9.18 EasyDiagnosis

EasyDiagnosis is an automatic online medical diagnosis expert system freely available to consumers and health care professionals. {1}

The system uses artificial intelligence to suggest a diagnosis ranked by probability according the symptoms presented, supposing those symptoms all arise from a single condition. Some 99 conditions are covered at present, including the most life-threatening. The service is not offered as an alternative to a professional examination by a physician, but as a useful adjunct or introductory screening.

Medical Expert Online is similar, but is an online model for demonstration purposes only. {2}

EasyDiagnosis employs an expert system (indeed MathMedics markets its AI software through this application) and therefore differs from sites like Symptom Diagnosis, {3} which offers a summary compilation of symptoms for each of more than 210 medical conditions, or Just Answer. Doctors, {4} which has actual online doctors answering health queries.

Technology

Easy Diagnosis has been built by MathMedics, {5} who do not disclose details of their proprietary software. Expert systems tend to be built in an artificial intelligence language, however, of which there are many. {6} The usual requirements are:

1. Language suits the particular domain of AI,
2. Coding can be done quickly, with a minimum of errors,
3. Resulting program runs at an acceptable speed, and (sometimes)
4. Code and data have similar structures.

A paper describing an asthma diagnosis expert system by Lee {7} integrates three units:

1. Guideline engine that supplies the physician with consistent treatment strategies.
2. Clinical knowledge base.
3. Patient's clinical record.

The workflow description language adopted was BPEL, {8} a web-based Business Executive Language which allows data to flow independently of function if recorded in XML format.

Typically, the service would start with a physical examination of the patient, with salient findings being stored and compared with previous information on the patient's record. The doctor would then consult the treatment strategies suggested by the expert system, decide to what extent these were applicable, and discuss options with the patient. Once agreed, the option would be added to the patient's record, and any prescribed medication printed out.

Smartphone Applications

Billed as a 'doctor in your pocket', Afridoctor is a free add-on for Nokia phones that provides basic, health-related information. {9} Though more intended for minor ailments, Afridoctor features:

1. 'Find a doctor' service uses Google Maps to locate local health services: doctors, hospitals and emergency clinics.
2. 'Distress' service that enables users to quickly contact a family member or friend.
3. First aid tips.
4. Symptom checker.
5. 'Snapdiagnosis', where patients send pictures of their ailments to a panel of doctors who reply with a diagnosis within 48 hours.

Probably as an indication of things to come, most downloads of the application were from outside Africa: over 20,000 by June 2010. {10}

Points to Note

1. Internet technology creates new services, sometimes unimaginable a decade ago.

Questions

1. Explain how Easy Diagnosis works.
2. What is the business model for the service?
3. Your company offers financial services. How would you persuade your boss to adopt the Easy Diagnosis model? Provide a cost benefit study.
4. You're a busy local health authority facing staff cutbacks. How could something like Easy Diagnosis help, in what areas, and how would you start such a project?

Sources and Further Reading

Need the references and resources for further study? Consider our affordable (US $ 4.95)  pdf ebook. It includes extensive (3,000) references, plus text, tables and illustrations you can copy, and is formatted to provide comfortable sequential reading on screens as small as 7 inches.

   Get your eBook here.