Analyze Informant-Based Questionnaire for The Early Diagnosis of Senile Dementia Using Deep Learning

Abstract

Objective: This paper proposes a multiclass deep learning method for the classification of dementia using an informant-based questionnaire. Methods: A deep neural network classification model based on Keras framework is proposed in this paper. To evaluate the advantages of our proposed method, we compared the performance of our model with industry-standard machine learning approaches. We enrolled 6,701 individuals, which were randomly divided into training data sets (6030 participants) and test data sets (671 participants). We evaluated each diagnostic model in the test set using accuracy, precision, recall, and F1-Score. Results: Compared with the seven conventional machine learning algorithms, the DNN showed higher stability and achieved the best accuracy with 0.88, which also showed good results for identifying normal (F1-score = 0.88), mild cognitive impairment (MCI) (F1-score = 0.87), very mild dementia (VMD) (F1-score = 0.77) and Severe dementia (F1-score = 0.94). Conclusion: The deep neural network (DNN) classification model can effectively help doctors accurately screen patients who have normal cognitive function, mild cognitive impairment (MCI), very mild dementia (VMD), mild dementia (Mild), moderate dementia (Moderate), and severe dementia (Severe). INDEX TERMS Dementia, information gain, deep neural network, machine learning.

Motor Dysfunction Questionnaire and Dopamine Transporter Imaging Composite Scale Improve Differentiating Dementia With Lewy Bodies From Alzheimer's Disease With Motor Dysfunction

“…This was a two-phase study to design and test the MDQ embedded in the History-based Artificial Intelligence Clinical Dementia Diagnostic System (HAICDDS), which is currently used to register patients with dementia or motor dysfunction in the Show Chwan Healthcare System (Lin et al, 2018 ; Chiu et al, 2019a , b ; Wang et al, 2020 ; Zhu et al, 2020 ). Before beginning the project, 30 patients with their caregivers were tested by neuropsychologists from three centers, and the reproducibility was studied using the interrater reliability analysis.…”

Operational Determination of Subjective Cognitive Decline, Mild Cognitive Impairment, and Dementia Using Sum of Boxes of the Clinical Dementia Rating Scale

“…This study is a retrospective analysis of the dementia registry database from the Show Chwan Healthcare System, currently applied in three centers in Taiwan (two in central Taiwan and one in southern Taiwan). In the database, the detailed clinical history of each participant was recorded using a structured questionnaire called the History-Based Artificial Intelligent Clinical Dementia Diagnostic System (HAICDDS), which has been well-validated (Lin et al, 2018 ; Chiu et al, 2019a , b , 2020 ; Tsai and Chiu, 2019 ; Chang et al, 2020 ; Wang et al, 2020 ; Zhu et al, 2020a , b ; Huang et al, 2021 ). In addition, CDR was used for staging dementia, and the daily function was assessed using the Instrumental Activities of Daily Living (IADL) Scale (Lawton and Brody, 1969 ) and Barthel Index (BI) (Mahoney and Barthel, 1965 ).…”

“…In our recent experience and studies using artificial intelligence (AI) for the diagnosis of cognitive impairment (CI) and dementia, the CDR and CDR-SB have become perfect references for machine learning in our newly designed questionnaires (Chiu et al, 2019a ; Chang et al, 2020 ; Zhu et al, 2020a , b ). Therefore, we highly recommend CDR as a further AI study.…”

Assessing Significance of Cognitive Assessments for Diagnosing Alzheimer’s Disease with Fuzzy-Rough Feature Selection

“…Each of such assessment carries a score, which is interchangeably termed cognitive assessment factor or cognitive feature hereafter. Yet, among numerous cognitive assessments available, there is no global standard for what assessments are more appropriate to be applied to patients, as reflected by different practices observed in recent literature [5,6], while the separate use of such assessment factors individually also tends to perform poorly. For instance, when considering only the Mini-Mental State Examination (MMSE) [7], one of the most widely used assessment features, a mere less than 70% accuracy is attainable even with the powerful support vector machine and neural networks employed as the classifier.…”

“…For instance, when considering only the Mini-Mental State Examination (MMSE) [7], one of the most widely used assessment features, a mere less than 70% accuracy is attainable even with the powerful support vector machine and neural networks employed as the classifier. Fortunately, the use of multiple assessment factors has generally been shown to provide a good indicator of AD in a number of case studies [6].…”