Rule-based systems Directory UMM :Data Elmu:jurnal:A:Aquacultural Engineering:Vol23.Issue1-3.Sept2000:

by parasites, viruses, bacteria and fungal agents Post, 1983; Stoskopf, 1993. All of this make for a huge problem space, and a major challenge for anyone trying to develop a program that will reach reasonable and timely diagnoses. Most of the existing examples of disease diagnosis programs are in the field of human medicine. Most of them tackle only a single disease or a subset of related diseases and none to our knowledge attempts to diagnose all human diseases. During the development of Fish-Vet, we looked into rule-based, case-based, neural-net and fuzzy logic systems. Each of these ‘pure’ systems has advantages as well as deficiencies. Our decision was to create a working program and, where real-world constraints were in conflict with the ‘pure’ systems, we ‘polluted’ that system as long as the end result was faster andor more accurate.

2. General problems in fish disease diagnosis

There are several problems inherent in a disease diagnostic process. These have to be taken into account by any software package trying to aid the diagnosis. “ No disease exhibits all the signs described in the literature. In most cases there are acute and chronic phases of a disease having differing signs. Therefore, the program has to be able to reach the right diagnosis with a partial set of signs. “ There is a time progression for every disease. A disease seen when the first clinical signs appear will exhibit different signs than when mortalities are already occurring. “ Since the program has to obtain input from a human user, the problem of terminology looms large. Until today and despite efforts made by international organizations, no accepted vocabulary has been agreed on for veterinary termi- nology CAP, 1998; HL7, 1999. This is now changing with the incorporation of veterinary terminology in SNOMED. Moreover, cultural differences will also result in different terms being used for the same condition. “ In many cases, by the time the fish exhibit signs of a problem, there is already a secondary agent involved virus + bacteria, fungus + bacteria, etc.. Therefore the signs observed by the user may ‘belong’ to more than one disease in the program’s database. “ ‘It is human to err’, but never more so than in our case. We have to take into account that the signs chosen by a user to describe a condition are influenced by his knowledge and experience. Therefore we have to deal with the possibility that ‘wrong’ signs will be entered by the user.

3. Rule-based systems

These systems are the ‘classical’ artificial intelligence AI applications, and include the majority of the expert systems ES deployed during the 1970 – 1980s Buchanan and Shortliffe, 1984; Ignizio, 1991. In principle, a rule-based ES contains a database of rules relating to the problem domain in question. These rules are in the form of ‘IF — THEN — ’ sentences. After the database is ready, we can query the rules using an inference engine IE, which is at the heart of the ES. We present the IE with observed signs, and it will search the rules, looking for those which fit the observations. Using those rules, the IE will try to find diseases whose rules have ‘fired’. Going the other way, using a list of possible diseases, the IE will enable us to see other rules and define other observations needed to make the differential diagnosis. The advantages of this technique are that it is mature, there are many systems to choose from, the system can ‘explain’ it’s results by showing those rules used to reach an answer and rules can be edited so that the IE is tuned to obtain better results. Another strength is that in most cases the IE uses an existing ES-shell so the development will not entail extensive programming as such, only the entry of rules Tu´nez et al., 1996. The main deficiency of rule-based systems is that they require some ‘deep’ knowledge in order to be truly effective. This means we need to know the causal relationships between signs and disease. Another problem is that the rule database grows very rapidly as the problem space expands. As a result, the computation time grows and for the problem space we needed, it became totally impractical. In Fish-Vet we use a rule based system with a very limited set of rules, whose only purpose is to cut down the problem space to a manageable size quickly. This entails rules for species membership i.e. some diseases are species-specific so that for any diagnosis we can ‘throw out’ all species-specific diseases not belonging to the species in question. A similar approach is used for water type i.e. seawater specific diseases can be disregarded when looking at a freshwater fish, etc.

4. Case based reasoning CBR