeted therapies", "veterinary medicine", "genetic variation", and "free radicals and oxidative stress". Inclusion criteria included studies focusing on species-specific therapeutic responses, genetic influences, and oxidative stress. To ensure that only the most recent and relevant evidence was included, only peer-reviewed publications from the last two decades were considered. Each study selected for analysis was critically appraised, with a particular emphasis on methodological quality, experimental design, and scientific contribution to the understanding of how environmental and biological factors influence therapeutic outcomes. A special emphasis was placed on studies that used a comparative, cross-species approach to assess variability in therapeutic responses and potential adverse effects. The review synthesises evidence on the role of epigenetic and paragenetic factors and highlights the importance of cross-species studies to understand how environmental and biological factors influence treatment outcomes. By highlighting genetic variation, oxidative stress, and individual species differences, the review argues for personalised and species-specific therapeutic approaches. The review emphasises that such an approach would improve veterinary care and inform future research aimed at optimising targeted therapies, ultimately leading to better animal health and treatment efficacy. A key contribution of the review is its emphasis on the need for more personalised treatment protocols that take into account individual genetic profiles and environmental factors; it also calls for a greater integration of cross-species studies.

Exploring the potential and limitations of artificial intelligence in animal anatomy

BACKGROUND

Artificial intelligence (AI) is revolutionizing veterinary medicine, particularly in the domain of veterinary anatomy. At present, there is no existing review article in the literature that examines the prospects and challenges associated with the use of AI in animal anatomy education.

STUDY DESIGN

Narrative review.

OBJECTIVE

This review article explores the prospects and drawbacks of AI applications in veterinary anatomy. Anatomy and AI-powered diagnostic systems enhance clinical examination, diagnosis, and treatment by analyzing vast datasets, improving accuracy, and detecting subtle anomalies.

METHODS

We reviewed and analyzed recent literature on AI applications in veterinary anatomy education, emphasizing their potential, limitations, and future directions..

CONCLUSION

In veterinary anatomy education, AI integrates advanced tools like three-dimensional (3D) models, virtual reality (VR), and augmented reality (AR), offering dynamic and interactive learning experiences to students as well as the faculty of veterinary institutions across the globe. Despite these advantages, AI faces challenges such as the need for extensive, high-quality data, potential biases, and issues with algorithmic transparency. Additionally, virtual dissection and educational tools may impact hands-on learning and ethical and legal concerns regarding data privacy must be addressed. Balancing AI integration with traditional skills and addressing these challenges will maximize AI's benefits in veterinary anatomy and ensure comprehensive veterinary care.

Development and validation of a machine learning model for clinical wellness visit classification in cats and dogs

Introduction

Early disease detection in veterinary care relies on identifying subclinical abnormalities in asymptomatic animals during wellness visits. This study introduces a model designed to distinguish between wellness and other types of veterinary visits.

Objectives

The purpose of this study is to validate the use of a visit classification model compared to manual classification of veterinary visits by three board-certified veterinarians.

Materials and methods

The algorithm was initially trained using a Gradient Boosting Machine model with a dataset of 11,105 clinical visits from 2012 to 2017 involving 655 animals (85.3% dogs and 14.7% cats) across 544 U.S. veterinary practices. Three validators were tasked with classifying 400 visits, including both wellness and other types of visits, selected randomly from the same database used for initial model training, aiming to maintain consistency and relevance between the training and application phases; visit classifications were subsequently categorized into "wellness" or "other" based on majority consensus among validators to assess the model's performance in identifying wellness visits.

Results

The model demonstrated a specificity of 0.94 (95% CI: 0.91 to 0.96), implying its accuracy in distinguishing non-wellness visits. The model had a sensitivity of 0.86 (95% CI: 0.80 to 0.92), indicating its ability to correctly identify wellness visits as compared to the annotations provided by veterinary experts. The balanced accuracy, calculated as 0.90 (95% CI: 0.87 to 0.93), further confirms the model's overall effectiveness.

Clinical significance

The model exhibits high specificity and sensitivity, ensuring accurate identification of a high proportion of wellness visits. Overall, this model holds promise for advancing research on preventive care's role in subclinical disease identification, but prospective studies are needed for validation.

The TiHoCL panel for canine lymphoma: a feasibility study integrating functional genomics and network biology approaches for comparative oncology targeted NGS panel design

Targeted next-generation sequencing (NGS) enables the identification of genomic variants in cancer patients with high sensitivity at relatively low costs, and has thus opened the era to personalized human oncology. Veterinary medicine tends to adopt new technologies at a slower pace compared to human medicine due to lower funding, nonetheless it embraces technological advancements over time. Hence, it is reasonable to assume that targeted NGS will be incorporated into routine veterinary practice in the foreseeable future. Many animal diseases have well-researched human counterparts and hence, insights gained from the latter might, in principle, be harnessed to elucidate the former. Here, we present the TiHoCL targeted NGS panel as a proof of concept, exemplifying how functional genomics and network approaches can be effectively used to leverage the wealth of information available for human diseases in the development of targeted sequencing panels for veterinary medicine. Specifically, the TiHoCL targeted NGS panel is a molecular tool for characterizing and stratifying canine lymphoma (CL) patients designed based on human non-Hodgkin lymphoma (NHL) research outputs. While various single nucleotide polymorphisms (SNPs) have been associated with high risk of developing NHL, poor prognosis and resistance to treatment in NHL patients, little is known about the genetics of CL. Thus, the ~100 SNPs featured in the TiHoCL targeted NGS panel were selected using functional genomics and network approaches following a literature and database search that shielded ~500 SNPs associated with, in nearly all cases, human hematologic malignancies. The TiHoCL targeted NGS panel underwent technical validation and preliminary functional assessment by sequencing DNA samples isolated from blood of 29 lymphoma dogs using an Ion Torrent™ PGM System achieving good sequencing run metrics. Our design framework holds new possibilities for the design of similar molecular tools applied to other diseases for which limited knowledge is available and will improve drug target discovery and patient care.

Intersecting Pathways in Bioinformatics and Translational Informatics: A One Health Perspective on Key Contributions and Future Directions

OBJECTIVES

To identify and summarize the top bioinformatics and translational informatics (BTI) papers published in 2022 for the International Medical Informatics Association (IMIA) Yearbook 2023.

METHODS

We conducted a comprehensive literature search to identify the top BTI papers, resulting in a set of ten candidate papers. The candidates were reviewed by the section co-editors and external reviewers to select the top three papers from 2022.

RESULTS

From a total of 558 papers, we identified a final candidate list of ten BTI papers for peer-review. These papers apply new statistical frameworks and experimental designs to better capture individual variability in disease and incorporate data that captures differences between single cells and across environmental exposures. In addition, they highlight the importance of model generalization across diverse cohorts and scalability to large medical centers.

CONCLUSIONS

We note several important trends in the candidate top BTI papers this year, including a continued focus on developing accurate and scalable computational models to predict disease risk across diverse cohorts and new strategies to capture the molecular heterogeneity of disease.

Artificial Intelligence to Predict the BRAF V595E Mutation in Canine Urinary Bladder Urothelial Carcinomas

In dogs, the BRAF mutation (V595E) is common in bladder and prostate cancer and represents a specific diagnostic marker. Recent advantages in artificial intelligence (AI) offer new opportunities in the field of tumour marker detection. While AI histology studies have been conducted in humans to detect BRAF mutation in cancer, comparable studies in animals are lacking. In this study, we used commercially available AI histology software to predict BRAF mutation in whole slide images (WSI) of bladder urothelial carcinomas (UC) stained with haematoxylin and eosin (HE), based on a training (n = 81) and a validation set (n = 96). Among 96 WSI, 57 showed identical PCR and AI-based BRAF predictions, resulting in a sensitivity of 58% and a specificity of 63%. The sensitivity increased substantially to 89% when excluding small or poor-quality tissue sections. Test reliability depended on tumour differentiation (p < 0.01), presence of inflammation (p < 0.01), slide quality (p < 0.02) and sample size (p < 0.02). Based on a small subset of cases with available adjacent non-neoplastic urothelium, AI was able to distinguish malignant from benign epithelium. This is the first study to demonstrate the use of AI histology to predict BRAF mutation status in canine UC. Despite certain limitations, the results highlight the potential of AI in predicting molecular alterations in routine tissue sections.

Using a gradient boosted model for case ascertainment from free-text veterinary records

Case ascertainment for prevalence and incidence studies from veterinary clinical data poses a major challenge because medical notes are not consistently structured or complete. Using natural language processing (NLP) and machine learning, this study aimed to obtain accurate case recognition for feline upper respiratory tract infections (primarily caused by viruses such as feline herpes virus (FHV-1) and feline calici virus (FCV), and bacteria such as Chlamydophila felis, Mycoplasma felis and Bordetella bronchiseptica using retrospective electronic veterinary records from the Royal Society for Prevention of Cruelty to Animals, Queensland (RSPCA Qld). Data cleaning and NLP on eight years of free-text veterinary records from RSPCA Queensland was carried out to derive text-based predictors. The NLP steps included sorting records by length of stay, vectorising, tokenising and spell checking against a bespoke veterinary database. A gradient boosted model (GBM) was trained to predict the probability of each animal having a diagnosis of upper respiratory infection. A manually annotated dataset was used for training the algorithm to learn dominant patterns between predictors (frequencies of n-grams) and responses (manual binary case classification). The GBM's performance was tested against an out of sample validation dataset, and model agnostics were used to interrogate the model's learning process. The GBM used patient-level frequencies of 1250 unique n-grams as predictor variables and was able to predict the probability of cases in the validation dataset with an accuracy of 0.95 (95% CI 0.92, 0.97) and F1 score of 0.96. Predictors that exerted the highest influence on the model included frequencies of "doxycycline", "flu", "sneezing", "doxybrom" and "ocular". The trained GBM was deployed on the full dataset spanning eight years, comprising 60,258 clinical entries. The prevalence in the full dataset was predicted to be 23.59%, which is in line with domain expertise from practicing veterinarians at the shelter. Case ascertainment is a crucial step for further epidemiological study of cat flu. Ultimately, this tool can be extended to other clinical procedures, conditions, and diseases such as intensive care treatment due to snake bites and tick paralysis, physical injuries such as orthopaedic fractures or chest injuries and labour-intensive infectious diseases like parvovirus, canine cough, and ringworm, all of which require prolonged quarantine and care.

Artificial intelligence in veterinary diagnostic imaging: A literature review

Artificial intelligence in veterinary medicine is an emerging field. Machine learning, a subfield of artificial intelligence, allows computer programs to analyze large imaging datasets and learn to perform tasks relevant to veterinary diagnostic imaging. This review summarizes the small, yet growing body of artificial intelligence literature in veterinary imaging, provides necessary background to understand these papers, and provides author commentary on the state of the field. To date, less than 40 peer-reviewed publications have utilized machine learning to perform imaging-associated tasks across multiple anatomic regions in veterinary clinical and biomedical research. Major challenges in this field include collection and cleaning of sufficient image data, selection of high-quality ground truth labels, formation of relationships between veterinary and machine learning professionals, and closure of the gap between academic uses of artificial intelligence and currently available commercial products. Further development of artificial intelligence has the potential to help meet the growing need for radiological services through applications in workflow, quality control, and image interpretation for both general practitioners and radiologists.

Case Distribution, Sources, and Breeds of Dogs Presenting to a Veterinary Behavior Clinic in the United States from 1997 to 2017

The purpose of this retrospective case study was to evaluate trends over time in case distribution, sources, and breeds of dogs presenting to the behavioral medicine service at a veterinary college referral hospital in the United States. For case distribution and sources, the available records from the behavior service (n = 1923) from 1997 to 2017 were evaluated. Breeds of dogs presenting to all services (n = 51,052) were compared to behavior cases (n = 822) from 2007 to 2016. Over twenty years, 72.2% of dogs presented for aggression, 20.1% for anxieties/fears/phobias, and 7.4% for miscellaneous behavioral problems. Dogs acquired from breeders decreased and dogs from shelters, rescues, or adopted as a stray increased over twenty years (p < 0.0001). The Herding (p = 0.0124) and Terrier (p < 0.0001) groups were overrepresented for behavior problems as compared to all other services over ten years. Variations in terminology and diagnostic approach made comparisons with earlier studies difficult, which underscores a need for a more consistent methodology in veterinary behavioral medicine. Understanding trends in sources of dogs could direct resources aimed at guiding owners when acquiring a pet dog and preventing behavioral problems. Findings related to breeds could help guide research focused on the genetic contributions to behavior.

Machine learning for modeling the progression of Alzheimer disease dementia using clinical data: a systematic literature review

OBJECTIVE

Alzheimer disease (AD) is the most common cause of dementia, a syndrome characterized by cognitive impairment severe enough to interfere with activities of daily life. We aimed to conduct a systematic literature review (SLR) of studies that applied machine learning (ML) methods to clinical data derived from electronic health records in order to model risk for progression of AD dementia.

MATERIALS AND METHODS

We searched for articles published between January 1, 2010, and May 31, 2020, in PubMed, Scopus, ScienceDirect, IEEE Explore Digital Library, Association for Computing Machinery Digital Library, and arXiv. We used predefined criteria to select relevant articles and summarized them according to key components of ML analysis such as data characteristics, computational algorithms, and research focus.

RESULTS

There has been a considerable rise over the past 5 years in the number of research papers using ML-based analysis for AD dementia modeling. We reviewed 64 relevant articles in our SLR. The results suggest that majority of existing research has focused on predicting progression of AD dementia using publicly available datasets containing both neuroimaging and clinical data (neurobehavioral status exam scores, patient demographics, neuroimaging data, and laboratory test values).

DISCUSSION

Identifying individuals at risk for progression of AD dementia could potentially help to personalize disease management to plan future care. Clinical data consisting of both structured data tables and clinical notes can be effectively used in ML-based approaches to model risk for AD dementia progression. Data sharing and reproducibility of results can enhance the impact, adaptation, and generalizability of this research.

Day-1 Competencies for Veterinarians Specific to Health Informatics

In 2015, the American Association of Veterinary Medical Colleges (AAVMC) developed the Competency-Based Veterinary Education (CBVE) framework to prepare practice-ready veterinarians through competency-based education, which is an outcomes-based approach to equipping students with the skills, knowledge, attitudes, values, and abilities to do their jobs. With increasing use of health informatics (HI: the use of information technology to deliver healthcare) by veterinarians, competencies in HI need to be developed. To reach consensus on a HI competency framework in this study, the Competency Framework Development (CFD) process was conducted using an online adaptation of Developing-A-Curriculum, an established methodology in veterinary medicine for reaching consensus among experts. The objectives of this study were to (1) create an HI competency framework for new veterinarians; (2) group the competency statements into common themes; (3) map the HI competency statements to the AAVMC competencies as illustrative sub-competencies; (4) provide insight into specific technologies that are currently relevant to new veterinary graduates; and (5) measure panelist satisfaction with the CFD process. The primary emphasis of the final HI competency framework was that veterinarians must be able to assess, select, and implement technology to optimize the client-patient experience, delivery of healthcare, and work-life balance for the veterinary team. Veterinarians must also continue their own education regarding technology by engaging relevant experts and opinion leaders.

A CTSA One Health Alliance (COHA) survey of clinical trial infrastructure in North American veterinary institutions

While a necessary step toward enhancing rigor and reproducibility of veterinary clinical trials conducted on the translational spectrum includes understanding the current state of the field, no broad assessment of existing veterinary clinical trial resources has been previously conducted. Funded by a CTSA One Health Alliance (COHA) pilot award, the goal of this project was to conduct an electronic survey of North American Veterinary Colleges regarding practices in veterinary clinical trial review, approval, conduct, and support in order to identify opportunities to leverage existing resources and develop new ones to enhance the impact of veterinary and translational health research.A total of 30 institutions were invited to participate in the survey and the survey response rate was 73 %. The most common source of funding noted for veterinary clinical research was industry (33 %); however, respondents reported that only 5 % (3.7-11.0) of studies were FDA-regulated. Respondents indicated that most studies (80 %); conducted at their institution were single site studies. Study review and approval involved the IACUC either solely, or in combination with a hospital review board, at 95.5 % of institutions. Workforce training related to clinical research best practices was variable across institutions. Opportunities were identified to strengthen infrastructure through harmonization of clinical research review and approval practices. This might naturally lead to expansion of multi-site studies. Based on respondent feedback, future workforce development initiatives might center on training in the specifics of conducting FDA-sponsored research, Good Clinical Practice (GCP), clinical study budget design, grants management, adverse event reporting, study monitoring and use of electronic data capture platforms.