Diagnosis of Dementia and Alzheimer's Disease Based on Classification Algorithms

  • Ci Song Northeastern University at Qinhuangdao
  • Shuxian Zong Northeastern University at Qinhuangdao
DOI: https://doi.org/10.18686/aem.v11i4.335
Keywords: Machine Learning, Smote, Dementia Diagnosis

Abstract

Alzheimer's disease is currently the most common kind of senile dementia. With the increasing aging degree of the global society, Alzheimer's disease will become an unavoidable social problem in an aging society. In order to improve this situation, artificial intelligence algorithms that are good at mining the internal laws of data are applied in the hope of more effectively diagnose this disease, which should be intervened as early as possible. After briefly restating the current situation of dementia and Alzheimer's disease, the diagnostic model for dementia is built using logistic regression, which achieves great accuracy despite the simplicity of the model. Then, two diagnostic models that can identify if the patient with dementia has Alzheimer's disease based on SVM and Random Forest are tested. Although both the algorithms perform poorly because of the sample imbalance, after processing the original data with SMOTE, their performances are largely improved.

References

[1] Zhao Q, Chen P, Zhu SQ, et al. Effects of nonpharmacological interventions on elderly people with subjective memory complaints: a systematic review[J]. Chinese General Practice, 2020, 23(29): 3719-3728.

[2] Li N, Zhao Y. Research progress on early recognition of mild cognitive impairment and related theoretical models[J]. Chinese Journal of Nursing, 2018, 53(5):6.

[3] Whitehouse P, Price D, Struble R, et al. Alzheimer's dementia: loss of neurons in the basal forebrain[J]. Annals of Neurology, 1982, 10:122-126.

[4] Yan An. Community Health Service in Early Diagnosis and Adjuvant Therapy of Alzheimer Disease[J]. Continuing Medical Education, 2015, 000(007):131-133.

[5] Alzheimer Features, Available from: https://www.kaggle.com/ datasets/brsdincer/ alzheimer-features.

[6] Liao JG, Chin KV. Logistic regression for disease classification using microarray data: model selection in a large p and small n case. [J]. Bioinformatics, 2007, 23(15):1945-1951.

[7] Alzheimer's clinical data, Available from: https://www.kaggle. com/datasets/ legendahmed/alzheimers-clinical-data.

[8] Vapnik, VladimirN. An Overview of Statistical Learning Theory. [J]. IEEE Transactions on Neural Networks, 1999.

[9] Long BT. Network Video Customer Churn Prediction and Analysis Based on Random Forest and K-means Algorithm[J]. Journal of Hubei Minzu University (Natural Science Edition), 2022,40(02):202-207.

[10] Hui H, Wang WY, Mao BH. Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning[J]. Lecture Notes in Computer Science, 2005.
Published
2023-03-02
Issue
Vol 11, No 4 (2022)
Section
Original Research Article

Copyright (c) 2023 Ci Song, Shuxian Zong

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Authors submitting to USP journals agree to publish their manuscript under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0) where authors agree to allow third parties to share their work (copy, distribute, transmit) and to adapt it, under the condition that the authors are given credit, and that in the event of reuse or distribution, the terms of this license are made clear

Authors retain copyright of their work, with first publication rights (online and print) granted to Universe Scientific Publishing or the owner of the journal in question.