Related Articles

The molecular fingerprints of 1,349 isolates of Mycobacterium bovis received between 1979 and August 2000 at Agence Française de Sécurité Sanitaire des Aliments (Afssa) have been obtained by spoligotyping. The majority of the isolates (1,266) were obtained from cattle living in France. An apparently high level of heterogeneity was observed between isolates. One hundred sixty-one spoligotypes were observed in total, of which 153 were from French isolates. The two predominant spoligotypes, designated BCG-like and GB54, accounted for 26 and 12% of the isolates, respectively. In addition, 84% of the spoligotypes were found fewer than 10 times. Analysis of the results by clustering and parsimony-based algorithms revealed that the majority of the spoligotypes were closely related. The predominant spoligotype was identical to that of the vaccine strain Mycobacterium bovis BCG, which was isolated in France at the end of the 19th century. Some spoligotypes were closely associated with restricted geographical areas. Interestingly, some spoligotypes, which were frequently observed in France, were also observed in neighboring countries. Conversely, few spoligotypes were common to France and England, and those that were shared were observed at very different frequencies. This last point illustrates the potential role for an international data bank, which could help trace the spread of M. bovis across national borders.

Three species of mycoplasma have been established as being of importance as causes of pneumonia in housed calves, based on pathogenicity studies and frequency of association with the disease. These three species are Mycoplasma bovis, M. dispar, and Ureaplasma diversum. M. bovis is the most pathogenic of these species but the disease outbreaks with which it is associated are sporadic. M. dispar is regularly isolated from pneumonic calves but is also found causing mild superficial and asymptomatic infections of the respiratory mucosa. The bovine ureaplasmas are serologically complex. They are distinct from ureaplasmas isolated from other non-ruminants by PAGE analysis, G + C content of DNA, and serology. A second species within the genus ureaplasma has been proposed to accommodate the bovine ureaplasmas, U. diversum. Control of mycoplasma respiratory infections of cattle based on immunization might be possible. Calves have been immunized against M. bovis and immunity has been related to antibody in the lung. M. dispar appears less immunogenic in calves than M. bovis and this may contribute to its pathogenicity.

Mycobacterium avium subsp. paratuberculosis (basonym M. paratuberculosis) is the etiologic agent of a severe gastroenteritis in ruminants known as Johne's disease. Economic losses to the cattle industry in the United States are staggering, reaching $1.5 billion annually. A potential pathogenic role in humans in the etiology of Crohn's disease is under investigation. In this article, we review the epidemiology, pathogenesis, diagnostics, and disease control measures of this important veterinary pathogen. We emphasize molecular genetic aspects including the description of markers used for strain identification, diagnostics, and phylogenetic analysis. Recent important advances in the development of animal models and genetic systems to study M. paratuberculosis virulence determinants are also discussed. We conclude with proposals for the applications of these models and recombinant technology to the development of diagnostic, control, and therapeutic measures.

The demography of bovine infections caused by Mycobacterium avium subspecies paratuberculosis (MAP) in Ireland is poorly defined. The objective of this study was to describe the demographics of cattle positive to MAP on faecal culture, based on submissions to the Cork Regional Veterinary Laboratory (Cork RVL) from 1994 to 2006. The study focused on all available faecal samples from adult cattle with non-responsive chronic diarrhoea that were submitted by private veterinary practitioners to Cork RVL for MAP culture. For each MAP-positive by faecal culture animal, data were collated from Cork RVL and Cattle Movement Monitoring Scheme (CMMS) records. Johne's disease (JD) was confirmed in 110 animals from 86 herds by the Cork RVL between 1994 and 2006, with a rate of positive cases between 15% and 18% over last four years of the study. Two breeds (Holstein/Friesian or Limousin) made up 78% of submissions. Movements were assessed for the 57 study animals with available movement information, 90% died within one year of the test and 26% tested positive in the herd they were born into. The study provides preliminary information about movement trends and demographics of animals with MAP positive submissions. Although the study area is restricted, it includes the most intensive (and economically-important) dairy region in Ireland. The demographics of JD infection from the study area are in agreement with international reports. Further work is required to determine demographic trends, incidence and prevalence of JD throughout Ireland. It is hoped this work may contribute to the development of a surveillance strategy for MAP by regional veterinary laboratories.

Strains formerly identified as Streptococcus bovis were allotted to two groups by sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of whole-cell proteins. Strains from humans with infections, mostly patients with endocarditis, and strains from pigeons with septicemia clustered with the recently described species Streptococcus gallolyticus. The original S. bovis type strain and strains exclusively from ruminants formed the second cluster. The findings indicate that S. gallolyticus is more likely to be involved in human and animal infections than S. bovis. Growth characteristics and several biochemical reactions were found to be useful in the differentiation of S. gallolyticus from S. bovis.

Bartonella spp. are small hemotropic bacteria infecting mammals. Four Bartonella species have been recently described in cattle and wild ruminants. To date, the biology and possible pathogenic role of Bartonella species isolated from ruminants are poorly understood. Therefore, a dairy herd of 448 cows and heifers was surveyed in order to establish the prevalence of Bartonella bovis and B. chomelii infections, the level of bacteremia, and the relationship between bacteremia and age or pregnancy status. The putative impact of Bartonella infection on production performance (individual milk cell count, milk yield) and reproductive status (success of artificial insemination [AI], placental retention, embryonic death, and abortion) was also assessed. The overall mean prevalence of B. bovis bacteremia was 59%, with the highest prevalence in heifers (92.5%). No B. chomelii was isolated, and 95% (114/120) of the B. bovis strains isolated and tested by PCR-restriction fragment length polymorphism belonged to type I. The level of bacteremia was higher in pregnant cows than in nonpregnant cows (P = 0.05), and the level of bacteremia rose during the last two-thirds of gestation (P < 0.001). There was no correlation between bacteremia and milk yield, individual milk cell count, success of first AI, interval between two calvings, or incidence of abortion and embryonic death. The interval from calving to first AI was shorter and the incidence of placental retention was lower in bacteremic animals than in nonbacteremic ones (P = 0.03 and P = 0.01, respectively).

Surveillance systems tracking health patterns in animals have potential for early warning of infectious disease in humans, yet there are many challenges that remain before this can be realized. Specifically, there remains the challenge of detecting early warning signals for diseases that are not known or are not part of routine surveillance for named diseases. This paper reports on the development of a hidden Markov model for analysis of frontline veterinary sentinel surveillance data from Sri Lanka. Field veterinarians collected data on syndromes and diagnoses using mobile phones. A model for submission patterns accounts for both sentinel-related and disease-related variability. Models for commonly reported cattle diagnoses were estimated separately. Region-specific weekly average prevalence was estimated for each diagnoses and partitioned into normal and abnormal periods. Visualization of state probabilities was used to indicate areas and times of unusual disease prevalence. The analysis suggests that hidden Markov modelling is a useful approach for surveillance datasets from novel populations and/or having little historical baselines.

Glucose-nonfermenting gram-negative bacilli (NFB) have been recognized recently as opportunistic pathogens of humans. With few exceptions, strains of NFB have not been considered important enough to be identified when isolated from animals. In this study, all NFB isolated during a 1-year period in a clinical veterinary microbiology laboratory were identified to determine their prevalence. Of the 347 strains of NFB obtained, the most common species were Pseudomonas aeruginosa, Acinetobacter calcoaceticus, Bordetella bronchiseptica, and Pseudomonas pseudoalcaligenes. Of all clinical veterinary specimens submitted for cultures, 10% contained nonfermenters.

Restriction fragment length polymorphism analysis of rRNA genes was employed to genetically compare Fusobacterium necrophorum subsp. necrophorum and F. necrophorum subsp. funduliforme isolates from multiple abscesses of the same liver and isolates from liver abscesses, the ruminal wall, and ruminal contents from the same animal. Four livers with multiple abscesses and samples of ruminal contents, ruminal walls, and liver abscesses were collected from 11 cattle at slaughter. F. necrophorum was isolated from all liver abscesses, nine ruminal walls, and six ruminal content samples. Chromosomal DNA of the isolates was extracted and single or double digested with restriction endonucleases (EcoRI, EcoRV, SalI, and HaeIII); then restriction fragments were hybridized with a digoxigenin-labeled cDNA probe transcribed from a mixture of 16S and 23S rRNAs from Escherichia coli. EcoRI alone or in combination with EcoRV yielded the most discriminating ribopatterns for comparison. Within the subspecies multiple isolates from the same liver were indistinguishable based on the ribopattern obtained with EcoRI. The hybridization patterns of liver abscess isolates were concordant with those of the corresponding isolates from ruminal walls in eight of nine sets of samples. None of the six ruminal content isolates matched either the liver abscess isolates or the ruminal wall isolates. The genetic similarity between the isolates from liver abscesses and ruminal walls supports the hypothesis that F. necrophorum isolates of liver abscesses originate from the rumen.

Background

The evolution of mycoplasmas from a common ancestor with Firmicutes has been characterized not only by genome down-sizing but also by horizontal gene transfer between mycoplasma species sharing a common host. The mechanisms of these gene transfers remain unclear because our knowledge of the mycoplasma mobile genetic elements is limited. In particular, only a few plasmids have been described within the Mycoplasma genus.

Results

We have shown that several species of ruminant mycoplasmas carry plasmids that are members of a large family of elements and replicate via a rolling-circle mechanism. All plasmids were isolated from species that either belonged or were closely related to the Mycoplasma mycoides cluster; none was from the Mycoplasma bovis-Mycoplasma agalactiae group. Twenty one plasmids were completely sequenced, named and compared with each other and with the five mycoplasma plasmids previously reported. All plasmids share similar size and genetic organization, and present a mosaic structure. A peculiar case is that of the plasmid pMyBK1 from M. yeatsii; it is larger in size and is predicted to be mobilizable. Its origin of replication and replication protein were identified. In addition, pMyBK1 derivatives were shown to replicate in various species of the M. mycoides cluster, and therefore hold considerable promise for developing gene vectors. The phylogenetic analysis of these plasmids confirms the uniqueness of pMyBK1 and indicates that the other mycoplasma plasmids cluster together, apart from the related replicons found in phytoplasmas and in species of the clade Firmicutes.

Conclusions

Our results unraveled a totally new picture of mycoplasma plasmids. Although they probably play a limited role in the gene exchanges that participate in mycoplasma evolution, they are abundant in some species. Evidence for the occurrence of frequent genetic recombination strongly suggests they are transmitted between species sharing a common host or niche.

Background

Bovine tuberculosis (TB) caused by Mycobacterium bovis is primarily a disease of ruminants, particularly cattle (Bos primigenius) and buffalo (Syncerus caffer), and is endemic in most developing countries. To date, studies done in Uganda have documented the prevalence of M. bovis in cattle, humans and wild life, in addition to non-tuberculous mycobacteria in pigs. Pigs are increasingly becoming an important component of the livestock sector and share the human ecosystem in rural Uganda. It is therefore of public health interest that they are not a source of human infections. As a follow up to previously published findings on mycobacteria in pigs, this study was aimed at investigating the occurrence and molecular characteristics of M. bovis detected in slaughter pigs in Mubende district, Uganda. One hundred fifty mesenteric lymph nodes with lesions suggestive of mycobacterial infections were collected from approximately one thousand slaughtered pigs in Mubende district over a period of five months. The isolation and identification of M. bovis was done using conventional mycobacteriological methods. Mycobacteria belonging to the Mycobacterium tuberculosis complex (MTC) were identified to species level using deletion analysis. Molecular typing was done using Spoligotyping and MIRU-VNTR analysis. Molecular data were analysed and interpreted using MIRU-VNTR plus, SpolDB4.0 and the Mycobacterium bovis spoligo database.

Results

Of the examined animals, one boar and two sows from Madudu Sub County were infected with M. bovis which presented as lesions of a deep yellow colour and a grit-like texture in the mesenteric lymph nodes. This represents 2% (3/150) of the lymph nodes where lesions suggestive of mycobacterial infections were detected. Molecular analysis revealed that the isolates from the infected pigs showed identical MIRU-VNTR profile and spoligotype (SB1469).

Conclusions

This is the first study documenting the occurrence of M. bovis in slaughter pigs in Uganda, revealing that one in fifty slaughter pigs with suspected lesions in mesenteric lymph nodes were infected. Molecular analysis revealed that the isolates were identical, showing a spoligotype previously reported from humans and cattle in the north eastern part of the Uganda cattle corridor. This finding is of public health importance, therefore there is a need for close cooperation between medical and veterinary professionals in designing and implementing control and prevention measures that safeguard the public from this potential source of zoonotic TB in Uganda.

BSE can infect small ruminants and could be misdiagnosed as scrapie.

Transmissible spongiform encephalopathies are fatal neurodegenerative diseases that include variant Creutzfeldt-Jakob disease in humans, scrapie in small ruminants, and bovine spongiform encephalopathy (BSE) in cattle. Scrapie is not considered a public health risk, but BSE has been linked to variant Creutzfeldt-Jakob disease. Small ruminants are susceptible to BSE, and in 2005 BSE was identified in a farmed goat in France. We confirm another BSE case in a goat in which scrapie was originally diagnosed and retrospectively identified as suspected BSE. The prion strain in this case was further characterized by mouse bioassay after extraction from formaldehyde-fixed brain tissue embedded in paraffin blocks. Our data show that BSE can infect small ruminants under natural conditions and could be misdiagnosed as scrapie. Surveillance should continue so that another outbreak of this zoonotic transmissible spongiform encephalopathy can be prevented and public health safeguarded.

Mycoplasma bovis is a small, cell wall-less bacterium that contributes to a number of chronic inflammatory diseases in both dairy and feedlot cattle, including mastitis and bronchopneumonia. Numerous reports have implicated M. bovis in the activation of the immune system, while at the same time inhibiting immune cell proliferation. However, it is unknown whether the specific immune-cell population M. bovis is capable of attaching to and potentially invading. Here, we demonstrate that incubation of M. bovis Mb1 with bovine peripheral blood mononuclear cells (PBMC) resulted in a significant reduction in their proliferative responses while still remaining viable and capable of gamma interferon secretion. Furthermore, we show that M. bovis Mb1 can be found intracellularly (suggesting a role for either phagocytosis or attachment/invasion) in a number of select bovine PBMC populations (T cells, B cells, monocytes, γδ T cells, dendritic cells, NK cells, cytotoxic T cells, and T-helper cells), as well as red blood cells, albeit it at a significantly lower proportion. M. bovis Mb1 appeared to display three main patterns of intracellular staining: diffuse staining, an association with the intracellular side of the cell membrane, and punctate/vacuole-like staining. The invasion of circulating immune cells and erythrocytes could play an important role in disease pathogenesis by aiding the transport of M. bovis from the lungs to other sites.

Neisseria cuniculi was isolated, between March 1987 and March 1989, from 38 cases of respiratory disease in small and large ruminants. In all but five cases N. cuniculi was cultured together with other potential respiratory pathogens. A survey was conducted to assess the prevalence of N. cuniculi in the pharyngeal region of Merino and Awassi purebred sheep, Awassi/East-Friesian and Merino/Romanov crossbred sheep and one exotic cross breed (goat/ibex). N. cuniculi was isolated from 80-88% of the animals under 1 month of age. Among older animals, the microorganism was isolated from 20.5% of the pure bred animals and 79.3% of the crossbred ones. This difference was significant (P less than 0.001) by the chi 2 test. The prevalence of N. cuniculi in the second age group coincides with the susceptibility of the breeds to respiratory pathology. This, we believe, is the first report of N. cuniculi involved in multiple cases of respiratory pathology and of a survey assessing the prevalence of this microorganism in small ruminants.

Stable mycoplasma antigens for the indirect hemagglutination test (IHA) were prepared employing glutaraldehyde treated sheep erythrocytes sensitized with Mycoplasma agalactiae subsp. bovis and Mycoplasma bovigenitalium antigens. Employing these antigens mycoplasma antibodies were detected in sera from cattle which had mastitic symptoms due to natural infection with either M. agalactiae subsp. bovis or M. bovigenitalium. A total of 200 cows from four herds were examined at varying intervals for the presence of M. agalactiae subsp. bovis and for the detection of antibody using growth inhibition and IHA tests. Mycoplasmas were isolated from 37 animals. Growth inhibiting antibody was detected from 56 of the 200 animals. In the IHA tests, antibody titer greater than or equal to 1:80 were detected in 148 animals, 76 of these having antibody titers greater than or equal to 1:160, while sera of 116 normal control animals had no growth inhibiting antibody and none had IHA antibody titers greater than 1:40. M. bovigenitalium was isolated from the milk of three of 26 animals in a fifth herd during an outbreak of mastitis. Growth inhibiting antibodies were demonstrated in the sera of ten of the 26 animals. However, the IHA test detected antibody titers of greater than or equal to 1:160 in 13 animals and of 1:80 in one of the 26 animals. To determine the specificity of the IHA tests, M. agalactiae subsp. bovis and M. bovigenitalium antigens were reacted with rabbit hyperimmune typing sera produced against 12 species of bovine mycoplasmatales. Homologous antisera showed IHA antibody titers of 1:1280 and 1:2560 against M. agalactiae subsp. bovis and M. bovigenitalium respectively, whereas heterologous antisera showed IHA antibody titers of less than or equal to 1:20. Also eight type-specific bovine antisera were reacted with M agalactiae subsp. bovis and M. bovigenitalium antigens in homologous and heterologous tests. Homoogous reactions showed IHA antibody titers greater than or equal to 1:320, whereas heterologous reactions showed IHA titers of less than or equal to 1:20. This IHA test promises to be useful for the detection of bovine mycoplasma antibodies in sera from cattle infected with M. agalactiae subsp. bovis or M. bovigenitalium. Thes test is sensitive, reproducible and specific and the technique is relatively simple and rapid. The antigens were stable for at least seven months.

The bacterium Mycoplasma agalactiae is responsible for contagious agalactia (CA) in small domestic ruminants, a syndrome listed by the World Organization for Animal Health and responsible for severe damage to the dairy industry. Recently, we frequently isolated this pathogen from lung lesions of ibexes during a mortality episode in the French Alps. This situation was unusual in terms of host specificity and tissue tropism, raising the question of M. agalactiae emergence in wildlife. To address this issue, the ibex isolates were characterized using a combination of approaches that included antigenic profiles, molecular typing, optical mapping, and whole-genome sequencing. Genome analyses showed the presence of a new, large prophage containing 35 coding sequences (CDS) that was detected in most but not all ibex strains and has a homolog in Mycoplasma conjunctivae, a species causing keratoconjunctivitis in wild ungulates. This and the presence in all strains of large integrated conjugative elements suggested highly dynamic genomes. Nevertheless, M. agalactiae strains circulating in the ibex population were shown to be highly related, most likely originating from a single parental clone that has also spread to another wild ungulate species of the same geographical area, the chamois. These strains clearly differ from strains described in Europe so far, including those found nearby, before CA eradication a few years ago. While M. agalactiae pathogenicity in ibexes remains unclear, our data showed the emergence of atypical strains in Alpine wild ungulates, raising the question of a role for the wild fauna as a potential reservoir of pathogenic mycoplasmas.

The purpose of this study was to determine the frequency of selected pathogens in the tissues of a group of feedlot cattle with chronic disease (most often respiratory disease and/or arthritis). Samples of lung and joint tissues from 49 feedlot animals that had failed to respond to antibiotic therapy were tested by immunohistochemical staining for the antigens of Mycoplasma bovis, Haemophilus somnus, Pasteurella (Mannheimia) hemolytica, and bovine viral diarrhea virus (BVDV). Mycoplasma bovis was demonstrated in over 80% of cases, including in 45% of joints and 71% of lungs tested. Mycoplasma bovis was the only bacterial pathogen identified in the joints. Haemophilus somnus and Pasteurella (Mannheimia) haemolytica were found in 14% and 23% of cases, respectively, and were confined to the lungs in all instances. Infection with BVDV was demonstrated in over 40% of cases. Mycoplasma bovis and BVDV were the most common pathogens persisting in the tissues of these animals that had failed to respond to antibiotic therapy.

Streptococcus bovis is a normal inhabitant of the rumen but has been implicated as a causative agent for ruminal lactic acidosis and related problems. While rarely isolated from humans, S. bovis has been identified as a causative agent for endocarditis, meningitis, and septicemia. Recent reports have also suggested a correlation between human colonic carcinoma and increased levels of S. bovis. Identification of S. bovis strains of human origin has been problematic because of variations in results of biochemical tests compared with results for ruminal strains. We have tested a cloned amylase gene from the ruminal strain S. bovis JB1 as a potential DNA probe for rapid and accurate identification of S. bovis strains from all sources. DNAs from strains identified as S. bovis, of both human and ruminal origin, were found to hybridize with the probe under stringent conditions. The probe also hybridized with variants of S. bovis that did not grow on starch. The probe did not hybridize with DNA isolated from other bacteria of human colonic and ruminal origin, including Bacteroides thetaiotaomicron, Bacteroides ruminicola, Butyrivibrio fibrisolvens, and Enterococcus faecalis but did demonstrate hybridization with Streptococcus salivarius.

Images

Chronic, antibiotic-resistant pneumonia, sometimes with concurrent polyarthritis, occurs in feedlot cattle in western Canada. The prevalence of Mycoplasma bovis, bovine viral diarrhea virus, and Haemophilus somnus was determined by using immunohistochemical staining of lung and heart tissue from 2 groups of animals with this history. Mycoplasma bovis antigen was present in 44/48 cases submitted between 1995 and 1998 (retrospective group) and 15/16 of cases from 1999 (prospective group), and was associated with pulmonary necrosis. Bovine viral diarrhea virus antigen was present in association with microscopic vascular lesions in 31/48 retrospective and 9/16 of prospective cases. Types Ib and II bovine viral diarrhea virus were isolated from 4/16 prospective cases. Haemophilus somnus antigen was present in heart, lung, or both of 15/48 retrospective and 8/16 prospective cases. The results suggest that there may be synergism between bovine viral diarrhea virus and M. bovis in this pneumonia with arthritis syndrome.

The Staphylococcus (S.) intermedius group (SIG) has been a main research subject in recent years. S. pseudintermedius causes pyoderma and otitis in companion animals as well as foodborne diseases. To prevent SIG-associated infection and disease outbreaks, identification of both staphylococcal exotoxins and staphylococcal cassette chromosome mec (SCCmec) types among SIG isolates may be helpful. In this study, it was found that a single isolate (one out of 178 SIG isolates examined) harbored the canine enterotoxin SEC gene. However, the S. intermedius exfoliative toxin gene was found in 166 SIG isolates although the S. aureus-derived exfoliative toxin genes, such as eta, etb and etd, were not detected. SCCmec typing resulted in classifying one isolate as SCCmec type IV, 41 isolates as type V (including three S. intermedius isolates), and 10 isolates as non-classifiable. Genetic relatedness of all S. pseudintermedius isolates recovered from veterinary staff, companion animals, and hospital environments was determined by pulsed-field gel electrophoresis. Strains having the same band patterns were detected in S. pseudintermedius isolates collected at 13 and 18 months, suggesting possible colonization and/or expansion of a specific S. pseudintermedius strain in a veterinary hospital.

Escherichia coli has been used as an indicator organism for fecal contamination of water and other environments and is often a commensal organism in healthy animals, yet a number of strains can cause disease in young or immunocompromised animals. In this study, 281 E. coli isolates from bovine, porcine, chicken, canine, equine, feline, and other veterinary sources were analyzed by BOXA1R PCR and by virulence factor profiling of 35 factors to determine whether they had utility in identifying the animal source of the isolates. The results of BOXA1R PCR analysis demonstrated a high degree of diversity; less than half of the isolates fell into one of 27 clusters with at least three isolates (based on 90% similarity). Nearly 60% of these clusters contained isolates from more than one animal source. Conversely, the results of virulence factor profiling demonstrated clustering by animal source. Three clusters, named Bovine, Chicken, and Porcine, based on discriminant components analysis, were represented by 90% or more of the respective isolates. A fourth group, termed Companion, was the most diverse, containing at least 84% of isolates from canine, feline, equine, and other animal sources. Based on these results, it appears that virulence factor profiling may have utility, helping identify the likely animal host species sources of certain E. coli isolates.

SYNOPSIS

The threat of bioterrorism and emerging infectious diseases has prompted various public health agencies to recommend enhanced surveillance activities to supplement existing surveillance plans. The majority of emerging infectious diseases and bioterrorist agents are zoonotic. Animals are more sensitive to certain biological agents, and their use as clinical sentinels, as a means of early detection, is warranted.

This article provides design methods for a local integrated zoonotic surveillance plan and materials developed for veterinarians to assist in the early detection of bioevents. Zoonotic surveillance in the U.S. is currently too limited and compartmentalized for broader public health objectives. To rapidly detect and respond to bioevents, collaboration and cooperation among various agencies at the federal, state, and local levels must be enhanced and maintained. Co-analysis of animal and human diseases may facilitate the response to infectious disease events and limit morbidity and mortality in both animal and human populations.

Background

Recent focus on earlier detection of pathogen introduction in human and animal populations has led to the development of surveillance systems based on automated monitoring of health data. Real- or near real-time monitoring of pre-diagnostic data requires automated classification of records into syndromes–syndromic surveillance–using algorithms that incorporate medical knowledge in a reliable and efficient way, while remaining comprehensible to end users.

Methods

This paper describes the application of two of machine learning (Naïve Bayes and Decision Trees) and rule-based methods to extract syndromic information from laboratory test requests submitted to a veterinary diagnostic laboratory.

Results

High performance (F1-macro = 0.9995) was achieved through the use of a rule-based syndrome classifier, based on rule induction followed by manual modification during the construction phase, which also resulted in clear interpretability of the resulting classification process. An unmodified rule induction algorithm achieved an F1-micro score of 0.979 though this fell to 0.677 when performance for individual classes was averaged in an unweighted manner (F1-macro), due to the fact that the algorithm failed to learn 3 of the 16 classes from the training set. Decision Trees showed equal interpretability to the rule-based approaches, but achieved an F1-micro score of 0.923 (falling to 0.311 when classes are given equal weight). A Naïve Bayes classifier learned all classes and achieved high performance (F1-micro = 0.994 and F1-macro = .955), however the classification process is not transparent to the domain experts.

Conclusion

The use of a manually customised rule set allowed for the development of a system for classification of laboratory tests into syndromic groups with very high performance, and high interpretability by the domain experts. Further research is required to develop internal validation rules in order to establish automated methods to update model rules without user input.

The unique properties of nanomaterials related to structural stability and quantum-scale reactive properties open up a world of possibilities that could be exploited to design and to target drug delivery or create truly microscale biological sensors for veterinary applications. We developed cost-saving and solvent-free nanoemulsions. Formulated with a low-energy method, these nanoemulsions can find application in the delivery of controlled amounts of drugs into the beverage of breeding animals (such as poultry, cattle, pigs) or be used for the controlled release of injectable poorly water-soluble drugs.