Machine Learning Approaches for Prediction of Alzheimer’s Disease

Kadriye Filiz Balbal

Makale detayları

Machine Learning Approaches for Prediction of Alzheimer’s Disease

Yazarlar:

Kadriye Filiz Balbal

Yayıncı:

Izmir Katip Celebi University

Ana Başlıklar:

Makine Öğrenme (Diğer)

Dergi:

Journal of Artificial Intelligence and Data Science Vol.4 Issue 2, pp.110-118

Belge Tipi:

Makale

Yayın Tarihi:

12/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001335933

pISSN:

2791-8335

Anahtar kelimeler:

Özet:

 Alzheimers Disease AD is a disorder that significantly impacts an individuals behavior, memory, and cognitive functions, ultimately leading to a loss of independence. Early and accurate diagnosis of AD is critical to mitigating its progression and improving patient outcomes, especially as no definitive cure is currently available. This study investigates the application of machine learning algorithms to predict and diagnose AD based on patient symptoms and clinical data. The dataset used in this research includes comprehensive health information from 2,149 patients, with 35 features covering demographic, lifestyle, and medical factors, and no missing values. Seven widely recognized machine learning algorithmsKNN, GNB, SVM, DT, RF, AdaBoost, and XGBoostwere evaluated to determine their effectiveness in disease prediction. Performance was assessed using recall, precision, accuracy, and F1-score metrics, providing a robust evaluation of each model. XGBoost achieved the highest accuracy rate of 95.35, highlighting its superior predictive capability, while KNN recorded the lowest accuracy at 75.54. The results demonstrate the strength of machine learning algorithms, particularly ensemble methods like XGBoost, in analyzing complex clinical data for the early detection of Alzheimers Disease. These findings underscore the critical role of machine learning in enhancing diagnostic accuracy and enabling timely interventions, which are essential for improving the quality of life for individuals at risk of Alzheimers Disease.

 Alzheimers Disease AD is a disorder that significantly impacts an individuals behavior, memory, and cognitive functions, ultimately leading to a loss of independence. Early and accurate diagnosis of AD is critical to mitigating its progression and improving patient outcomes, especially as no definitive cure is currently available. This study investigates the application of machine learning algorithms to predict and diagnose AD based on patient symptoms and clinical data. The dataset used in this research includes comprehensive health information from 2,149 patients, with 35 features covering demographic, lifestyle, and medical factors, and no missing values. Seven widely recognized machine learning algorithmsKNN, GNB, SVM, DT, RF, AdaBoost, and XGBoostwere evaluated to determine their effectiveness in disease prediction. Performance was assessed using recall, precision, accuracy, and F1-score metrics, providing a robust evaluation of each model. XGBoost achieved the highest accuracy rate of 95.35, highlighting its superior predictive capability, while KNN recorded the lowest accuracy at 75.54. The results demonstrate the strength of machine learning algorithms, particularly ensemble methods like XGBoost, in analyzing complex clinical data for the early detection of Alzheimers Disease. These findings underscore the critical role of machine learning in enhancing diagnostic accuracy and enabling timely interventions, which are essential for improving the quality of life for individuals at risk of Alzheimers Disease.

 Alzheimers Disease AD is a disorder that significantly impacts an individuals behavior, memory, and cognitive functions, ultimately leading to a loss of independence. Early and accurate diagnosis of AD is critical to mitigating its progression and improving patient outcomes, especially as no definitive cure is currently available. This study investigates the application of machine learning algorithms to predict and diagnose AD based on patient symptoms and clinical data. The dataset used in this research includes comprehensive health information from 2,149 patients, with 35 features covering demographic, lifestyle, and medical factors, and no missing values. Seven widely recognized machine learning algorithmsKNN, GNB, SVM, DT, RF, AdaBoost, and XGBoostwere evaluated to determine their effectiveness in disease prediction. Performance was assessed using recall, precision, accuracy, and F1-score metrics, providing a robust evaluation of each model. XGBoost achieved the highest accuracy rate of 95.35, highlighting its superior predictive capability, while KNN recorded the lowest accuracy at 75.54. The results demonstrate the strength of machine learning algorithms, particularly ensemble methods like XGBoost, in analyzing complex clinical data for the early detection of Alzheimers Disease. These findings underscore the critical role of machine learning in enhancing diagnostic accuracy and enabling timely interventions, which are essential for improving the quality of life for individuals at risk of Alzheimers Disease.

 Alzheimer Hastalığı AD, bir bireyin davranışını, hafızasını ve bilişsel işlevlerini önemli ölçüde etkileyen ve nihayetinde bağımsızlık kaybına yol açan bir hastalıktır. ADnin erken ve doğru teşhisi, özellikle şu anda kesin bir tedavi bulunmadığından, ilerlemesini hafifletmek ve hasta sonuçlarını iyileştirmek için kritik öneme sahiptir. Bu çalışma, hasta semptomlarına ve klinik verilere dayanarak ADyi tahmin etmek ve teşhis etmek için makine öğrenimi algoritmalarının uygulamasını araştırmaktadır. Bu araştırmada kullanılan veri seti, demografik, yaşam tarzı ve tıbbi faktörleri kapsayan 35 özellik ve eksik değer içermeyen 2.149 hastadan kapsamlı sağlık bilgilerini içerir. Hastalık tahminindeki etkinliklerini belirlemek için yaygın olarak tanınan yedi makine öğrenimi algoritması - KNN, GNB, SVM, DT, RF, AdaBoost ve XGBoost - değerlendirildi. Performans, her modelin sağlam bir değerlendirmesini sağlayan geri çağırma, kesinlik, doğruluk ve F1 puanı ölçümleri kullanılarak değerlendirildi. XGBoost, üstün tahmin yeteneğini vurgulayan 95,35lik en yüksek doğruluk oranına ulaşırken, KNN 75,54 ile en düşük doğruluğu kaydetti. Sonuçlar, Alzheimer Hastalığının erken teşhisi için karmaşık klinik verileri analiz etmede makine öğrenimi algoritmalarının, özellikle XGBoost gibi topluluk yöntemlerinin gücünü göstermektedir. Bu bulgular, Alzheimer Hastalığı riski taşıyan bireylerin yaşam kalitesini iyileştirmek için gerekli olan tanı doğruluğunu artırma ve zamanında müdahaleleri mümkün kılmada makine öğreniminin kritik rolünü vurgulamaktadır.

Makale detayları

Machine Learning Approaches for Prediction of Alzheimer’s Disease

Yazarlar:

Kadriye Filiz Balbal

Yayıncı:

Izmir Katip Celebi University

Ana Başlıklar:

Makine Öğrenme (Diğer)

Dergi:

Journal of Artificial Intelligence and Data Science Vol.4 Issue 2, pp.110-118

Belge Tipi:

Makale

Yayın Tarihi:

12/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001335933

pISSN:

2791-8335

machine learning

classification

Alzheimer

disease predictionmakine öğrenmesi

sınıflandırma

hastalık tahmini

Alzheimer Hastalığı

Sayı

Makale detayları

Yazarlar: Kadriye Filiz Balbal

Yayıncı: Izmir Katip Celebi University

Ana Başlıklar: Makine Öğrenme (Diğer)

Dergi: Journal of Artificial Intelligence and Data Science Vol.4 Issue 2, pp.110-118

Belge Tipi: Makale

Yayın Tarihi: 12/2024

Belge Dili: İngilizce

Document ID (Org No.): 2001335933

pISSN: 2791-8335

Sayı

Yazarlar:

Kadriye Filiz Balbal

Yayıncı:

Izmir Katip Celebi University

Ana Başlıklar:

Makine Öğrenme (Diğer)

Dergi:

Journal of Artificial Intelligence and Data Science Vol.4 Issue 2, pp.110-118

Belge Tipi:

Makale

Yayın Tarihi:

12/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001335933

pISSN:

2791-8335