Application of Artificial Neural Networks in Dementia and Alzheimer's Diagnosis

Yigit A., Işık Z.

26th IEEE Signal Processing and Communications Applications Conference (SIU), İzmir, Türkiye, 2 - 05 Mayıs 2018 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2018.8404447
  • Basıldığı Şehir: İzmir
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: Artificial Neural Networks, Alzheimer Disease, Dementia, Machine Learning, Medical Diagnosis
  • Dokuz Eylül Üniversitesi Adresli: Evet

Özet

Diagnosis in the early phases of many diseases makes it possible to treat the disease and affects the treatment process positively. This is especially important for diseases like Alzheimer in the field of neurology. The use of a computerized support system, which can autonomously perform the diagnostic process by the expert in this process, saves time and helps to reduce the most human errors. In this study, machine learning models with the ability to diagnose dementia and Alzheimer's disease were developed by predicting the Clinical Dementia Rating (CDR) value. Artificial Neural Networks (ANN), Logistic Regression (LR), k-nearest neighbors (KNN), and Decision Tree (DT) classifiers were applied to compare the classification performances. The Open Access Series of Imaging Studies (OASIS) longitudinal and cross-sectional datasets have been used to train models. As a result of the tests, best performance of the detection and identification of Alzheimer's disease has been shown by LR and YSA models.