PetBERT: automated ICD-11 syndromic disease coding for outbreak detection in first opinion veterinary electronic health records.

Effective public health surveillance requires consistent monitoring of disease signals such that researchers and decision-makers can react dynamically to changes in disease occurrence. However, whilst surveillance initiatives exist in production animal veterinary medicine, comparable frameworks for companion animals are lacking. First-opinion veterinary electronic health records (EHRs) have the potential to reveal disease signals and often represent the initial reporting of clinical syndromes in animals presenting for medical attention, highlighting their possible significance in early disease detection. Yet despite their availability, there are limitations surrounding their free text-based nature, inhibiting the ability for national-level mortality and morbidity statistics to occur. This paper presents PetBERT, a large language model trained on over 500 million words from 5.1 million EHRs across the UK. PetBERT-ICD is the additional training of PetBERT as a multi-label classifier for the automated coding of veterinary clinical EHRs with the International Classification of Disease 11 framework, achieving F1 scores exceeding 83% across 20 disease codings with minimal annotations. PetBERT-ICD effectively identifies disease outbreaks, outperforming current clinician-assigned point-of-care labelling strategies up to 3 weeks earlier. The potential for PetBERT-ICD to enhance disease surveillance in veterinary medicine represents a promising avenue for advancing animal health and improving public health outcomes.

Seasonality and other risk factors for fleas infestations in domestic dogs and cats.

Fleas in the genus Ctenocephalides are the most clinically important parasitic arthropods of dogs and cats worldwide yet risk factors that might increase the risk of infestation in small animals remains unclear. Here we developed a supervised text mining approach analysing key aspects of flea epidemiology using electronic health records from domestic cats and dogs seen at a sentinel network of 191 voluntary veterinary practices across Great Britain between March 2014 and July 2020. Our methods identified fleas as likely to have been present during 22,276 of 1,902,016 cat consultations (1.17%) and 12,168 of 4,844,850 dog consultations (0.25%). Multivariable logistic regression modelling found that animals originating from areas of least deprivation were associated with 50% reductions in odds of veterinary-recorded flea infestation compared to the most deprived regions in England. Age of the animal was significantly associated with flea presentation in both cats and dogs, with cases peaking before animals reached 12 months. Cases were recorded through each study years, peaking between July and October, with fluctuations between each year. Our findings can be used towards healthcare messaging for veterinary practitioners and owners.

Using topic modelling for unsupervised annotation of electronic health records to identify an outbreak of disease in UK dogs.

A key goal of disease surveillance is to identify outbreaks of known or novel diseases in a timely manner. Such an outbreak occurred in the UK associated with acute vomiting in dogs between December 2019 and March 2020. We tracked this outbreak using the clinical free text component of anonymised electronic health records (EHRs) collected from a sentinel network of participating veterinary practices. We sourced the free text (narrative) component of each EHR supplemented with one of 10 practitioner-derived main presenting complaints (MPCs), with the 'gastroenteric' MPC identifying cases involved in the disease outbreak. Such clinician-derived annotation systems can suffer from poor compliance requiring retrospective, often manual, coding, thereby limiting real-time usability, especially where an outbreak of a novel disease might not present clinically as a currently recognised syndrome or MPC. Here, we investigate the use of an unsupervised method of EHR annotation using latent Dirichlet allocation topic-modelling to identify topics inherent within the clinical narrative component of EHRs. The model comprised 30 topics which were used to annotate EHRs spanning the natural disease outbreak and investigate whether any given topic might mirror the outbreak time-course. Narratives were annotated using the Gensim Library LdaModel module for the topic best representing the text within them. Counts for narratives labelled with one of the topics significantly matched the disease outbreak based on the practitioner-derived 'gastroenteric' MPC (Spearman correlation 0.978); no other topics showed a similar time course. Using artificially injected outbreaks, it was possible to see other topics that would match other MPCs including respiratory disease. The underlying topics were readily evaluated using simple word-cloud representations and using a freely available package (LDAVis) providing rapid insight into the clinical basis of each topic. This work clearly shows that unsupervised record annotation using topic modelling linked to simple text visualisations can provide an easily interrogable method to identify and characterise outbreaks and other anomalies of known and previously un-characterised diseases based on changes in clinical narratives.

Gastric mucosal pathology in Belgian Shepherd dogs with and without clinical signs of gastric disease.

BACKGROUND: Gastric carcinoma (GC) is uncommon in dogs, except in predisposed breeds such as Belgian Shepherd dogs (BSD) of the Tervuren and Groenendael varieties. When GC is diagnosed in dogs it is often late in the disease, resulting in a poorer prognosis. The aim of this prospective clinical study was to investigate possible associations of gastric mucosal pathologies with clinical signs, laboratory test results and GC in BSD. An online survey gathered epidemiological data to generate potential risk factors for vomiting as the predominant gastric clinical sign, and supported patient recruitment for endoscopy. Canine Chronic Enteropathy Clinical Activity Index (CCECAI) score and signs of gastroesophageal reflux (GER) were used to allocate BSD older than five years to either Group A, with signs of gastric disease, or Group B, without signs. Findings in the clinical history, laboratory tests and gastric histopathology of endoscopic biopsies were statistically analysed in search of associations. RESULTS: The online survey included 232 responses. Logistic regression analysis recognized an association of vomiting with gagging, poor appetite and change in attitude. Recruitment for endoscopy included 16 BSD in Group A (mean age 9.1 ± 1.8 years, mean CCECAI = 3.1 ± 2.2 and signs of GER); and 11 in Group B (mean age 9.8 ± 1.4 years, CCECAI = 0, no signs of GER). Seven (25.9%) of the 27 BSD (Group A 4/16, Group B 3/11) had leukopenia. Serum C-reactive protein tended to be increased with more advanced GC (P = 0.063). Frequency of GC, mucosal atrophy, mucous metaplasia, or glandular dysplasia did not differ between groups. GC was frequently diagnosed (6/27), even without clinical signs (2/11). The odds ratio for vomiting (OR = 9.9; P = 0.016) was increased only when glandular dysplasia was present. GC was associated with mucous metaplasia (P = 0.024) and glandular dysplasia (P = 0.006), but not with mucosal atrophy (P = 1). CONCLUSIONS: GC can develop as an occult disease, associated with metaplasia and dysplasia of the gastric mucosa. Suggestive clinical signs, notably vomiting, should warrant timely endoscopy in BSD. Extensive endoscopic screening of asymptomatic dogs remains, however, unrealistic. Therefore, biomarkers of mucosal pathology preceding clinical illness are needed to support an indication for endoscopy and enable early diagnosis of GC.