| DC 欄位 |
值 |
語言 |
| DC.contributor | 資訊工程學系 | zh_TW |
| DC.creator | 張采庭 | zh_TW |
| DC.creator | Tsai-Ting Chang | en_US |
| dc.date.accessioned | 2022-8-11T07:39:07Z | |
| dc.date.available | 2022-8-11T07:39:07Z | |
| dc.date.issued | 2022 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=109522061 | |
| dc.contributor.department | 資訊工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 輕度認知障礙 (MCI) 和阿爾茨海默病 (AD) 都是對全球具有嚴重影響的認知疾病。有研究指出，如果這些症狀能夠及早被發現，將可以減緩疾病的惡化，甚至痊癒。目前常見的認知評估和診斷方法大多依賴人為操作，而且結果常常被質疑是否過於主觀，因此一種方便、客觀和準確率高的新型診斷方法是目前迫切需要的。隨著資訊爆炸時代的到來，虛擬現實(VR)和人工智慧(AI)的技術受到了高度的關注。本研究提出了一種智慧認知評估和診斷方法，利用機器學習(ML)和深度學習(DL)結合VR技術，打造了一款3D心理旋轉測試模組。與過去傳統的紙本和2D心理旋轉測試相比，它收集了使用者在解題時的數據，並且可以通過操作手把玩成模組任務，藉此來進行空間認知訓練，同時會記錄答題過程中的手把操作狀態。通過這些原始數據和設計的192個特徵，再使用AI 技術構建分類器，能夠將MCI或AD患者與健康人區分開來，最後完成一個準確率約90%的認知功能智慧診斷模型，這也證明了空間訓練在認知評估中的可行性。 | zh_TW |
| dc.description.abstract | Mild Cognitive Impairment (MCI) and Alzheimer′s disease (AD) are both cognitive diseases with enormous global impact. The studies have shown that if these symptoms are detected and acted on early, disease progression can be slowed or even cured. The current common cognitive assessment and screening methods rely more on manual operations, and the results are often questioned as being overly subjective. As a result, a convenient, objective and highly accurate diagnostic method is necessary. With the advent of the era of information explosion, the technologies of virtual reality (VR) and artificial intelligence (AI) have received leaping attention. This study proposes an intelligent cognitive assessment and diagnosis method. Utilizing machine learning and deep learning combined with VR technology to create a 3D mental rotation test module. Compared with the traditional paper and 2D mental rotation tests in the past, it has collected richer user-solving data. Users can conduct spatial cognition training on this module by manipulating the handle, and at the same time record the operation status in the process. Through these raw data and the 192 designed features, AI technology was integrated to construct a classifier to distinguish MCI or AD patients from healthy people. Finally, an intelligent diagnostic model of cognitive function with an accuracy rate of nearly 90% was integrated, which also proved the feasibility of spatial cognition in cognitive assessment. | en_US |
| DC.subject | 心理旋轉(MR) | zh_TW |
| DC.subject | 認知障礙 | zh_TW |
| DC.subject | 虛擬現實(VR) | zh_TW |
| DC.subject | 機器學習(ML) | zh_TW |
| DC.subject | 深度學習(DL) | zh_TW |
| DC.subject | Mental Rotation (MR) | en_US |
| DC.subject | Cognitive Impairment | en_US |
| DC.subject | Virtual reality (VR) | en_US |
| DC.subject | Machine Learning (ML) | en_US |
| DC.subject | Deep Learning (DL) | en_US |
| DC.title | 基於深度學習模型的3D心理旋轉對認知障礙的診斷與評估 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | A Deep Learning-based Diagnosis and Assessment of Cognitive Impairment with 3D Mental Rotation | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |