The Application of Machine Learning Technique for Malaria Diagnosis

2. Literature Review

The concepts of Artificial Intelligence in medicine have been researched upon in several respects. Hong (1988) summarized the potential of AI techniques in medicine as follows:

• •

  Produces new tools to support medical decision making, training and research.

• •

  Integrates activities in medical, computer, cognitive and other sciences, etc.

Early studies in intelligent medical system such as MYCIN, (ASNET, PIP and internist-I have shown to out perfume manual practice of diagnosis in several disease domain (Shortlitte, 1987).

Machine learning is a branch of Artificial Intelligence; Artificial intelligence (AI) is the science and technology whose goal is to develop computers that can think, see, perceive, hear, talk and feel etc. Anigbogu (2003) in order words, artificial intelligence involves developing a machine (computer system), which functions are normally associated with human intelligence, which include; reasoning, inference, hearing and problem solving etc (Patterson, 1990).

• Diagnosis is the identification of abnormal condition that afflicts a specific patient, based on manifested clinical data or lesions. If the final diagnosis agrees with a disease that afflicts a patient, the diagnostic process is correct; otherwise, a misdiagnosis occurred (Feder, 2006). The diagnostic algorithm will be based on disease models stored in the computer knowledge base including the name of disease with the cause, pathogenesis, lesion, pathophysiology, clinical data, syndromes, clinical presentation and complications.

• Symptoms, in a strict medical sense, are subjective clues (e.g. pain, nausea) that the patient experiences. These are revealed by the patient during history taking.

• Signs are objective clues (e.g. swelling, wheezing) that a clinician detects during steps of physical examination.

• Results of test are clues obtained through laboratory test and other techniques (Feder, 2006)

• A Decision Tree is a logical model represented as a binary (two-way split) tree that shows how the value of a target variable can be predicated by using the values of a set of predicator variables (Figure 1) (Quinlan, 1986).