1. NCCU Academic Hub
  2. Publications
  3. College of Commerce
  4. NSC Projects
  5. Item

Publications-NSC Projects

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 數據驅動的學習與解讀：阿茲海默症
Data Driven Learning and Storytelling: Alzheimer＇s Disease
作者 周珮婷
貢獻者 統計系
關鍵詞 機器學習; 大數據; 非同質資料; 數據融合; 失智症; 阿茲海默; 老人; 長照; 人工智慧
Machine learning; Big data; Nonhomogeneous data; Data fusion; Dementia; AD; Elderly; Long-term care; AI
日期 2019-12
上傳時間 4-Jun-2026 09:08:14 (UTC+8)
摘要 世界人口老化，失智症病患大幅增加，花費在失智症的照護成本將大幅增加，阿茲海默症為失智症的一種，是老年人失智的主要原因，它是一種神經退化疾病，是一種進程緩慢、會隨時間惡化的持續性神經功能障礙。未來除了病人本身造成健保醫療上負擔外，長期照顧失智病患的家人的經濟壓力及生活品質、醫護人員的壓力，及要如何維持病人的生活功能與各階段生活品質維護，從預防、治療、復健、到末期至安寧照護，如何建立失智症照護的團隊與國家所能提供的幫助都是該被討論的，這些因病人造成的社會和經濟負擔將成為重要的公共健康問題。儘管失智症的重要性，卻相對較少的研究專注在病人認知功能衰退的過程，或解釋功能衰退軌跡的差異。Everyday Cognition （ECog; Farias et al., 2008）為長期追蹤資料，紀錄病人在主要生活領域和日常生活的基本活動功能行為。精細地描述正常生活功能到有失智之間的中間階段和過渡點。我們將以機器學習方法分析此資料與ADNI資料庫，以數據融合處理分析病人的認知功能下降過程及大腦結構退化過程。同時也分析病人與正常人的認知退化差異，及病人本身其他疾病與失智關係。這對早期診斷、預防、治療的早期治療具有重要的臨床意義，能了解哪些人未來可能發生認知功能問題，哪些生活領域將受到影響，功能衰退將如何進展，以及認知和非認知因素對功能性問題的貢獻，並且可以改善疾病進展的測量以及潛在的疾病改善治療的效果。失智前生活功能的解析也可以為診斷提供依據，並提供早期治療的機會，以適應老年人的需求，並幫助他們降低損失，進而幫助政府在長照規劃時的參考。 本研究計畫為一年期計畫，分析美國ECog資料及ADNI fMRI資料，為阿茲海默解碼，了解病人的生活功能、大腦退化、患病過程，未來近一步結合健保資料庫分析。以大數據分析與機器學習，進而建立醫療人工智慧。本計畫屬於跨領域計畫，包含統計、資訊、生醫、心理、認知科學領域。此類研究較少在長期追蹤量表資料上，但是此方法有潛力協助醫學診斷，建立預警系統，協助長照規劃，而且資料相對好獲取。本計畫預期提升台灣創新醫療之研究，並且配合政府5+2產業創新及國家科學技術發展四大計畫，發展醫療人工智能，建立國際研究團隊。
The number of elderly people is growing rapidly. Greater numbers of people will need long-term care as longevity rates increase. Health care costs will raise too. Alzheimer’s Disease (AD) is a major cause of disability in the elderly and represents a significant public health concern because of its associated personal, social, and economic burden. Despite its importance, relatively little research has focused on understanding the course of functional decline or the mechanisms that account for individual differences in trajectories of functional decline. ECog, a sensitive approach to measuring functional limitations can help researcher measure functional limitations across multiple domains. It shows that Mild Cognitive Impairment (MCI) is associated with functional limitations intermediate to normal aging and dementia. We propose to analyze the ECog data set with Machine Learning methods. Also, with data fusion methods, ECog data set will be combined with ADNI fMRI data set. The relevance of function decline and medical image will be analyzed. Furthermore, dementia cases from National health insurance research database, NHIRD will be compared with normal cases to find the association in the future. Furthermore, association analysis will be performed to other symptoms of dementia cases since dementia and other age-related pathologies may also lead to other clinical manifestations. The significant clinical impact of early diagnosis, prevention, treatment, understanding who may have cognitive problems in the future, which functions will be affected, how functional decline will progress, and the impact of cognitive and non-cognitive factors to functional decline. In addition, we can improve the measurement of disease progression as well as find the potential method to treat the disease. Detecting functional decline before dementia can also provide early diagnosis and proper treatment for elderly. Furthermore, the results of proposed grant can help the government to make decision about long-term care. It is an one-year project. ECog and ADNI dataset will be analyzed. Eventually, decoding dementia by analyzing functional decline and finding association of these datasets. Big data analysis and Machine learning techniques will be applied to build an artificial intelligence in medical diagnosis systems. The proposed project is multidisciplinary, which involves different subjects like statistics, computer science, biomedical, psychology, and neuroscience. Not too many research focus on longitudinal datasets like the proposed project does. The proposed project will provide a further help for medical diagnosis in building an alarm systems and provide assistance in long-term care systems.
關聯 科技部, MOST107-2118-M004-004, 107.08-108.07
資料類型 report
dc.contributor 統計系
dc.creator (作者) 周珮婷
dc.date (日期) 2019-12
dc.date.accessioned 4-Jun-2026 09:08:14 (UTC+8)-
dc.date.available 4-Jun-2026 09:08:14 (UTC+8)-
dc.date.issued (上傳時間) 4-Jun-2026 09:08:14 (UTC+8)-
dc.identifier.uri (URI) https://ah.lib.nccu.edu.tw/item?item_id=182733-
dc.description.abstract (摘要) 世界人口老化，失智症病患大幅增加，花費在失智症的照護成本將大幅增加，阿茲海默症為失智症的一種，是老年人失智的主要原因，它是一種神經退化疾病，是一種進程緩慢、會隨時間惡化的持續性神經功能障礙。未來除了病人本身造成健保醫療上負擔外，長期照顧失智病患的家人的經濟壓力及生活品質、醫護人員的壓力，及要如何維持病人的生活功能與各階段生活品質維護，從預防、治療、復健、到末期至安寧照護，如何建立失智症照護的團隊與國家所能提供的幫助都是該被討論的，這些因病人造成的社會和經濟負擔將成為重要的公共健康問題。儘管失智症的重要性，卻相對較少的研究專注在病人認知功能衰退的過程，或解釋功能衰退軌跡的差異。Everyday Cognition （ECog; Farias et al., 2008）為長期追蹤資料，紀錄病人在主要生活領域和日常生活的基本活動功能行為。精細地描述正常生活功能到有失智之間的中間階段和過渡點。我們將以機器學習方法分析此資料與ADNI資料庫，以數據融合處理分析病人的認知功能下降過程及大腦結構退化過程。同時也分析病人與正常人的認知退化差異，及病人本身其他疾病與失智關係。這對早期診斷、預防、治療的早期治療具有重要的臨床意義，能了解哪些人未來可能發生認知功能問題，哪些生活領域將受到影響，功能衰退將如何進展，以及認知和非認知因素對功能性問題的貢獻，並且可以改善疾病進展的測量以及潛在的疾病改善治療的效果。失智前生活功能的解析也可以為診斷提供依據，並提供早期治療的機會，以適應老年人的需求，並幫助他們降低損失，進而幫助政府在長照規劃時的參考。 本研究計畫為一年期計畫，分析美國ECog資料及ADNI fMRI資料，為阿茲海默解碼，了解病人的生活功能、大腦退化、患病過程，未來近一步結合健保資料庫分析。以大數據分析與機器學習，進而建立醫療人工智慧。本計畫屬於跨領域計畫，包含統計、資訊、生醫、心理、認知科學領域。此類研究較少在長期追蹤量表資料上，但是此方法有潛力協助醫學診斷，建立預警系統，協助長照規劃，而且資料相對好獲取。本計畫預期提升台灣創新醫療之研究，並且配合政府5+2產業創新及國家科學技術發展四大計畫，發展醫療人工智能，建立國際研究團隊。
dc.description.abstract (摘要) The number of elderly people is growing rapidly. Greater numbers of people will need long-term care as longevity rates increase. Health care costs will raise too. Alzheimer’s Disease (AD) is a major cause of disability in the elderly and represents a significant public health concern because of its associated personal, social, and economic burden. Despite its importance, relatively little research has focused on understanding the course of functional decline or the mechanisms that account for individual differences in trajectories of functional decline. ECog, a sensitive approach to measuring functional limitations can help researcher measure functional limitations across multiple domains. It shows that Mild Cognitive Impairment (MCI) is associated with functional limitations intermediate to normal aging and dementia. We propose to analyze the ECog data set with Machine Learning methods. Also, with data fusion methods, ECog data set will be combined with ADNI fMRI data set. The relevance of function decline and medical image will be analyzed. Furthermore, dementia cases from National health insurance research database, NHIRD will be compared with normal cases to find the association in the future. Furthermore, association analysis will be performed to other symptoms of dementia cases since dementia and other age-related pathologies may also lead to other clinical manifestations. The significant clinical impact of early diagnosis, prevention, treatment, understanding who may have cognitive problems in the future, which functions will be affected, how functional decline will progress, and the impact of cognitive and non-cognitive factors to functional decline. In addition, we can improve the measurement of disease progression as well as find the potential method to treat the disease. Detecting functional decline before dementia can also provide early diagnosis and proper treatment for elderly. Furthermore, the results of proposed grant can help the government to make decision about long-term care. It is an one-year project. ECog and ADNI dataset will be analyzed. Eventually, decoding dementia by analyzing functional decline and finding association of these datasets. Big data analysis and Machine learning techniques will be applied to build an artificial intelligence in medical diagnosis systems. The proposed project is multidisciplinary, which involves different subjects like statistics, computer science, biomedical, psychology, and neuroscience. Not too many research focus on longitudinal datasets like the proposed project does. The proposed project will provide a further help for medical diagnosis in building an alarm systems and provide assistance in long-term care systems.
dc.format.extent 116 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) 科技部, MOST107-2118-M004-004, 107.08-108.07
dc.subject (關鍵詞) 機器學習; 大數據; 非同質資料; 數據融合; 失智症; 阿茲海默; 老人; 長照; 人工智慧
dc.subject (關鍵詞) Machine learning; Big data; Nonhomogeneous data; Data fusion; Dementia; AD; Elderly; Long-term care; AI
dc.title (題名) 數據驅動的學習與解讀：阿茲海默症
dc.title (題名) Data Driven Learning and Storytelling: Alzheimer＇s Disease
dc.type (資料類型) report