Introduction
Artificial intelligence is generally used to describe algorithms capable of producing output previously thought to require human intelligence.1 It is a significant evolution from older machine learning algorithms that learn patterns from data in a human-assisted way.2,3 Generative AI is a recent development, with algorithms that generate text or images from a simple request written in a natural language (the prompt).4
Technological innovations generally mature following a hype cycle5 (Figure 1) characterized by an initial phase of inflated expectations (the hype), followed by a trough of disillusionment before tangible value creation takes off. Artificial intelligence is no exception. Backed by extensive research, AI has generated—and is still generating—massive expectations in many domains including human and veterinary medicine.6–12 But where are we on tangible value creation in small animal veterinary medicine?
Hype cycle for technological innovation5 with the AI use cases discussed in this paper.
Citation: Journal of the American Veterinary Medical Association 263, 3; 10.2460/javma.24.09.0617
Value creation starts when technological innovations provide practical solutions to real-world problems or use cases. This review will assess current AI value creation in a diverse set of use cases. Findings will be used to convey a realistic outlook and suggestions for strategies to speed up AI value creation in small animal veterinary medicine.
Use Case 1: Diagnostic Imaging
Diagnostic imaging is one of the most common diagnostic tests in veterinary medicine. Unfortunately, the demand for veterinary radiologists outstrips their availability, leading to potential treatment delays and impacting patient welfare.13 Teleradiology, in which the radiologist interprets images remotely, makes it easier to access radiologists but is not enough, as the low number of experts in many specialist areas still limits throughput. Therefore, obtaining rapid, high-quality specialist assessments of images remains a significant problem.
The computational discipline that designs AI algorithms for medical images is called radiomics.14 Veterinary radiomics is already quite advanced, for two main reasons. Firstly, there are large amounts of accurate digitalized expert labeled data available for training new algorithms (ie, routinely collected radiographs that have a diagnosis label). Secondly, applications often build on highly optimized algorithms previously applied to other images or domains.15,16 This has led to the publication of several high-performing algorithms that can identify potential quality problems with images17 or can extract medically relevant information,18–24 with several commercial solutions for image annotations now being available.25–29
In terms of adoption, radiomics has the benefit that its outputs can be displayed on the image so that these can always be validated by a specialist whose feedback can also be used to improve the algorithms. This greatly alleviates common explainability concerns linked to the black box nature of AI models30–32 and prevents being misguided by AI-induced artifacts or hallucinations. It is unlikely that radiomics will replace trained experts any time soon,33 but it does already reduce their workload by simplifying the tasks of identifying important features in the images. Taken together, AI for diagnostic imaging is surely beyond the hype and is in the initial stages of adding tangible value to veterinary workflows.
Use Case 2: Early Disease Detection
In current small animal veterinary medicine settings, many cases involve managing diseases already at advanced stages at the time of diagnosis. Earlier detection would obviously be desirable for potential improvements in prognosis, cost, and more timely, shorter, and less aggressive interventions. Early disease detection starts by individual-level risk models that allow veterinarians to identify pets at risk on the basis of demographics and clinical information so that, when appropriate, additional diagnostic tests can be performed during routine yearly checkups. Predictive models can then use all available information to support an early diagnosis. As it stands, these approaches, which are common in human medicine to predict mortality34,35 or risk of cardiovascular diseases36 or cancer,37 for example, are rarely applied in small animal veterinary medicine.
Individual-level risk models estimate the risk of a specific individual to develop a medical condition at a given point in their life.38,39 The development of accurate veterinary risk models requires data on pet signalments, health information, and other observations, with less common conditions generally requiring larger datasets.12,40,41 Electronic medical record systems42–45 are the most common source of such data,46 with specific initiatives such as VetCompass47–49 making such datasets available to researchers. A few risk models have been developed for pets,47,48,50 but important challenges remain. One of the main problems is that the data used can be biased due to being extracted from specific pet owner populations or due to unequal access to care, attitudes, economic status, or geographical location.51 Therefore, advanced methodologies that are capable of explicitly addressing biases in the data should be considered to ensure the accuracy of models, such as Bayesian52–54 or causal55,56 approaches.
Early disease prediction models are types of individual-level risk models that use patient demographics supplemented with diagnostic test data, clinical signs, genetic data, or other pieces of information to support an early diagnosis of a disease with confidence and often do so before a reliable diagnosis can be obtained on the basis of clinical signs only. Artificial intelligence models have been developed to evaluate the risk of several veterinary conditions including anterior cruciate ligament rupture,57 chronic kidney disease,58–61 Cushing syndrome,62 dermatitis,63 epilepsy,64 heart failure,65 peritonitis,66 hyperthyroidism,67 and hypoadrenocorticism.68,69 A major barrier to applying these in a clinical setting is the black box nature of many of these models, as they provide risk profiles without describing how or why a given result has been obtained.70–72 Moreover, most of these models are only validated by the same researchers who created the model, often using data very similar to those used for model development. The lack of independent validation is a significant problem that affects veterinary and human models alike.73,74 This concern is valid and real, as some human AI models have been found to be biased when applied in real-word scenarios.75,76
Overall, while progress is being made, the opportunities provided by AI have not yet moved the needle significantly for early disease detection in the clinic. Data availability is an important limiting factor that is apparent at the model-building stage, limiting performance, and at the independent model validation stage, limiting trust building. The black box nature of the AI models is an additional challenge that limits veterinarians’ understanding of the models on the journey to embrace them. Therefore, the use of AI in this space is mostly still hype. The scientific community is already addressing some of these concerns by developing technology to make the model explainable.77,78 However, it will take some more time for this use case to mature, driven by gradual improvements in data collection and workflow adaptations that accommodate risk and early prediction models into clinical practice.
Use Case 3: Administrative Workload
The burden of bureaucracy or tedious administration work have been described as major generators of veterinary workload,79 with 40% of all veterinarians stating that administration was their number one problem.80
Artificial intelligence has the potential to automate many routine tasks such as scheduling appointments, managing patient records, taking consultation notes, tracking inventory, and handling basic customer inquiries. While developments are generally spearheaded in human medicine,6 there is good progress in veterinary medicine–tailored solutions as well. For note-taking, the AI solutions are a combination of automatic speech recognition algorithms to transcribe conversation into text and natural language understanding algorithms to summarize key concepts of the text into clinical notes. Automatic speech recognition and natural language understanding are well established and widely used technologies that have allowed a speedy commercial deployment of a variety of tools.25,81–84
Artificial intelligence is also used to automate the patient-owner journey by scheduling and keeping track of appointments; automatically populating subjective, objective, assessment, and plan (SOAP) workflows; supporting veterinary decision-making; and sending invoices. This type of solution can be integrated into practice information management systems25,85–88 or can be provided as plug-ins.89,90 Some practice information management systems also have additional AI-based functionality for tracking inventory and measuring clinic performance.
In the context of this use case, the use of AI to communicate with pet owners is a particularly interesting topic. Artificial intelligence–based chatbots allow fast response at any time of the day, thus freeing up time from veterinarians and increasing pet owner engagement.91 They can also be trained to be inclusive and avoid biases so that communication is more effective.92 At the same time, there is a risk that chatbots provide inaccurate information and pet owners are uncomfortable with their messages being owned by a third party.93
Clearly, many AI-based solutions to perform administrative tasks are readily available to veterinarians. This type of AI is currently already creating value, and future work is likely to focus on refinement and improved IT and workflow integration.
Use Case 4: Disease Surveillance
Disease surveillance, a cornerstone of veterinary epidemiology, aims to support the health of pet populations through the identification of outbreaks, monitoring of disease trends, and early detection of emerging threats. One of the main challenges is monitoring animal populations at a sufficiently large scale and high frequency given constraints on data collection, data access, and available resources. Several real-world data sources can be used to this end without the need for additional data collection,94,95 such as primary care and referral clinical records,96 pet insurance databases,97 and internet search data.98 However, the scale and complexity of these data make them unusable without the incorporation of AI.
In terms of supporting disease surveillance, AI can streamline data processing, knowledge extraction, quantification, and prediction.99–101 A nice illustration is the work of the Small Animal Veterinary Surveillance Network101,102 that used primary care clinical records to effectively track an outbreak of acute vomiting in dogs by means of unsupervised modeling and simple visualizations.103,104 Artificial intelligence can also increase the usability of complex, nuanced free-text medical notes through processing via large language models specifically tailored for veterinary data such as DeepTag,105 VetTag,106 and PetBERT.95 Confounding due to biases and missing data, which has the potential to limit reliable decision-making, can be addressed through causal inference methods, such as potential outcomes models, structural equation modeling, Bayesian methods, and directed acyclic graphs.107–112
While there are some good examples for where AI is starting to enable surveillance at scale, there remains scope and opportunity for it to create value. Analysis of real-world data presents as a huge opportunity; however, considerable effort and resources are still required to generate and manage the quality of these data.
Conclusion and Outlook
Artificial intelligence value creation in small animal medicine is clearly not uniform across different veterinary applications, as highlighted by the use cases discussed (Figure 1). For diagnostic imaging and administrative workload, AI applications are mature and in the initial stages of value creation, as shown by commercial solutions being available as stand-alone services or as plug-ins that can be integrated into standard software. Early disease detection and disease surveillance are currently still in the hype phase, with lots of academic activity but limited real-world applications. As the underlying AI technology is quite similar, this poses the question of what drives the maturity differences.
In terms of value creation, it is important to understand that AI as a technology is merely a collection of models that become useful only after being trained on appropriate datasets. Consequently, use case–specific data are key to unlocking the value of AI, and that is where the heterogeneity clearly emerges. For diagnostic imaging, abundant training data can be created with limited effort. Similarly, administrative-workload AI models either do not require specific training (eg, speech recognition models) or can be trained with available datasets (eg, clinical note vocabularies, SOAP manuals, diagnosis textbooks). The less mature early disease detection and disease surveillance use cases heavily depend on electronic health record data, and unfortunately that type of data is not available at a level of quantity and quality that is needed for effective real-world applications. Data challenges are linked to quantity (number of pets), quality (consistency of terminology), sparseness (big gaps due to limited number of visits), and coherence (unique pet identifiers across data sources). Low quantity and large sparseness will result in models unable to generalize properly to all expected scenarios, while low quality and low coherence will result in model outputs that may be inaccurate or even wrong. It thus seems that everything is ready for AI, except the data.
Leveraging AI in small animal medicine will require the profession to step up in terms of electronic health data collection and data sharing. While the technical approach to this has long been described,113 who will drive the execution is less obvious. A collaborative consortium-based approach114 like for human medicine is an option to consider and could involve academics, subject matter experts (eg, the Association for Veterinary Informatics), practicing veterinarians, pharmaceutical companies, kennel clubs, and pet owners. This will be a journey but one that can be a step change in the field.
Acknowledgments
The authors acknowledge Janet Patterson-Kane.
Disclosures
Dr. Albergante, Ciaran O’Flynn, and Dr. De Meyer are employees of Mars Petcare. The RapidRead and Easyvet products discussed in this article are Mars Petcare products.
No AI-assisted technologies were used in the generation of this manuscript.
Funding
The authors have nothing to disclose.
ORCID
L. Albergante https://orcid.org/0000-0001-8151-6989
G. De Meyer https://orcid.org/0000-0002-7073-2691
C. O’Flynn https://orcid.org/0000-0002-8022-1748
References
- 1.↑
Heaven WD. What is AI? MIT Technology Review. Accessed December 3, 2024. https://www.technologyreview.com/2024/07/10/1094475/what-is-artificial-intelligence-ai-definitive-guide/
- 2.↑
Kubat M. An Introduction to Machine Learning. 2nd ed. Springer International Publishing AG; 2017:1-348.
- 3.↑
Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed. Springer Science + Business Media LLC; 2009. doi:10.1007/978-0-387-84858-7
- 4.↑
Epstein Z, Hertzmann A; Investigators of Human Creativity. Art and the science of generative AI. Science. 1979;2023(380):1110-1111.
- 5.↑
Blosch M, Fenn J. Understanding Gartner’s hype cycles. Gartner Research. Accessed July 19, 2024. https://www.gartner.com/en/documents/3887767
- 6.↑
Topol E. Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again. Basic Books Inc; 2019.
- 7.
Appleby RB, Basran PS. Artificial intelligence in veterinary medicine. J Am Vet Med Assoc. 2022;260(8):819-824. doi:10.2460/javma.22.03.0093
- 8.
Basran PS, Appleby RB. The unmet potential of artificial intelligence in veterinary medicine. Am J Vet Res. 2022;83(5):385-392. doi:10.2460/ajvr.22.03.0038
- 9.
Bouchemla F, Akchurin SV, Akchurina IV, Dyulger GP, Latynina ES, Grecheneva AV. Artificial intelligence feasibility in veterinary medicine: a systematic review. Vet World. 2023;16(10):2143-2149. doi:10.14202/vetworld.2023.2143-2149
- 10.
Owens A, Vinkemeier D, Elsheikha H. A review of applications of artificial intelligence in veterinary medicine. Companion Anim. 2023;28(6):78-85. doi:10.12968/coan.2022.0028a
- 11.
Ezanno P, Picault S, Beaunée G, et al. Research perspectives on animal health in the era of artificial intelligence. Vet Res. 2021;52(1):40. doi:10.1186/s13567-021-00902-4
- 12.↑
Akinsulie OC, Idris I, Aliyu VA, et al. The potential application of artificial intelligence in veterinary clinical practice and biomedical research. Front Vet Sci. 2024;11:1347550. doi:10.3389/fvets.2024.1347550
- 13.↑
Cima G. Specialists in short supply: universities, private practices struggle to find certain specialists, blame lack of residency training programs. J Am Vet Med Assoc. 2018;253(8):965-968.
- 14.↑
Bibault JE, Xing L, Giraud P, et al. Radiomics: a primer for the radiation oncologist. Cancer Radiother. 2020;24(5):403-410. doi:10.1016/j.canrad.2020.01.011
- 15.↑
Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60-88. doi:10.1016/j.media.2017.07.005
- 16.↑
Chen B, Jin J, Liu H, et al. Trends and hotspots in research on medical images with deep learning: a bibliometric analysis from 2013 to 2023. Front Artif Intell. 2023;6:1289669. doi:10.3389/frai.2023.1289669
- 17.↑
Banzato T, Wodzinski M, Burti S, Vettore E, Muller H, Zotti A. An AI-based algorithm for the automatic evaluation of image quality in canine thoracic radiographs. Sci Rep. 2023;13(1):17024. doi:10.1038/s41598-023-44089-4
- 18.↑
Burti S, Longhin Osti V, Zotti A, Banzato T. Use of deep learning to detect cardiomegaly on thoracic radiographs in dogs. Vet J. 2020;262:105505. doi:10.1016/j.tvjl.2020.105505
- 19.
Li S, Wang Z, Visser LC, Wisner ER, Cheng H. Pilot study: application of artificial intelligence for detecting left atrial enlargement on canine thoracic radiographs. Vet Radiol Ultrasound. 2020;61(6):611-618. doi:10.1111/vru.12901
- 20.
Boissady E, de La Comble A, Zhu X, Hespel AM. Artificial intelligence evaluating primary thoracic lesions has an overall lower error rate compared to veterinarians or veterinarians in conjunction with the artificial intelligence. Vet Radiol Ultrasound. 2020;61(6):619-627. doi:10.1111/vru.12912
- 21.
Boissady E, De La Comble A, Zhu X, Abbott J, Hespel AM. Comparison of a deep learning algorithm vs. humans for vertebral heart scale measurements in cats and dogs shows a high degree of agreement among readers. Front Vet Sci. 2021;8:764570. doi:10.3389/fvets.2021.764570
- 22.
Banzato T, Wodzinski M, Burti S, et al. Automatic classification of canine thoracic radiographs using deep learning. Sci Rep. 2021;11(1):3964. doi:10.1038/s41598-021-83515-3
- 23.
Fitzke M, Stack C, Dourson A, et al. RapidRead: global deployment of state-of-the-art radiology AI for a large veterinary teleradiology practice. ArXiv. Published online November 9, 2021. doi:10.48550/arXiv.2111.08165
- 24.↑
Parkinson MJ, Fitzke M, Stack JC, Dourson A, inventors; Mars Inc, assignee. Systems and methods for assessing pet radiology images. US patent US20240054637A1. February 15, 2024.
- 25.↑
Danylenko G. A guide to AI tools for veterinary medicine. Full Slice. Accessed August 6, 2024. https://fullslice.agency/blog/a-guide-to-ai-tools-for-veterinary-medicine/
- 26.
Veterinary radiology reimagined. Radimal. Accessed December 4, 2024. https://radimal.ai/
- 27.
RapidRead. Antech Imaging Services. Accessed December 4, 2024. https://info.antechimagingservices.com/rapidread/
- 28.
AI designed for veterinarians. PicoxIA. Accessed December 4, 2024. https://picoxia.com/en/
- 29.↑
Instant point-of-care radiology results. SignalPET. Accessed December 4, 2024. https://www.signalpet.com/
- 30.↑
Basran PS, Appleby RB. What’s in the box? A toolbox for safe deployment of artificial intelligence in veterinary medicine. J Am Vet Med Assoc. 2024;262(8):1090-1098. doi:10.2460/javma.24.01.0027
- 31.
Amann J, Blasimme A, Vayena E, Frey D, Madai VI; Precise4Q Consortium. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med Inform Decis Mak. 2020;20(1):310. doi:10.1186/s12911-020-01332-6
- 32.↑
Arsomngern P, Numcharoenpinij N, Piriyataravet J, et al. Computer-aided diagnosis for lung lesion in companion animals from x-ray images using deep learning techniques. In: 2019 Institute of Electrical and Electronics Engineers (IEEE) 10th International Conference on Awareness Science and Technology. IEEE; 2019:1-6. doi:10.1109/ICAwST.2019.8923126
- 33.↑
Langlotz CP. Will artificial intelligence replace radiologists? Radiol Artif Intell. 2019;1(3):e190058. doi:10.1148/ryai.2019190058
- 34.↑
Maurer LR, Bertsimas D, Bouardi HT, et al. Trauma outcome predictor: an artificial intelligence interactive smartphone tool to predict outcomes in trauma patients. J Trauma Acute Care Surg. 2021;91(1):93-99. doi:10.1097/TA.0000000000003158
- 35.↑
Bertsimas D, Dunn J, Velmahos GC, Kaafarani HMA. Surgical risk is not linear: derivation and validation of a novel, user-friendly, and machine-learning-based Predictive OpTimal Trees in Emergency Surgery Risk (POTTER) calculator. Ann Surg. 2018;268(4):574-583. doi:10.1097/SLA.0000000000002956
- 36.↑
Pate A, Emsley R, Ashcroft DM, Brown B, van Staa T. The uncertainty with using risk prediction models for individual decision making: an exemplar cohort study examining the prediction of cardiovascular disease in English primary care. BMC Med. 2019;17(1):134. doi:10.1186/s12916-019-1368-8
- 37.↑
Richter AN, Khoshgoftaar TM. A review of statistical and machine learning methods for modeling cancer risk using structured clinical data. Artif Intell Med. 2018;90:1-14. doi:10.1016/j.artmed.2018.06.002
- 38.↑
Fröhlich H, Balling R, Beerenwinkel N, et al. From hype to reality: data science enabling personalized medicine. BMC Med. 2018;16(1):150. doi:10.1186/s12916-018-1122-7
- 39.↑
Delpierre C, Lefèvre T. Precision and personalized medicine: what their current definition says and silences about the model of health they promote. Implication for the development of personalized health. Front Sociol. 2023;8:1112159. doi:10.3389/fsoc.2023.1112159
- 40.↑
Ngufor C, Caraballo PJ, O’Byrne TJ, et al. Development and validation of a risk stratification model using disease severity hierarchy for mortality or major cardiovascular event. JAMA Netw Open. 2020;3(7):e208270. doi:10.1001/jamanetworkopen.2020.8270
- 41.↑
Ahmed I, Debray TPA, Moons KGM, Riley RD. Developing and validating risk prediction models in an individual participant data meta-analysis. BMC Med Res Methodol. 2014;14(1):3. doi:10.1186/1471-2288-14-3
- 42.↑
Electronic health records in veterinary medicine. IEEE Pulse. January 23, 2017. Accessed July 30, 2024. https://www.embs.org/pulse/articles/electronic-health-records-veterinary-medicine/
- 43.
Krone LM, Brown CM, Lindenmayer JM. Survey of electronic veterinary medical record adoption and use by independent small animal veterinary medical practices in Massachusetts. J Am Vet Med Assoc. 2014;245(3):324-332. doi:10.2460/javma.245.3.324
- 44.
Reimer Fender K. Paperless practice at Banfield, the pet hospital. dvm360. Accessed December 4, 2024. https://www.dvm360.com/view/paperless-practice-banfield-pet-hospital
- 46.↑
Saito EK, Moore GE, Morrison J, Spofford N, Yang M. Younger dogs of specific breeds are more likely to experience a rare adverse event after administration of extended-release injectable moxidectin heartworm preventive. J Am Vet Med Assoc. 2023;261(11):1666-1672. doi:10.2460/javma.23.04.0217
- 47.↑
Pegram C, Diaz-Ordaz K, Brodbelt DC, et al. Later-age neutering causes lower risk of early-onset urinary incontinence than early neutering: a VetCompass target trial emulation study. PLoS One. 2024;19(7):e0305526. doi:10.1371/journal.pone.0305526
- 48.↑
Hall EJ, Carter AJ, Chico G, et al. Risk factors for severe and fatal heat-related illness in UK dogs—a VetCompass study. Vet Sci. 2022;9(5):231. doi:10.3390/vetsci9050231
- 50.↑
Kim Y, Kim J, Kim S, Youn H, Choi J, Seo K. Machine learning-based risk prediction model for canine myxomatous mitral valve disease using electronic health record data. Front Vet Sci. 2023;10:1189157. doi:10.3389/fvets.2023.1189157
- 51.↑
Bir C, Ortez M, Olynk Widmar NJ, Wolf CA, Hansen C, Ouedraogo FB. Familiarity and use of veterinary services by US resident dog and cat owners. Animals (Basel). 2020;10(3):483. doi:10.3390/ani10030483
- 52.↑
Kelly DL, Smith CL. Bayesian inference in probabilistic risk assessment—the current state of the art. Reliab Eng Syst Saf. 2009;94(2):628-643. doi:10.1016/j.ress.2008.07.002
- 53.
Dunbar D, Babayan SA, Krumrie S, Haining H, Hosie MJ, Weir W. Assessing the feasibility of applying machine learning to diagnosing non-effusive feline infectious peritonitis. Sci Rep. 2024;14(1):2517. doi:10.1038/s41598-024-52577-4
- 54.↑
Cummings CO, Mitchell MA, Perry SM, et al. Bayesian decision analysis: an underutilized tool in veterinary medicine. Animals (Basel). 2022;12(23):3414. doi:10.3390/ani12233414
- 55.↑
Hurley JC. How to apply structural equation modelling to infectious diseases concepts. Clin Microbiol Infect. 2022;28(12):1567-1571. doi:10.1016/j.cmi.2022.05.028
- 56.↑
Bello NM, Ferreira VC, Gianola D, Rosa GJM. Conceptual framework for investigating causal effects from observational data in livestock. J Anim Sci. 2018;96(10):4045-4062. doi:10.1093/jas/sky277
- 57.↑
Baker LA, Momen M, Chan K, et al. Bayesian and machine learning models for genomic prediction of anterior cruciate ligament rupture in the canine model. G3 (Bethesda). 2020;10(8):2619-2628.
- 58.↑
Kokkinos Y, Morrison J, Bradley R, et al. An early prediction model for canine chronic kidney disease based on routine clinical laboratory tests. Sci Rep. 2022;12(1):14489. doi:10.1038/s41598-022-18793-6
- 59.
Renard J, Faucher MR, Combes A, Concordet D, Reynolds BS. Machine-learning algorithm as a prognostic tool in non-obstructive acute-on-chronic kidney disease in the cat. J Feline Med Surg. 2021;23(12):1140-1148. doi:10.1177/1098612X211001273
- 60.
Bradley R, Tagkopoulos I, Kim M, et al. Predicting early risk of chronic kidney disease in cats using routine clinical laboratory tests and machine learning. J Vet Intern Med. 2019;33(6):2644-2656. doi:10.1111/jvim.15623
- 61.↑
Biourge V, Delmotte S, Feugier A, Bradley R, McAllister M, Elliott J. An artificial neural network-based model to predict chronic kidney disease in aged cats. J Vet Intern Med. 2020;34(5):1920-1931. doi:10.1111/jvim.15892
- 62.↑
Schofield I, Brodbelt DC, Niessen SJM, et al. Development and internal validation of a prediction tool to aid the diagnosis of Cushing’s syndrome in dogs attending primary-care practice. J Vet Intern Med. 2020;34(6):2306-2318. doi:10.1111/jvim.15851
- 63.↑
Fraser MA, McNeil PE, Girling SJ. Prediction of future development of canine atopic dermatitis based on examination of clinical history. J Small Anim Pract. 2008;49(3):128-132. doi:10.1111/j.1748-5827.2007.00439.x
- 64.↑
Flegel T, Neumann A, Holst AL, et al. Machine learning algorithms predict canine structural epilepsy with high accuracy. Front Vet Sci. 2024;11:1406107. doi:10.3389/fvets.2024.1406107
- 65.↑
Reynolds CA, Brown DC, Rush JE, et al. Prediction of first onset of congestive heart failure in dogs with degenerative mitral valve disease: the PREDICT cohort study. J Vet Cardiol. 2012;14(1):193-202. doi:10.1016/j.jvc.2012.01.008
- 66.↑
Doguc O, Bilgi SB, Cagdas S, et al. A decision support system for detecting FIP disease in cats based on machine learning methods. In: García Márquez FP, Jamil A, Hameed AA, et al., eds. Emerging Trends and Applications in Artificial Intelligence. Springer Nature Switzerland; 2024:176-186. doi:10.1007/978-3-031-56728-5_16
- 67.↑
Hyperthyroid cats and the IDEXX SDMA test. IDEXX UK. Accessed December 4, 2024. https://www.idexx.co.uk/en-gb/veterinary/reference-laboratories/sdma/hyperthyroid-cats-and-idexx-sdma-test/
- 68.↑
Fowlie S, Spence S, Roberts E, Ramsey IK. Electrolyte measurements differ between point-of-care and reference analysers in dogs with hypoadrenocorticism. J Small Anim Pract. 2020;61(10):599-608. doi:10.1111/jsap.13205
- 69.↑
Study reveals differing electrolyte measurements between point-of-care and reference analysers in dogs with hypoadrenocorticism. Improve Veterinary Practice. December 4, 2024. https://www.veterinary-practice.com/2020/study-reveals-differing-electrolyte-measurements-between-point-of-care-and-reference-analysers-in-dogs-with-hypoadrenocorticism
- 70.↑
Farah L, Murris JM, Borget I, Guilloux A, Martelli NM, Katsahian SIM. Assessment of performance, interpretability, and explainability in artificial intelligence-based health technologies: what healthcare stakeholders need to know. Mayo Clin Proc Digit Health. 2023;1(2):120-138. doi:10.1016/j.mcpdig.2023.02.004
- 71.
Reddy S. Explainability and artificial intelligence in medicine. Lancet Digit Health. 2022;4(4):e214-e215. doi:10.1016/S2589-7500(22)00029-2
- 72.↑
Ali S, Abuhmed T, El-Sappagh S, et al. Explainable artificial intelligence (XAI): what we know and what is left to attain trustworthy artificial intelligence. Inf Fusion. 2023;99:101805. doi:10.1016/j.inffus.2023.101805
- 73.↑
Ebrahimian S, Kalra MK, Agarwal S, et al. FDA-regulated AI algorithms: trends, strengths, and gaps of validation studies. Acad Radiol. 2022;29(4):559-566. doi:10.1016/j.acra.2021.09.002
- 74.↑
Habib AR, Gross CP. FDA regulations of AI-driven clinical decision support devices fall short. JAMA Intern Med. 2023;183(12):1401-1402. doi:10.1001/jamainternmed.2023.5006
- 75.↑
Domalpally A, Channa R. Real-world validation of artificial intelligence algorithms for ophthalmic imaging. Lancet Digit Health. 2021;3(8):e463-e464. doi:10.1016/S2589-7500(21)00140-0
- 76.↑
Chen F, Wang L, Hong J, Jiang J, Zhou L. Unmasking bias in artificial intelligence: a systematic review of bias detection and mitigation strategies in electronic health record-based models. J Am Med Inform Assoc. 2024;31(5):1172-1183. doi:10.1093/jamia/ocae060
- 77.↑
Burton J, Al Moubayed N. SHAP explanations for multimodal text-tabular models. Res Sq. Published online October 27, 2023. doi:10.21203/rs.3.rs-3405528/v1
- 78.↑
Burton J, Farrell S, Mäntylä Noble PJ, Al Moubayed N. Explainable text-tabular models for predicting mortality risk in companion animals. Sci Rep. 2024;14(1):14217. doi:10.1038/s41598-024-64551-1
- 79.↑
Spendelow J, Cripwell C, Stott R, et al. Workplace stressor factors, profiles and the relationship to career stage in UK veterinarians, veterinary nurses and students. Vet Med Sci. 2024;10(4):e1547. doi:10.1002/vms3.1547
- 80.↑
McNeill C, McNeill E. Key challenges in the veterinary profession. Royal Canin. Accessed July 19, 2024. https://vetfocus.royalcanin.com/en/scientific/key-challenges-in-the-veterinary-profession
- 82.
Automatic notetaking for vets. VetNotes. Accessed December 4, 2024. https://vetnotes.com/
- 83.
AI digital scribe for busy veterinarians. ScribbleVet. Accessed December 4, 2024. https://www.scribblevet.com/
- 85.↑
Mackay S. Five best veterinary practice management software with client communication and engagement in 2023. PetsApp. Accessed August 8, 2024. https://petsapp.com/blog/best-veterinary-practice-management-software
- 86.
The all-in-one patient journey bundle for veterinary practices. PetDesk. Accessed December 4, 2024. https://petdesk.com/
- 87.
Veterinary engagement and communication platform. PetsApp. Accessed December 4, 2024. https://petsapp.com/
- 91.↑
Masood S, Al Bashrawi MA, Khan MA, Nazir A. Exploring ChatGPT applications in healthcare: a comprehensive overview. Vis Comput. Published online November 27, 2024. doi:10.1007/s00371-024-03697-5
- 92.↑
Grassini E, Buzzi M, Leporini B, Vozna A. A systematic review of chatbots in inclusive healthcare: insights from the last 5 years. Univers Access Inf Soc. Published online May 10, 2024. doi:10.1007/s10209-024-01118-x
- 93.↑
Esmaeilzadeh P. Use of AI-based tools for healthcare purposes: a survey study from consumers’ perspectives. BMC Med Inform Decis Mak. 2020;20(1):170. doi:10.1186/s12911-020-01191-1
- 94.↑
O’Neill DG, Church DB, McGreevy PD, Thomson PC, Brodbelt DC. Prevalence of disorders recorded in cats attending primary-care veterinary practices in England. Vet J. 2014;202(2):286-291. doi:10.1016/j.tvjl.2014.08.004
- 95.↑
Farrell S, Appleton C, Noble PM, Al Moubayed N. PetBERT: automated ICD-11 syndromic disease coding for outbreak detection in first opinion veterinary electronic health records. Sci Rep. 2023;13(1):18015. doi:10.1038/s41598-023-45155-7
- 96.↑
Paynter AN, Dunbar MD, Creevy KE, Ruple A. Veterinary big data: when data goes to the dogs. Animals (Basel). 2021;11(7):1872. doi:10.3390/ani11071872
- 97.↑
Burnett E, Brand CL, O’Neill DG, et al. How much is that Doodle in the window? Exploring motivations and behaviours of UK owners acquiring designer crossbreed dogs (2019-2020). Canine Med Genet. 2022;9(1):8. doi:10.1186/s40575-022-00120-x
- 98.↑
Woolley CSC, Handel IG, Bronsvoort BM, Schoenebeck JJ, Clements DN. Surveillance of a vomiting outbreak in dogs in the UK using owner-derived and internet search data. Vet Rec. 2021;189(9):e308. doi:10.1002/vetr.308
- 99.↑
Doherr MG, Audigé L. Monitoring and surveillance for rare health-related events: a review from the veterinary perspective. Philos Trans R Soc Lond B Biol Sci. 2001;356(1411):1097-1106. doi:10.1098/rstb.2001.0898
- 100.
Salman MD. Surveillance tools and strategies for animal diseases in a shifting climate context. Anim Health Res Rev. 2013;14(2):147-150. doi:10.1017/S1466252313000121
- 101.↑
Sánchez-Vizcaíno F, Jones PH, Menacere T, et al. Small animal disease surveillance. Vet Rec. 2015;177(23):591-594. doi:10.1136/vr.h6174
- 102.↑
Radford A, Tierney A, Coyne KP, et al. Developing a network for small animal disease surveillance. Vet Rec. 2010;167(13):472-474. doi:10.1136/vr.c5180
- 103.↑
Radford AD, Singleton DA, Jewell C, et al. Outbreak of severe vomiting in dogs associated with a canine enteric coronavirus, United Kingdom. Emerg Infect Dis. 2021;27(2):517-528. doi:10.3201/eid2702.202452
- 104.↑
Noble PM, Appleton C, Radford AD, Nenadic G. Using topic modelling for unsupervised annotation of electronic health records to identify an outbreak of disease in UK dogs. PLoS One. 2021;16(12):e0260402. doi:10.1371/journal.pone.0260402
- 105.↑
Nie A, Zehnder A, Page RL, et al. DeepTag: inferring diagnoses from veterinary clinical notes. NPJ Digit Med. 2018;1(1):60. doi:10.1038/s41746-018-0067-8
- 106.↑
Zhang Y, Nie A, Zehnder A, Page RL, Zou J. VetTag: improving automated veterinary diagnosis coding via large-scale language modeling. NPJ Digit Med. 2019;2(1):35. doi:10.1038/s41746-019-0113-1
- 107.↑
Kratzer G, Lewis FI, Willi B, et al. Bayesian network modeling applied to feline calicivirus infection among cats in Switzerland. Front Vet Sci. 2020;7:73. doi:10.3389/fvets.2020.00073
- 108.
Cox R, Revie CW, Hurnik D, Sanchez J. Use of Bayesian Belief Network techniques to explore the interaction of biosecurity practices on the probability of porcine disease occurrence in Canada. Prev Vet Med. 2016;131:20-30. doi:10.1016/j.prevetmed.2016.06.015
- 109.
Dohoo IR. Quantitative epidemiology: progress and challenges. Prev Vet Med. 2008;86(3-4):260-269. doi:10.1016/j.prevetmed.2008.02.012
- 110.
Jayasundara VK, Khatibi A, Tham J. Application of structural equation modeling (SEM) for free-roaming dog population management. Adv Anim Vet Sci. 2024;12(7):1230-1238. doi:10.17582/journal.aavs/2024/12.7.1230.1238
- 111.
Powers JA, Chiu ES, Kraberger SJ, et al. Feline leukemia virus (FeLV) disease outcomes in a domestic cat breeding colony: relationship to endogenous FeLV and other chronic viral infections. J Virol. 2018;92(18):e00649-18. doi:10.1128/JVI.00649-18
- 112.↑
Tennant PWG, Murray EJ, Arnold KF, et al. Use of directed acyclic graphs (DAGs) to identify confounders in applied health research: review and recommendations. Int J Epidemiol. 2021;50(2):620-632. doi:10.1093/ije/dyaa213
- 113.↑
Wilkinson MD, Dumontier M, Aalbersberg IJ, et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data. 2016;3(1):160018. doi:10.1038/sdata.2016.18
- 114.↑
Martínez-García A, Alvarez-Romero C, Román-Villarán E, Bernabeu-Wittel M, Luis Parra-Calderón C. FAIR principles to improve the impact on health research management outcomes. Heliyon. 2023;9(5):e15733. doi:10.1016/j.heliyon.2023.e15733
