Using Variational Autoencoder to Develop and Validate a Compact, Deep Representation of Digital Clock Drawing Test for Classifying Dementia

The Clock Drawing Test (CDT) is an inexpensive tool to screen for dementia. In this study, we examined if a semi-supervised deep learning (DL) system using Variational Autoencoder (VAE) can extract atypical clock features from a large dataset of unannotated CDTs (n=13,580) and use them to classify dementia (n=18) from non-dementia (n=20) peers. The classification model built with VAE latent space features adequately classified dementia from non-dementia (0.78 Area Under Receiver Operating Characteristics (AUROC)). The VAE-identified atypical clock features were then reviewed by domain experts and compared with existing literature on clock drawing errors. This study shows that a semi-supervised deep learning (DL) analysis of the CDT can extract important clock drawing anomalies that are predictive of dementia.

Extended Application of Digital Clock Drawing Test in the Evaluation of Alzheimer’s Disease Based on Artificial intelligence and the Neural Basis

Introduction: This study aimed to build the supervised learning model to predict the state of cognitive impairment, Alzheimer’s Disease (AD) and cognitive domains including memory, language, action, and visuospatial based on Digital Clock Drawing Test (dCDT) precisely. Methods: 207 normal controls, 242 Mild Cognitive Impairment (MCI) patients, 87 dementia patients, including 53 AD patients, were selected from Shanghai Tongji Hospital. The electromagnetic tablets were used to collect the trajectory points of dCDT. By combining dynamic process and static results, different types of features were extracted, and the prediction models were built based on the feature selection approaches and machine learning methods. Results: The optimal AUC of cognitive impairment’s screening, AD’s screening and differentiation are 0.782, 0.919 and 0.818, respectively. In addition, the cognitive state of the domains with the best prediction result based on the features of dCDT is action with the optimal AUC 0.794, while the other three cognitive domains got the prediction results between 0.744-0.755. Discussion: By extracting dCDT features, cognitive impairment and AD patients can be identified early. Through dCDT feature extraction, a prediction model of single cognitive domain damage can be established.

Clock drawing test: comparison between the Pfizer and the Shulman systems

ABSTRACT Cognitive decline can be screened by the clock drawing test (CDT), which has several versions. Objective: This survey aimed to analyze the correlation between two simple methods for scoring the CDT. Methods: This cross-sectional study was nested in the Elo-Creati cohort from Passo Fundo, Brazil and comprised 404 subjects. Two raters underwent previous training and scored the subjects’ CDT according to both the Pfizer and Shulman systems. The inter-observer and intra-observer concordance within each method was analyzed with the Spearman’s rank correlation coefficient, as well as the concordance of the scores between the two methods. Age and scholarity were also correlated with the scores. Results: Most of the participants were women (93.8%) and Caucasian (84.6%), with a mean age of 66.9 (±7.8) years and a scholarity of 10.9 years (±5.6). There was significant inter-observer (Pfizer: r=0.739, p£0.001; Shulman: r=0.727, p£0.001) and intra-observer correlation (Pfizer: rater 1, r=0.628, p≤0.001; rater 2, r=0.821, p≤0.001; Shulman: rater 1, r=0.843, p≤0.001; rater 2: r=0.819; p≤0.001). Intra-observer correlation was also observed comparing Pfizer and Shulman methods (rater 1: r=0.744; p≤0.001; rater 2: r=0.702; p≤0.001). There was weak correlation of the scores with scholarity (Pfizer: r=0.283, p£0.001; Shulman: r=0.244, p£0.001) and age (Pfizer: r=-0.174, p£0.001; Shulman: r=-0.170, p£0.001). More participants were classified with decreased cognition through the Pfizer system (rater 1: 44.3 vs. 26.5%; rater 2: 42.1 vs. 16.3%; p≤0.001). Conclusions: For this population, our results suggest that the Pfizer system of scoring CDT is more suitable for screening cognitive decline.

Taking the Time to Assess Cognition in Parkinson’s Disease: The Clock Drawing Test

Background: Cognitive impairment is common and disabling in Parkinson’s disease (PD). Cognitive testing can be time consuming in the clinical setting. One rapid test to detect cognitive impairment in non-PD populations is the Clock Drawing Test (CDT), which calls upon the brain’s executive and visuospatial abilities to draw a clock designating a certain time. Objective: Test the hypothesis that PD participants would perform worse on CDT compared to controls and that CDT would correlate with other measures of cognition. Methods: This study evaluated two independent CDT scoring systems and differences in CDT performance between PD (N = 97) and control (N = 54) participants using a two-sample t-test. Pearson’s correlations were conducted between the CDT and tests of sleepiness (Epworth Sleepiness Scale) and vigilance (Psychomotor Vigilance Test); executive function (Trails B-A); and global cognition (Montreal Cognitive Assessment). Receiver operating characteristic curves were used to determine cut points on the CDT that identify individuals who need additional cognitive testing. Results: PD participants had worse performance on CDT compared to controls. The CDT was correlated with executive function (Trails B-A) and global cognition (Montreal Cognitive Assessment). The CDT correlated with vigilance (Psychomotor Vigilance Task) only in healthy controls. However, the CDT was not correlated with measures of sleepiness (Epworth Sleepiness Scale) in either group. A cut point of 9 on the Rouleau scale and 18 on the Mendez scale identified PD participants with cognitive impairment. Conclusion: The CDT is a rapid clinical cognitive assessment that is feasible in PD and correlates with other measures of cognition.

Faktor Prediktor Gangguan Kognitif 30 hari Pasca Stroke Iskemik Ringan-Sedang

Pendahuluan: Angka kejadian gangguan kognitif yang ditimbulkan akibat stroke iskemik saat ini semakin meningkat. Gangguan kognitif pascastroke iskemik seringkali terlambat didiagnosis. Penelitian yang mengidentifikasi faktor prediktor gangguan kognitif pascastroke iskemik akut masih terbatas dilaporkan di Indonesia. Tujuan penelitian ini adalah untuk mengidentifikasi faktor prediktor terhadap gangguan kognitif pada pasien pascastroke iskemik akut ringan-sedang.Metode: Penelitian kohort terhadap pasien berusia >18 tahun yang terdiagnosis stroke iskemik akut dan telah menjalani pemeriksaan Mini - Mental State Examination (MMSE) dan Clock Drawing Test (CDT) pada hari ke – 30 di Rumah Sakit Bethesda Yogyakarta diikutkan dalam penelitian. Luaran dari penelitian ini merupakan hasil MSSE dan CDT pada hari ke -30. Analisis penelitian ini menggunakan metode Chi-square intuk mengukur hubungan antara variabel bebas dengan variabel tergantung yang dilanjutkan dengan analisis multivariat regresi logistik. Nilai p <0,05 dianggap bermakna.Hasil: Sebanyak 140 pasien diikutkan dalam penelitian dengan rata-rata usia 62,8 tahun. Subjek berjenis kelamin laki – laki berjumlah 86 (61,4%) dan perempuan 54 (38,6%). Sembilan puluh satu subjek (65%) mengalami gangguan kognitif pascastroke iskemik akut. Analisis multivariat menunjukkan usia >70 tahun, tingkat pendidikan ≤ 6 tahun, skor Barhtel Index ≤4 dan skor mRS >3 saat terdiagnosis, jumlah lesi multipel dan lokasi lesi korteks merupakan faktor prediktor independen yang mempengaruhi gangguan kognitif 30 hari pascastroke iskemik akut. Kesimpulan: Usia >70 tahun, tingkat pendidikan ≤ 6 tahun, skor Barhtel Index ≤4 dan skor mRS >3 saat terdiagnosis, jumlah lesi multipel dan lokasi lesi korteks merupakan faktor prediktor independen terjadinya gangguan kognitif 30 hari pascastroke iskemik akut.