Risk Prediction Model for Dementia by Deep Learning Using Clinical Data

Abstract

It is estimated that dementia, which is the most important public health problem in the elderly, will increase day by day. It is stated that this situation will create great challenges for public health and aged care systems in all countries of the world. For this reason, it has become very important to determine the management and treatment procedures of dementia, to reduce the level of progression of the disease and to increase the quality of life of individuals exposed to the disease. The purpose of this study is to predict dementia and reveal the factors related to the disease with the deep learning approach. In the current study, open-access dementia data, which includes the information of 376 patients, was used. Dementia prediction was made using the deep learning method. Model results were evaluated with accuracy, balanced accuracy, sensitivity, selectivity, positive predictive value, negative predictive value, and F1-score performance metrics. In addition, 10-fold cross-validation method was used in the modeling phase. Finally, variable importance values were obtained by modeling. When the results are examined The highest metric values among the performance criteria achieved for group variable types were calculated for Demented; and were found that Accuracy, Sensitivity, Specificity, Positive predictive value, Negative predictive Value, and F1-score were 0.947, 0.946, 0.978, 0.966, 0.965 and 0.956 respectively. As a result, when the findings obtained from this study were examined, the dementia dataset, which consisted of imaging data and information about patients with clinical data, was classified with high accuracy using the deep learning method. The risk factors for dementia were determined with the variable importance values obtained as a result of the model.

Keywords

• Dementia
• classification
• machine learning
• deep learning
• variable importance.

References

1. [1] G. Demir, "Demans Ve Hemşirelik Bakımı," Black Sea Journal Of Health Science, Vol. 1, Pp. 35-39.
2. [2] M. Braun, U. Scholz, B. Bailey, S. Perren, R. Hornung, and M. Martin, "Dementia caregiving in spousal relationships: a dyadic perspective," Aging and mental health, vol. 13, pp. 426-436, 2009.
3. [3] M. Prince, M. Guerchet, and M. Prina, "Alzheimer’s Disease International (2013b) Policy Brief for Heads Of Government: The Global Impact of Dementia 2013-2050," Alzheimer's Disease International, London, 2014.
4. [4] Ö. G. D. İ. Akyar, "Demanslı hasta bakımı ve bakım modelleri," Hacettepe Üniversitesi Hemşirelik Fakültesi Dergisi, vol. 18, pp. 79-88, 2011.
5. [5] L. Deng and D. Yu, "Deep learning: methods and applications," Foundations and trends in signal processing, vol. 7, pp. 197-387, 2014.
6. [6] Y. Bengio and Y. LeCun, "Scaling learning algorithms towards AI," Large-scale kernel machines, vol. 34, pp. 1-41, 2007.
7. [7] A. Şeker, B. Diri, and H. H. Balık, "Derin öğrenme yöntemleri ve uygulamaları hakkında bir inceleme," Gazi Mühendislik Bilimleri Dergisi (GMBD), vol. 3, pp. 47-64, 2017.
8. [8] K. Umut, A. YILMAZ, and Y. Dikmen, "Sağlık alanında kullanılan derin öğrenme yöntemleri," Avrupa Bilim ve Teknoloji Dergisi, pp. 792-808, 2019.

9. [9] K. Kayaalp and A. Süzen, "Derin Öğrenme ve Türkiye’deki Uygulamaları," Yayın Yeri: IKSAD International Publishing House, Basım sayısı, vol. 1, 2018.
10. [10] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," nature, vol. 521, pp. 436-444, 2015.
11. [11] Y. Onadja, N. Atchessi, B. A. Soura, C. Rossier, and M.-V. Zunzunegui, "Gender differences in cognitive impairment and mobility disability in old age: a cross-sectional study in Ouagadougou, Burkina Faso," Archives of gerontology and geriatrics, vol. 57, pp. 311-318, 2013
12. [12] Ü. Görgülü and N. Akdemir, "İleri evre kanser hastalarına bakım verenlerin yorgunluk ve uyku kalitesinin değerlendirilmesi," Genel Tip Dergisi, vol. 20, 2010.
13. [13] M. N. Özata Değerli, "Orta Evre Alzheimer Tip Demans Tanılı Bireylerde Demansa Bağlı Davranışsal Ve Psikolojik Semptomlar İle Duyusal İşlemleme Süreci Arasındaki İlişkinin İncelenmesi," 2022.
14. [14] E. Alpaydin, Introduction to machine learning: MIT press, 2020.
15. [15] J. Schmidhuber, "Deep learning in neural networks: An overview," Neural networks, vol. 61, pp. 85-117, 2015.
16. [16] S. Bharati, P. Podder, D. N. H. Thanh, and V. Prasath, "Dementia classification using MR imaging and clinical data with voting based machine learning models," Multimedia Tools and Applications, pp. 1-22, 2022.