Abstract Food allergy is a disease that negatively affects quality of life, and in some cases its impact is serious. Diagnosing food allergies prior to exposure to the allergen(s) has significant costs and results in overdiagnosis leading to the avoidance of food to which patients are not allergic. Discovering relationships between features of food allergy data would support patients by finding their food allergens and avoiding the use of costly diagnostics. This paper presents the potential of using machine learning algorithms in discovering these relationships. The data was collected by the medical laboratory Intermedica through tests performed by patients with food allergies. The apriori algorithm is applied to these data. The relationships discovered in our data are implemented in a software application, which also has an interface to enter data about new patients being screened for food allergies. The set of discovered relationships leads to the creation of a list of food allergens for a new patient, which helps them eliminate the molecular allergy test when it is not necessary and as a result, reduce financial costs. The model also supports patients by not eliminating foods that do not harm them, thereby not risking a nutritional deficit.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|