| Summary: | 碩士 === 逢甲大學 === 應用數學系 === 106 === Dementia is a progressive deterioration of cognitive function caused by brain lesions or injuries, and the rate of deterioration is much higher than that of normal normal aging. The occurrence of dementia, in addition to the affected patients and their families, also has a huge impact on the country’s finances. Early detection of dementia can give appropriate treatment and care. In this study, Naive Bayes and Support Vector Machine (SVM) were used to train the model. The average accuracy of the trained model was about 82.4% and 86.3% respectively. Compared with the traditional Naive Bayes statistical learning method, SVM does not need to make any statistical assumptions except for its high prediction accuracy. Finally, we selected the SVM as a method of modeling, using the Matlab GUIDE tool, embedded classification model and developed into an application program interface to facilitate the operation of medical personnel. The future hopes to provide an objective decision-making tool for medical personnel through this dementia screening system to conduct preliminary pre-diagnosis.
|
|---|
