5GL-Doctor, Medical Diagnosis Aid/Expert System for Windows (32 or 64 bit), Suitable for the personal and professional user

All rights reserved © 5GL Software, Australia

A medical diagnosis aid/expert system. Large database with thousands of medical condition names (diseases) associated with a symptom/sign/event pattern. Suitable for personal or professional user. Lab analysis and drug reactions separate functions. Entries in the database, in their description, are intended to aid diagnosis. These rarely contain a complex description, more clues and hints to diagnosis. Treatment information is for the general user only.

Yipee

Need to relax? Who doesn’t? Enjoy good classical style science fiction?  Click on cover for link or search for Yipee Lulu.

RemindMe Links

Results shown depend on options set, country/region where used, how many patterns in the database, and other factors. Your results would be similar but not necessarily the same as shown in the images some of which are from past versions of 5GL-Doctor or development mode. There are number of analysis functions such as history analysis and an analysis that compares similar medical conditions to the selected one naming common symptoms/indicators. Filters can be used to make an analysis more generic. Example. The symptom high fever in the inquiry would only match with patterns that contain this, but if the fever filter is ticked then any reference to fever is considered valid. Your version of 5GL-Doctor would produce similar short lists but the P() values may differ.

An innovative way of looking at the evidence. In the above, the IRC is name of a disease Category, the Population the number of elements (symptoms/signs/events) in this universe, the Stars are those conditions that stand out relative to the inquiry, and the Constellations the number of conditions in this populated universe that missed a perfect inquiry match by one. When this is a large number, a re-think of the inquiry is appropriate. Selecting symptoms/signs/events at random from the Selection List is not likely to yield any results - because these have to fit a known IRC universe first. Or, if there is a message No Results, that may mean a patient has two separate conditions or one of the symptoms/signs/events in the inquiry is not directly related to the others. In the above, only the typical presenting symptoms/signs that matched the inquiry are shown. There may be more typical symptoms/signs - you need to look at the actual database entry. Also, a condition may have even thirty possible symptoms/signs as well, these defined in the “Possible Presentations” field in the actual record (which you can change). Clicking on any disease name, in most of the short lists, will take you into the disease entry itself. The IRC universe concept is a different way of looking at symptoms/signs/events. If you are a physician, you may find this analysis more useful than Analysis 1 (the expert system analysis). It is actually rather clever. Just a glance at the information on the left suggests the inquiry is related to a Pulmonary issue. (There may be some duplication of disease entries, usually with a suffix in such cases, this to assist the internal algorithms - or because medicine can be fuzzy. Some sources will claim Gallstones is also known as Cholecystitis - in fact, it is not the same thing. The latter usually caused by gallstones, but not exactly gallstones. However, this is quibbling - you can adapt the database to suit your way of understanding). Another issue in medicine is this: look at two expert sources in regards to symptoms/signs of a disease, and you may find the two don’t always look the same. The typical presentation field contains symptoms/signs all sources agree on - the possible is all the other possibilities. The symptom/signs from different sources are usually listed in the actual entry as well.

5GL-Doctor is about assisting with a timely and correct diagnosis. This is what the thrust of the information is about. Clues and hints how to make a diagnosis, these from medical expert sources, when found are usually included. A Conversation might, in response to multiple choice questions, add points and at the end based on the value of points present a diagnosis (i.e. a medical algorithm adapted to a conversation). Probably any medical algorithm can be adapted to a conversation.

Another interesting way of considering the evidence. This might get you right where you need to go. Unlike in Analysis 1 in which you sometimes need to wait for the results because (in some cases) of the volume of computations, this view is an instant response result. An inquiry element such as temperature or fever appears in over 470 disease patterns (probably more because some of the chemical exposure entries have a list 100 or more long of what the exposure is and what the symptoms are) so, in some settings, excluding some of the very high in appearance elements can yield a cleaner view. (Not sure if Dermatitis is the best category in the above case - that can be changed on the database of course, another category included in the Dermatomyositis entry. Thus this analysis ignores that which can confuse (i.e. many occurrences) and focuses on that which can provide a clearer view.