Horse-Expert: An aided expert system for diagnosing horse diseases

Abstract

In contrast to the rapid development of the horse husbandry in China, the ability of horse veterinarians to diagnose diseases has not been improved and only a few domain experts have considerable expertise. At present, many expert systems have been developed for diseases diagnosis, but few for horse diseases diagnosis have been studied in depth. This paper presents the design and development of a computer-aided expert system for diagnosing horse diseases. We suggest an approach for diagnosis of horse diseases based on the analysis of diagnostic characteristics and the experiential knowledge of domain experts. It is based on using evidence-weighted uncertainty reasoning theory, which is a combination of evidence theory and an uncertainty pass algorithm of confidence factors. It enables drawing of inferences with atypical clinical signs and the uncertainty of the user's subjective understanding. It reduces the influence of subjective factors on diagnostic accuracy. The system utilizes a user friendly interface for users and requests a confidence factor from users when feedback is given to the system. Horse-Expert combines the confidence factors with weight factors assigned to clinical signs by experts during the knowledge acquisition process to make diagnostic conclusions. The system can diagnose 91 common horse diseases, and provides suggestions for appropriate treatment options. In addition, users can check the medical record through statistical charts. The system has been tested in seven demonstration areas of Xinjiang province in northwestern China. By constantly maintaining and updating the knowledge base, the system has potential application in veterinary practice.

Artificial intelligence: Is it wizardry, witchcraft, or a helping hand for an equine veterinarian?

“…They might enable animal owners to assess the animal's clinical signs and take appropriate measures, for example, using a rule-based system for horses. 24 Given that similar systems for triaging in human medicine have produced encouraging results, 25 it could be expected that further research might be fruitful.…”

“…With widespread shortages of trained veterinary staff, it could be expected that some efforts might be invested in the AI virtual assistants to alleviate the pressure on overburdened staff. They might enable animal owners to assess the animal's clinical signs and take appropriate measures, for example, using a rule‐based system for horses 24 . Given that similar systems for triaging in human medicine have produced encouraging results, 25 it could be expected that further research might be fruitful.…”

Intelligent System for Solving Problems of Veterinary Medicine on the Example of Dairy Farms

“…It consisted of questions about symptoms, diagnoses and treatments. Based on their experience and requirements to develop an expert system, experts modified the questionnaire and provided information in tabular form [8]. Digestive system: Loss of Appetite (F01), Loss of Chewing Cud (F02), Salivation (F03), Stomatitis (F04), Oral Mucosa Hemorrhages (F05), Oral Lesions (F06), Aphta, Oral Ulcers (F07), Vesicles, Tubercules in Mouth (F08), Gastric Timpany (F09), Gastric Atony (F10), Abdomen Wall Disease (F11), Diarrhea (F12), Constipation (F13), Excrements admixed with blood, mucus, gas bubbles (F14).…”

Pet dog disease pre-diagnosis system for caregiver with possibilistic C-means clustering and disease database

“…There have been several researches to make computer aided decision making systems for cattles" health monitoring as the form of expert system [3,[13][14][15][16][17]. Those systems can handle a limited range of diseases for a certain cattle such as dog [3], horse [13], cow [14], fish [15], swine [16], and general cattle [17]. Usually their data collection method is primary and only can handle less than 10 related diseases with traditional certainty factor [13][14][15][16] or fuzzy logic [17].…”

“…Those systems can handle a limited range of diseases for a certain cattle such as dog [3], horse [13], cow [14], fish [15], swine [16], and general cattle [17]. Usually their data collection method is primary and only can handle less than 10 related diseases with traditional certainty factor [13][14][15][16] or fuzzy logic [17]. Thus the utility of such system is very limited and possibly unreliable or subjective.…”