請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/91843
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 盧子彬 | zh_TW |
dc.contributor.advisor | Tzu-Pin Lu | en |
dc.contributor.author | 吳昱弘 | zh_TW |
dc.contributor.author | Yu-Hung Wu | en |
dc.date.accessioned | 2024-02-23T16:15:29Z | - |
dc.date.available | 2024-02-24 | - |
dc.date.copyright | 2024-02-23 | - |
dc.date.issued | 2024 | - |
dc.date.submitted | 2024-01-29 | - |
dc.identifier.citation | 行政院,國家發展委員會(2022)。中華民國人口推估(2022年至2070年)。臺北市:國家發展委員會。
Lin, Y.-Y., & Huang, C.-S. (2015). Aging in Taiwan: Building a Society for Active Aging and Aging in Place. The Gerontologist, 56(2), 176-183. https://doi.org/10.1093/geront/gnv107 徐敏鐘, & 廖珮如. (2015). Financing National Health Insurance: The Challenge of Fast Population Aging [全民健康保險:快速人口老化之挑戰]. 經濟論文叢刊, 43(2), 145-182. https://doi.org/10.6277/ter.2015.432.1 MacNee, W., Rabinovich, R. A., & Choudhury, G. (2014). Ageing and the border between health and disease. Eur Respir J, 44(5), 1332-1352. https://doi.org/10.1183/09031936.00134014 Xia, X., Wang, Y., Yu, Z., Chen, J., & Han, J. J. (2021). Assessing the rate of aging to monitor aging itself. Ageing Res Rev, 69, 101350. https://doi.org/10.1016/j.arr.2021.101350 Lowsky, D. J., Olshansky, S. J., Bhattacharya, J., & Goldman, D. P. (2014). Heterogeneity in healthy aging. J Gerontol A Biol Sci Med Sci, 69(6), 640-649. https://doi.org/10.1093/gerona/glt162 Fuellen, G., Jansen, L., Cohen, A. A., Luyten, W., Gogol, M., Simm, A., Saul, N., Cirulli, F., Berry, A., Antal, P., Köhling, R., Wouters, B., & Möller, S. (2019). Health and Aging: Unifying Concepts, Scores, Biomarkers and Pathways. Aging Dis, 10(4), 883-900. https://doi.org/10.14336/ad.2018.1030 López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M., & Kroemer, G. (2013). The hallmarks of aging. Cell, 153(6), 1194-1217. https://doi.org/10.1016/j.cell.2013.05.039 Cole, J. H., Ritchie, S. J., Bastin, M. E., Valdés Hernández, M. C., Muñoz Maniega, S., Royle, N., Corley, J., Pattie, A., Harris, S. E., Zhang, Q., Wray, N. R., Redmond, P., Marioni, R. E., Starr, J. M., Cox, S. R., Wardlaw, J. M., Sharp, D. J., & Deary, I. J. (2018). Brain age predicts mortality. Molecular Psychiatry, 23(5), 1385-1392. https://doi.org/10.1038/mp.2017.62 Wang, J., Knol, M. J., Tiulpin, A., Dubost, F., de Bruijne, M., Vernooij, M. W., Adams, H. H. H., Ikram, M. A., Niessen, W. J., & Roshchupkin, G. V. (2019). Gray Matter Age Prediction as a Biomarker for Risk of Dementia. Proceedings of the National Academy of Sciences, 116(42), 21213-21218. https://doi.org/doi:10.1073/pnas.1902376116 Xia, X., Chen, X., Wu, G., Li, F., Wang, Y., Chen, Y., Chen, M., Wang, X., Chen, W., Xian, B., Chen, W., Cao, Y., Xu, C., Gong, W., Chen, G., Cai, D., Wei, W., Yan, Y., Liu, K., . . . Han, J.-D. J. (2020). Three-dimensional facial-image analysis to predict heterogeneity of the human ageing rate and the impact of lifestyle. Nature Metabolism, 2(9), 946-957. https://doi.org/10.1038/s42255-020-00270-x Horvath, S., & Raj, K. (2018). DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nature Reviews Genetics, 19(6), 371-384. https://doi.org/10.1038/s41576-018-0004-3 Zhu, Z., Shi, D., Guankai, P., Tan, Z., Shang, X., Hu, W., Liao, H., Zhang, X., Huang, Y., Yu, H., Meng, W., Wang, W., Ge, Z., Yang, X., & He, M. (2023). Retinal age gap as a predictive biomarker for mortality risk. Br J Ophthalmol, 107(4), 547-554. https://doi.org/10.1136/bjophthalmol-2021-319807 Crimmins, E. M. (2015). Lifespan and Healthspan: Past, Present, and Promise. Gerontologist, 55(6), 901-911. https://doi.org/10.1093/geront/gnv130 Lin, J. B., Tsubota, K., & Apte, R. S. (2016). A glimpse at the aging eye. npj Aging and Mechanisms of Disease, 2(1), 16003. https://doi.org/10.1038/npjamd.2016.3 Ma, J., Xu, X., Li, M., Zhang, Y., Zhang, L., Ma, P., Hou, J., Lei, Y., Liu, J., Huangfu, X., Yang, Y., Yi, X., Cheng, G., Bai, J., Zhong, X., Xu, X., & Wang, Y. (2021). Predictive models of aging of the human eye based on ocular anterior segment morphology. Journal of Biomedical Informatics, 120, 103855. https://doi.org/https://doi.org/10.1016/j.jbi.2021.103855 Salvi, S. M., Akhtar, S., & Currie, Z. (2006). Ageing changes in the eye. Postgrad Med J, 82(971), 581-587. https://doi.org/10.1136/pgmj.2005.040857 Grossniklaus, H. E., Nickerson, J. M., Edelhauser, H. F., Bergman, L. A., & Berglin, L. (2013). Anatomic alterations in aging and age-related diseases of the eye. Invest Ophthalmol Vis Sci, 54(14), Orsf23-27. https://doi.org/10.1167/iovs.13-12711 Sparrow, J. R., & Duncker, T. (2014). Fundus Autofluorescence and RPE Lipofuscin in Age-Related Macular Degeneration. J Clin Med, 3(4), 1302-1321. https://doi.org/10.3390/jcm3041302 Crawford, T. J., Smith, E. S., & Berry, D. M. (2017). Eye Gaze and Aging: Selective and Combined Effects of Working Memory and Inhibitory Control. Front Hum Neurosci, 11, 563. https://doi.org/10.3389/fnhum.2017.00563 Lee, W. J., Kim, J. H., Shin, Y. U., Hwang, S., & Lim, H. W. (2019). Differences in eye movement range based on age and gaze direction. Eye (Lond), 33(7), 1145-1151. https://doi.org/10.1038/s41433-019-0376-4 Dalrymple, K. A., Jiang, M., Zhao, Q., & Elison, J. T. (2019). Machine learning accurately classifies age of toddlers based on eye tracking. Scientific Reports, 9(1), 6255. https://doi.org/10.1038/s41598-019-42764-z Li, Y., Xia, X., & Paulus, Y. M. (2018). Advances in Retinal Optical Imaging. Photonics, 5(2). https://doi.org/10.3390/photonics5020009 Walker, H. K., Hall, W. D., & Hurst, J. W. (Eds.). (1990). Clinical Methods: The History, Physical, and Laboratory Examinations. (3rd ed.). Butterworths. Abu Sbeit, R. H. S., Mohammed, O. A., & Alamlih, L. I. (2021). Hypertension as an Unusual Cause of Proliferative Retinopathy: Case Report and Literature Review. Case Rep Ophthalmol, 12(2), 594-602. https://doi.org/10.1159/000515284 Viswanath, K., & McGavin, D. D. (2003). Diabetic retinopathy: clinical findings and management. Community Eye Health, 16(46), 21-24. Heckenlively, J. R., & Ferreyra, H. A. (2008). Autoimmune retinopathy: a review and summary. Semin Immunopathol, 30(2), 127-134. https://doi.org/10.1007/s00281-008-0114-7 Patton, N., Aslam, T., Macgillivray, T., Pattie, A., Deary, I. J., & Dhillon, B. (2005). Retinal vascular image analysis as a potential screening tool for cerebrovascular disease: a rationale based on homology between cerebral and retinal microvasculatures. J Anat, 206(4), 319-348. https://doi.org/10.1111/j.1469-7580.2005.00395.x Koronyo, Y., Biggs, D., Barron, E., Boyer, D. S., Pearlman, J. A., Au, W. J., Kile, S. J., Blanco, A., Fuchs, D. T., Ashfaq, A., Frautschy, S., Cole, G. M., Miller, C. A., Hinton, D. R., Verdooner, S. R., Black, K. L., & Koronyo-Hamaoui, M. (2017). Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer's disease. JCI Insight, 2(16). https://doi.org/10.1172/jci.insight.93621 Kim, Y. D., Noh, K. J., Byun, S. J., Lee, S., Kim, T., Sunwoo, L., Lee, K. J., Kang, S.-H., Park, K. H., & Park, S. J. (2020). Effects of Hypertension, Diabetes, and Smoking on Age and Sex Prediction from Retinal Fundus Images. Scientific Reports, 10(1), 4623. https://doi.org/10.1038/s41598-020-61519-9 Bourne, R. R. (2011). Ethnicity and ocular imaging. Eye (Lond), 25(3), 297-300. https://doi.org/10.1038/eye.2010.187 Fu, H., Wang, B., Shen, J., Cui, S., Xu, Y., Liu, J., & Shao, L. (2019). Evaluation of retinal image quality assessment networks in different color-spaces. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I 22 (pp. 48-56). Springer International Publishing. Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. Chen, W., Qian, W., Wu, G., Chen, W., Xian, B., Chen, X., Cao, Y., Green, C. D., Zhao, F., Tang, K., & Han, J.-D. J. (2015). Three-dimensional human facial morphologies as robust aging markers. Cell research, 25(5), 574-587. https://doi.org/10.1038/cr.2015.36 Weidner, C. I., Lin, Q., Koch, C. M., Eisele, L., Beier, F., Ziegler, P., Bauerschlag, D. O., Jöckel, K.-H., Erbel, R., Mühleisen, T. W., Zenke, M., Brümmendorf, T. H., & Wagner, W. (2014). Aging of blood can be tracked by DNA methylation changes at just three CpG sites. Genome Biology, 15(2), R24. https://doi.org/10.1186/gb-2014-15-2-r24 Hannum, G., Guinney, J., Zhao, L., Zhang, L., Hughes, G., Sadda, S., Klotzle, B., Bibikova, M., Fan, J. B., Gao, Y., Deconde, R., Chen, M., Rajapakse, I., Friend, S., Ideker, T., & Zhang, K. (2013). Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell, 49(2), 359-367. https://doi.org/10.1016/j.molcel.2012.10.016 Ahadi, S., Wilson, K. A., Babenko, B., McLean, C. Y., Bryant, D., Pritchard, O., Kumar, A., Carrera, E. M., Lamy, R., Stewart, J. M., Varadarajan, A., Berndl, M., Kapahi, P., & Bashir, A. (2023). Longitudinal fundus imaging and its genome-wide association analysis provide evidence for a human retinal aging clock. Elife, 12. https://doi.org/10.7554/eLife.82364 國家衛生研究院(2021)。2017年國民健康訪問調查結果報告。臺北市:國家衛生研究院。 Wong, T. Y., Islam, F. M., Klein, R., Klein, B. E., Cotch, M. F., Castro, C., Sharrett, A. R., & Shahar, E. (2006). Retinal vascular caliber, cardiovascular risk factors, and inflammation: the multi-ethnic study of atherosclerosis (MESA). Invest Ophthalmol Vis Sci, 47(6), 2341-2350. https://doi.org/10.1167/iovs.05-1539 Kelty, P. J., Payne, J. F., Trivedi, R. H., Kelty, J., Bowie, E. M., & Burger, B. M. (2008). Macular thickness assessment in healthy eyes based on ethnicity using Stratus OCT optical coherence tomography. Invest Ophthalmol Vis Sci, 49(6), 2668-2672. https://doi.org/10.1167/iovs.07-1000 | - |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/91843 | - |
dc.description.abstract | 研究背景:
根據台灣國家發展委員會的推估,台灣65歲以上的老年人口將在2025達到總人口數的20%,即將步入「超高齡社會」。這樣這種人口結構會對醫療體系造成龐大的壓力,為了因應這樣的困境,台灣積極推動長者健康老化、延緩失能等關鍵政策。生理年齡是身體器官老化的重要風險因子,雖然已經開發了各種判斷生理年裡的生物標記,如DNA甲基化狀態、大腦年齡等,但這些方法除了需要侵入性的採集檢體外,收費昂貴且需要時間來等待預測結果。因此本研究的目的想尋找一個可以非侵入性、成本低廉和迅速判斷年齡的方法。 眼睛是一個容易受到年齡變化影響的器官,隨著非侵入性眼底成像技術的進展,視網膜疾病診斷的準確率也顯著提高,此外眼底影像可以使血管成像並判斷全身健康狀態。在國外已經有使用深度學習方法以視網膜眼底鏡影像進行年齡預測的成功案例,但目前尚未有針對台灣人口的相關研究,因此本研究將使用來自台大醫院資料庫的眼底鏡影像來建立一個可以預測年齡的機器學習模型。 方法: 本研究中,我們使用了來自台大醫院的資料庫的眼底鏡影像資料,其中包含80,001張眼底鏡影像,還有其相關的年齡、性別和診斷資訊。這些資料來自於2017年至2021年間在台大醫院接受眼底檢查的8,907名受試者,資料已經過去識別化處理。排除資料缺失、年齡超過90歲和低於30歲的受試者,只保留高品質的眼底鏡影像後,我們共留下了來自7,798位受試者的 51,618張眼底鏡影像。從這些受試者進行的15,577次眼底檢查中,我們選擇了每次檢查中品質最好的眼底鏡影像。排除相關疾病的診斷資訊病史後,總共有來自7,174名受試者的14,307張眼底鏡影像會被用來進行生理年齡預測深度學習模型的訓練、驗證及測試。 我們使用VGG16結構作為年齡預測模型的基礎,加入性別和近視等變數進行調整,使用平均絕對誤差(MAE)作為評估模型表現的損失函數,Grad-CAM則用來識別輸入的眼底鏡影像中對生理年齡預測結果貢獻最多的區域。 結果: 在我們資料集來自7174名受試者的14,307次眼底檢查中,有59.4%的受試者是女性,40.6%為男性。近視的參與者占全部資料中的16%,受試者接受眼底檢查的平均年齡為59.0歲,標準差為13.26歲。 在評估年齡預測模型的表現時,未加入性別及近視變數調整的模型MAE為6.28歲,模型預測年齡的平均誤差為6.28歲,實際年齡與預測年齡的相關係數為0.79 (95%CI,0.76-0.81)。進行性別和近視的調整後,預測的準確性有明顯的提高,調整後的模型MAE為4.58歲,平均絕對誤差降低許多,實際年齡與預測年齡的相關係數也提到到0.90 (95%CI,0.88-0.91)。而Grad-CAM圖像也顯現出了視網膜血管對年齡預測模型影像的重要性。 結論: 本研究建立一個深度學習模型,使用來自台灣30 到 90 歲成年人的眼底鏡影像,進行年齡的預測。與其他研究比較後,我們的模型與其他侵入性方法相比(如腦部MRI、3D面部影像和DNA甲基化時鐘研究)預測年齡的表現非常出色。此外,跟上述方法相比,眼底鏡的成本更低、產生年齡預測結果的速度更快,因此希望這項研究可以成為台灣人口預測年齡的新途徑。 | zh_TW |
dc.description.abstract | Introduction:
As Taiwan anticipates a notable increase in the proportion of individuals aged 65 and above, projected to reach around 20% by 2025, the nation is on the brink of becoming a super-aged society, posing challenges to healthcare systems. To address this, Taiwan is actively promoting healthy aging. While chronological age is a recognized risk factor for age-related health issues, variations in health outcomes among individuals of the same age emphasize the importance of assessing biological age for accurate risk stratification and personalized interventions. This study aims to pioneer a non-invasive, cost-effective, and expedited approach for assessing aging using fundus images. Despite existing methods like brain age and DNA methylation, which are invasive and time-consuming, this research focuses on leveraging non-invasive fundus images for efficient age assessment. The eye, susceptible to age-related changes, serves as a significant risk factor for various eye conditions. Recent advancements in non-invasive fundus imaging have improved the diagnosis of retinal diseases, offering insights into systemic diseases by visualizing blood vessels and reflecting systemic circulation. While deep learning models have been successfully applied in age prediction using retinal fundus images, no specific studies have targeted the Taiwanese population. This research introduces a deep learning model tailored for predicting age based on fundus images from Taiwanese individuals, aiming to be non-invasive, cost-effective, and efficient. In summary, this study addresses Taiwan's demographic challenges, emphasizing the importance of biological age assessment. By utilizing non-invasive fundus images and deep learning modeling, it contributes to a novel approach for estimating human age, with potential applications in personalized healthcare. Methods: The study employed a dataset of 80,001 fundus images and relevant demographic details from individuals who underwent fundus examinations at the National Taiwan University Hospital between 2017 and 2021. Data cleansing involved the removal of incomplete, irrelevant, and low-quality data, resulting in a dataset of 74,018 fundus images from 8,276 individuals. After quality checks, the dataset was further narrowed down to include 51,618 images of acceptable quality from 7,798 participants. The highest quality fundus image from each of the 15,577 fundus examination visits was selected, resulting in 14,307 images from 7,174 participants without a reported medical history of disease. These images were used to construct the deep learning model for age prediction. The model assumed chronological age equates to biological age and utilized the dataset, randomly divided into training, validation, and testing sets. It employed the VGG16 network with adjustments for sex and myopia, utilizing the Root Mean Square Propagation as the optimization algorithm. A dropout layer was introduced to mitigate potential overfitting, and model performance was assessed using the mean absolute value (MAE) as the loss function. Grad-CAM was employed for interpretability. Results: The age prediction model, developed through deep learning, engaged a dataset of 14,307 visits from 7,174 participants at the National Taiwan University Hospital, with a gender distribution of 59.4% female and 40.6% male participants. Myopic participants constituted 16% of the dataset, with an average chronological age of 59.0 years. The initial model without sex and myopia adjustments yielded a mean absolute error of 6.28 years and a correlation coefficient of 0.79. The refined model, incorporating adjustments for sex and myopia, showed improved accuracy, with a reduced mean absolute error of 4.58 years and an elevated correlation coefficient of 0.90. Grad-CAM heat-maps highlighted the influence of retinal arcade vessels in the age prediction model. Conclusion: This study introduces a novel deep learning model using fundus images from a diverse Taiwanese population aged 30 to 90 to predict age with precision. Comparative analyses showcase its competitiveness against invasive methods, emphasizing its cost-effectiveness and efficiency. This pioneering study positions the model as a promising avenue for Taiwanese population age prediction. | en |
dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2024-02-23T16:15:29Z No. of bitstreams: 0 | en |
dc.description.provenance | Made available in DSpace on 2024-02-23T16:15:29Z (GMT). No. of bitstreams: 0 | en |
dc.description.tableofcontents | 誌謝 i
中文摘要 ii ABSTRACT iv CONTENTS vii LIST OF FIGURES ix LIST OF TABLES xi Chapter 1 Introduction 1 1.1 A Focus on Age Assessment 1 1.2 The Potential of Ocular Structures for Age Prediction 2 1.3 The Scope of this Study 4 Chapter 2 Methods 5 2.1 Data Source and Study Population 5 2.2 Data cleaning 5 2.3 Image Preprocessing 6 2.4 Deep Learning Model for Age Prediction 7 2.4.1 Overview 7 2.4.2 Convolutional Neural Network (CNN) 8 2.4.3 Model Architecture 10 2.4.4 Model Performance 11 2.4.5 Grad-CAM 12 Chapter 3 Results 14 3.1 Descriptive Statistics 14 3.2 Age Prediction Model Performance 14 3.2.1 Model Performance 14 3.2.2 Grad-CAM 16 3.3 Subgroup Analyses 17 3.3.1 Subgroup Analyses stratified by Myopia 17 3.3.2 Subgroup Analyses stratified by Sex 18 Chapter 4 Discussion 21 4.1 Strengths and Novelties 21 4.2 Limitations 22 REFERENCES 24 FIGURES 30 TABLES 53 | - |
dc.language.iso | en | - |
dc.title | 膜齡兩可:使用深度學習方法以眼底影像建立臺灣年齡預測模型 | zh_TW |
dc.title | Age Prediction in Taiwan: A Deep Learning Approach Using Fundus Images | en |
dc.type | Thesis | - |
dc.date.schoolyear | 112-1 | - |
dc.description.degree | 碩士 | - |
dc.contributor.oralexamcommittee | 王彥雯;陳翔瀚;林昭文 | zh_TW |
dc.contributor.oralexamcommittee | Charlotte Wang;Hsiang-Han Chen;Chao-Wen Lin | en |
dc.subject.keyword | 深度學習,年齡預測,生理年齡,視網膜影像,眼底鏡影像, | zh_TW |
dc.subject.keyword | Deep learning,Age prediction,Biological age,Retinal fundus images, | en |
dc.relation.page | 55 | - |
dc.identifier.doi | 10.6342/NTU202400348 | - |
dc.rights.note | 未授權 | - |
dc.date.accepted | 2024-01-30 | - |
dc.contributor.author-college | 公共衛生學院 | - |
dc.contributor.author-dept | 流行病學與預防醫學研究所 | - |
顯示於系所單位: | 流行病學與預防醫學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-112-1.pdf 目前未授權公開取用 | 13.1 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。