隨著人口老化以及失智症盛行率的提升，失智症越來越受到重視。許多研究提到加強預防和早期發現是應對失智症的關鍵，但因為失智症的症狀與疾病進程複雜，因此診斷的方式一直是個臨床挑戰。目前台灣臨床上最常使用是以Cognitive Abilities Screening Instrument (CASI)去評估患者的認知功能，但使用CASI的缺點為檢測過程耗時和專業人力需求，加上神經心理測驗師人力缺乏，目前通常病患要進行評估需要等待相當時間，所以需要一種快速、無須專業，且又能評估多項認知向度的測驗。在許多認知測驗中，畫時鐘測驗(clock drawing test，簡稱CDT)被認為是一項可以鑑別多樣認知功能的簡易測驗，但是目前僅限用於大略評估認知狀態。近年來相關研究指出，透過數位式CDT(dCDT)可以從繪圖過程獲取更多資訊。本篇研究提出了一種利用 dCDT 提取的特徵，透過以 XGBoost 作為適應度函數自動估測之基因演算法，用以預測 CASI 的九項認知向度分數的方法。此外，本研究也探討了繪圖特徵與各種認知向度間的相關性。在九項認知向度的預測中，本研究所提出的方法可以有效預測短期記憶、注意力、心理操作和抽象思考等分數，其中所選的重要特徵與先前的研究文獻一致。然而，對於遠程記憶和繪圖的領域，雖然預測的誤差小，但由於資料集的不均勻分佈，有其模型在各向度分數預測之準度限制，對於定向感、語言和言語流暢性的向度獲得了較差的預測結果。綜合上述，本項研究建立與探討dCDT評估多項認知功能的可能性與局限性。

With the aging population and increasing prevalence of dementia, dementia has attracted growing attention. Many studies emphasize that strengthening prevention and early detection are key to managing dementia. However, diagnosing dementia remains a clinical challenge due to the complexity of its symptoms and disease progression. Currently in Taiwan, the Cognitive Abilities Screening Instrument (CASI) is the most used clinically to assess patients' cognitive function. Yet, CASI's drawbacks include time-consuming testing processes and the need for trained professionals. Moreover, there is a shortage of neuropsychologists, resulting in significant waiting times for assessments. Therefore, there is a need for a rapid, non-professionally administered test that can evaluate multiple cognitive dimensions. Among various cognitive tests, the Clock Drawing Test (CDT) is considered a simple test capable of identifying diverse cognitive functions, albeit primarily used for rough cognitive status assessment. In recent years, studies have suggested that using a digital version of CDT (dCDT) can yield more information from the drawing process. This study proposes a method that utilizes features extracted from dCDT to predict scores on nine cognitive dimensions of CASI using a genetic algorithm with XGBoost as the fitness function estimation. Additionally, the study explores the correlations between drawing features and various cognitive dimensions. In predicting the nine cognitive dimensions, the proposed method effectively predicts scores for short-term memory, attention, mental manipulation, and abstract thinking using the previously selected important features. However, for remote memory and drawing domains, despite lower prediction errors, limitations in model accuracy across dimension scores are observed due to the uneven distribution of the dataset. Particularly, poorer predictive results were obtained for orientation, language, and verbal fluency dimensions. In summary, this study establishes and discusses the potential and limitations of using dCDT to assess multiple cognitive functions.

[1] W. H. Organization, "Global status report on the public health response to dementia," 2021.
[2]台灣失智症協會. (n.d.). “關於失智症-認識失智症,” 台灣失智症協會, http://www.tada2002.org.tw/About/IsntDementia, Mar. 2023 (accessed Jun. 2024).
[3] A. Wimo, K. Seeher, R. Cataldi, et al., "The worldwide costs of dementia in 2019," Alzheimers & Dementia, vol. 19, no. 7, pp. 2865-2873, 2023.
[4] K. S. Santacruz and D. Swagerty, "Early diagnosis of dementia," American Family Physician, vol. 63, no. 4, pp. 703-714, 2001.
[5] B. Reisberg, "Diagnostic criteria in dementia: a comparison of current criteria, research challenges, and implications for DSM-V," Geriatric Psychiatry and Neurology, vol. 19, no. 3, pp. 137-46, Sep. 2006.
[6] D. Wechsler, “WPPSI-III administration and scoring manual,” Psychological Corporation, 2002.
[7] T. N. Tombaugh, "Trail making test A and B: normative data stratified by age and education," Archives of Clinical Neuropsychology, vol. 19, no. 2, pp. 203-214, 2004.
[8] B. W. Williams, W. Mack, and V. W. Henderson, "Boston naming test in Alzheimer's disease," Neuropsychologia, vol. 27, no. 8, pp. 1073-1079, 1989.
[9] I. Arevalo‐Rodriguez, N. Smailagic, M. R. Figuls, et al., "Mini‐mental state examination (MMSE) for the detection of Alzheimer's disease and other dementias in people with mild cognitive impairment (MCI)," Cochrane Database of Systematic Reviews, no. 3, 2015.
[10] R. H. Benedict, D. Schretlen, L. Groninger, M. Dobraski, and B. Shpritz, "Revision of the brief visuospatial memory test: studies of normal performance, reliability, and validity," Psychological Assessment, vol. 8, no. 2, pp. 145, 1996.
[11] M. deLeon and G. Small, "Diagnosis of dementia," Neurobiology of Aging, vol. 22, no. 2, pp. 332-332, Apr. 2001.
[12] D. S. Knopman and R. C. Petersen, "Mild cognitive impairment and mild dementia: a clinical perspective," Mayo Clin Proc, vol. 89, no. 10, pp. 1452-1459, Oct. 2014.
[13] Z. S. Nasreddine, N. A. Phillips, V. Bédirian, et al., "The montreal cognitive assessment, MoCA: A brief screening tool for mild cognitive impairment," Journal of the American Geriatrics Society, vol. 53, no. 4, pp. 695-699, Apr. 2005.
[14] K. N. Lin, P. N. Wang, H. C. Liu, and E. L. Teng, "Cognitive abilities screening instrument, Chinese version 2.0 (CASI C-2.0): administration and clinical application," Acta Neurol Taiwan, vol. 21, no. 4, pp. 180-189, Dec. 2012.
[15] R. C. Tsai, K. N. Lin, H. J. Wang and H. C. Liu, "Evaluating the uses of the total score and the domain scores in the cognitive abilities screening instrument, Chinese version (CASI C-2.0): results of confirmatory factor analysis," International Psychogeriatrics, vol. 19, no. 6, pp. 1051-1063, Dec. 2007.
[16] E. Pinto and R. Peters, “Literature review of the clock drawing test as a tool for cognitive screening,” Dementia and Geriatric Cognitive Disorders, vol. 27, no. 3, pp. 201-213, 2009.
[17] M. Rakusa, J. Jensterle and J. Mlakar, "Clock drawing test: a simple scoring system for the accurate screening of cognitive impairment in patients with mild cognitive impairment and dementia," Dementia and Geriatric Cognitive Disorders, vol. 45, no. 5-6, pp. 326-334, 2018.
[18] D. J. Libon, R. A. Swenson, E. J. Barnoski and L. P. Sands, "Clock drawing as an assessment tool for dementia," Archives of Clinical Neuropsychology, vol. 8, no. 5, pp. 405-415, Oct. 1993.
[19] H. F. Reichenfeld and G. A. Wells, "Clock drawing: a neuropsychological analysis," Journal of Psychiatry and Neuroscience, vol. 20, no. 2, p. 155, 1995.
[20] S. Kim, S. Jahng, K. Yu, B. Lee and Y. Kang, “Usefulness of the clock drawing test as a cognitive screening instrument for mild cognitive impairment and mild dementia: an evaluation using three scoring systems,” Dementia and Neurocognitive Disorders, vol. 17, no. 3, pp. 100-109, 2018.
[21] L. Ehreke, M. Luppa, H. H. König and S. G. Riedel-Heller, “Is the clock drawing test a screening tool for the diagnosis of mild cognitive impairment? A systematic review,” International Psychogeriatrics, vol. 22, no. 1, pp. 56-63, 2010.
[22] I. Park and U. Lee, “Automatic, qualitative scoring of the clock drawing test (CDT) based on U-Net, CNN and mobile sensor data,” Sensors, vol. 21, no. 15, pp. 32, Aug. 2021.
[23] D. Handzlik, L. L. Richmond, S. Skiena, M. A. Carr, S. A. P. Clouston and B. J. Luft, "Explainable automated evaluation of the clock drawing task for memory impairment screening," Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, vol. 15, no. 2, Apr. 2023.
[24] A. Toomela, “Digital clock drawing test implementation and analysis,” M.S. thesis, Faculty of Information Technology Institute of Computer Science, TalTech IT College, Estonia, 2017.
[25] S. Amini, L. F. Zhang, B. R. Hao, A. Gupta, M. T. Song, C. Karjadi, H. H. Lin, V. B. Kolachalama, R. Au and I. C. Paschalidis, “An artificial intelligence-assisted method for dementia detection using images from the clock drawing test,” Journal of Alzheimers Disease, vol. 83, no. 2, pp. 581-589, 2021.
[26] X. Zheng, W. Zhang, X. Wang, et al., "Extended application of digital clock drawing test in the evaluation of Alzheimer's disease based on artificial intelligence and the neural basis," Current Alzheimer Research, vol. 18, no. 14, pp. 1127-1139, 2021.
[27] W. Souillard-Mandar, R. Davis, C. Rudin, R. Au, D. J. Libon, R. Swenson, C. C. Price, M. Lamar and D. L. Penney, “Learning classification models of cognitive conditions from subtle behaviors in the digital clock drawing test,” Machine Learning, vol. 102, no. 3, pp. 393-441, Mar. 2016.
[28] R. Binaco, N. Calzaretto, J. Epifano, S. McGuire, M. Umer, S. Emrani, V. Wasserman, D. J. Libon and R. Polikar, “Machine learning analysis of digital clock drawing test performance for differential classification of mild cognitive impairment subtypes versus Alzheimer’s disease,” Journal of the International Neuropsychological Society, vol. 26, no. 7, pp. 690-700, 2020.
[29] C. Dion, F. Arias, S. Amini, R. Davis, D. Penney, D. J. Libon and C. C. Price, “Cognitive correlates of digital clock drawing metrics in older adults with and without mild cognitive impairment,” Journal of Alzheimers Disease, vol. 75, no. 1, pp. 73-83, 2020.
[30] R. J. Piers, K. N. Devlin, B. T. Ning, Y. L. Liu, B. Wasserman, J. M. Massaro, M. Lamar, C. C. Price, R. Swenson, R. Davis, D. L. Penney, R. Au and D. J. Libon, “Age and graphomotor decision making assessed with the digital clock drawing test: The Framingham Heart Study,” Journal of Alzheimers Disease, vol. 60, no. 4, pp. 1611-1620, 2017.
[31] J. Yuan, D. J. Libon, C. Karjadi, A. F. A. Ang, S. Devine, S. H. Auerbach, R. Au and H. H. Lin, “Association between the digital clock drawing test and neuropsychological test performance: Large community-based prospective cohort (Framingham Heart Study),” Journal of Medical Internet Research, vol. 23, no. 6, pp. 9, Jun. 2021.
[32] M. Shehab, L. Abualigah, Q. Shambour, M. A. Abu-Hashem, M. K. Y. Shambour, A. I. Alsalibi and A. H. Gandomi, "Machine learning in medical applications: a review of state-of-the-art methods," Computers in Biology and Medicine, vol. 145, pp.105458, Jun. 2022.
[33] M. M. Ahsan, S. A. Luna and Z. Siddique, "Machine learning based disease diagnosis: a comprehensive review," Healthcare, vol. 10, no. 3, pp.541, Mar. 2022.
[34] R. A. Buckley, K. J. Atkins, E. Fortunato, B. Silbert, D. A. Scott and L. Evered, "A novel digital clock drawing test as a screening tool for perioperative neurocognitive disorders: A feasibility study," Acta Anaesthesiologica Scandinavica, vol. 65, no. 4, pp. 473-480, Apr. 2021.
[35] J. Redmon, S. Divvala, R. Girshick and A. Farhadi, "You only look once: unified, real-time object detection," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 779-788.
[36] C. Cervellera and D. Macciò, "Distribution preserving stratified sampling for learning problems," IEEE Transactions on Neural Networks and Learning Systems, vol. 29, no. 7, pp. 2886-2895, 2017.
[37] A. Bommert, X. D. Sun, B. Bischl, J. Rahnenführer and M. Lang, "Benchmark for filter methods for feature selection in high-dimensional classification data," Computational Statistics & Data Analysis, vol. 143, Mar. 2020.
[38] R. Kohavi and G. H. John, "Wrappers for feature subset selection," Artificial Intelligence, vol. 97, no. 1-2, pp. 273-324, Dec. 1997.
[39] C. W. Chen, Y. H. Tsai, F. R. Chang and W. C. Lin, "Ensemble feature selection in medical datasets: Combining filter, wrapper, and embedded feature selection results," Expert Systems, vol. 37, no. 5, Oct. 2020.
[40] Z. Arabasadi, R. Alizadehsani, M. Roshanzamir, H. Moosaei and A. A. Yarifard, "Computer aided decision making for heart disease detection using hybrid neural network-Genetic algorithm," Computer Methods and Programs in Biomedicine, vol. 141, pp. 19-26, Apr. 2017.
[41] R. Cerf, "The Quasispecies Regime for the Simple Genetic Algorithm with Roulette-Wheel Selection," Advances in Applied Probability, vol. 49, no. 3, pp. 903-926, Sep. 2017.
[42] T. T. Wong and P. Y. Yeh, "Reliable accuracy estimates from k-fold cross validation," IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 8, pp. 1586-1594, Aug. 2020.
[43] T. Chen and C. Guestrin, "Xgboost: A scalable tree boosting system," Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785-794.
[44] C. S. Zhang, Y. Zhang, X. J. Shi, G. Almpanidis, G. J. Fan and X. J. Shen, "On incremental learning for gradient boosting decision trees," Neural Processing Letters, vol. 50, no. 1, pp. 957-987, Aug. 2019.
[45] S. K. Henderson, K. A. Peterson, K. Patterson, M. A. Lambon Ralph and J. B. Rowe, "Verbal fluency tests assess global cognitive status but have limited diagnostic differentiation: Evidence from a large-scale examination of six neurodegenerative diseases," Brain Commun, vol. 5, no. 2, 2023.