隨著人口老化，阿茲海默症成為最常見的失智症類型，嚴重影響患者生活品質。
本研究旨在探討結合腦室體積與認知評估量表的診斷方法，以期在阿茲海默症的診斷
中取得優異表現。首先對3D磁振造影（MRI）影像進行資料預處理，計算每張切片中
的腦室面積，最終透過切片堆疊的方式加總成腦室體積。接著，將腦室體積及認知評
估量表數據輸入深度學習中的全連接層神經網絡，區分認知正常對照組（CN）、輕度認
知障礙患者（MCI）和阿茲海默症患者（AD）。結果顯示，三元分類準確率為0.87，認
知正常與輕度認知障礙分類準確率為0.95，輕度認知障礙與阿茲海默症分類準確率為
0.84，認知正常與阿茲海默症分類準確率為0.99。研究結果表明，腦室體積擴大與認知
功能下降具有相關性，為阿茲海默症的早期診斷和進展監測提供了新的思路。然而，
本研究在區分輕度認知障礙與阿茲海默症方面仍有進步空間，未來可增加與阿茲海默
症高度相關的特徵，以改進診斷方法的準確性。總結來說，本研究提出的綜合診斷方
法在提高阿茲海默症診斷準確性方面具有潛力，為未來相關研究和臨床應用提供了重
要參考。

With the aging population, Alzheimer’s disease has become the most common type of de
mentia, severely affecting the quality of life of patients. This study aims to explore a diagnostic
method that combines ventricular volume and cognitive assessment scales to achieve superior
performance in diagnosing Alzheimer’s disease. First, we preprocessed the data from 3D mag
netic resonance imaging (MRI) to calculate the ventricular area in each slice, and then summed
these areas through slice stacking to obtain the total ventricular volume. Next, we input the
ventricular volume and cognitive assessment scale data into a fully connected layer neural net
work in deep learning to distinguish between cognitively normal controls (CN), mild cognitive
impairment patients (MCI), and Alzheimer’s disease patients (AD).
The results showed that the accuracy of the ternary classification is 0.87, the accuracy of
distinguishing between cognitively normal and mild cognitive impairment is 0.95, the accu
racy of distinguishing between mild cognitive impairment and Alzheimer’s disease is 0.84, and
the accuracy of distinguishing between cognitively normal and Alzheimer’s disease is 0.99. The
results indicate that there is a correlation between increased ventricular volume and cognitive de
cline, providing new insights for the early diagnosis and progression monitoring of Alzheimer’s
disease.
However, there is still room for improvement in distinguishing between mild cognitive im
pairment and Alzheimer’s disease. Future work could focus on incorporating features highly
correlated with Alzheimer’s disease to improve the accuracy of the diagnostic method. In con
clusion, the comprehensive diagnostic method proposed in this study has potential in improving
the accuracy of Alzheimer’s disease diagnosis, providing an important reference for future re
search and clinical applications.

[1] J. Gaugler, B. James, T. Johnson, et al., “2022 alzheimer’s disease facts and figures,” Alzheimers
&Dementia, vol. 18, no. 4, pp. 700–789, 2022.
[2] R.C.Petersen, G. E. Smith, S. C. Waring, R. J. Ivnik, E. G. Tangalos, and E. Kokmen, “Mild cog
nitive impairment: Clinical characterization and outcome,” Archives of neurology, vol. 56, no. 3,
pp. 303–308, 1999.
[3] D.R. Roalf, P. J. Moberg, S. X. Xie, D. A. Wolk, S. T. Moelter, and S. E. Arnold, “Comparative
accuracies of two common screening instruments for classification of alzheimer’s disease, mild
cognitive impairment, and healthy aging,” Alzheimer’s & dementia, vol. 9, no. 5, pp. 529–537,
2013.
[4] S. L. Warren and A. A. Moustafa, “Functional magnetic resonance imaging, deep learning, and
alzheimer’s disease: A systematic review,” Journal of Neuroimaging, vol. 33, no. 1, pp. 5–18,
2023.
[5] C.Jiang,X.Zhang,W.Huang,andC.Meinel,“Segmentationandquantificationofbraintumor,”in
2004 IEEE Symposium on Virtual Environments, Human-Computer Interfaces and Measurement
Systems, 2004.(VCIMS)., IEEE, 2004, pp. 61–66.
[6] J. Wang, A. Ekin, and G. de Haan, “Shape analysis of brain ventricles for improved classification
of alzheimer＇spatients,” in 2008 15th IEEE International Conference on Image Processing, IEEE,
2008, pp. 2252–2255.
[7] S. M. Nestor, R. Rupsingh, M. Borrie, et al., “Ventricular enlargement as a possible measure of
alzheimer’s disease progression validated using the alzheimer’s disease neuroimaging initiative
database,” Brain, vol. 131, no. 9, pp. 2443–2454, 2008.
[8] O. Colliot, G. Chételat, M. Chupin, et al., “Discrimination between alzheimer disease, mild cog
nitive impairment, and normal aging by using automated segmentation of the hippocampus,” Ra
diology, vol. 248, no. 1, pp. 194–201, 2008.
[9] “Linear/fully-connected layers user’s guide.” (), [Online]. Available: https://docs.nvidia.
com/deeplearning/performance/dl-performance-fully-connected/index.html.
[10] S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing
internal covariate shift,” in International conference on machine learning, pmlr, 2015, pp. 448
456.
[11] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A sim
ple way to prevent neural networks from overfitting,” The journal of machine learning research,
vol. 15, no. 1, pp. 1929–1958, 2014.
[12] D.P.KingmaandJ.Ba,“Adam:Amethodforstochasticoptimization,”arXivpreprintarXiv:1412.6980,
2014.
[13] “Confusion matrix.” (), [Online]. Available: https://www.ml-science.com/confusion
matrix.
[14] “Alzheimer＇s disease neuroimaging initiative.” (), [Online]. Available: https://adni.loni.
usc.edu/.
[15] M.F.Folstein, S. E. Folstein, and P. R. McHugh, ““mini-mental state＂: A practical method for
grading the cognitive state of patients for the clinician,” Journal of psychiatric research, vol. 12,
no. 3, pp. 189–198, 1975.
[16] J. A. Yesavage, T. L. Brink, T. L. Rose, et al., “Development and validation of a geriatric de
pression screening scale: A preliminary report,” Journal of psychiatric research, vol. 17, no. 1,
pp. 37–49, 1982.
[17] J. C. Morris, “The clinical dementia rating (cdr) current version and scoring rules,” Neurology,
vol. 43, no. 11, pp. 2412–2412, 1993.
[18] R.I.Pfeffer, T. T. Kurosaki, C. Harrah Jr, J. M. Chance, and S. Filos, “Measurement of functional
activities in older adults in the community,” Journal of gerontology, vol. 37, no. 3, pp. 323–329,
1982.
[19] J. L. Cummings, M. Mega, K. Gray, S. Rosenberg-Thompson, D. A. Carusi, and J. Gornbein,
“The neuropsychiatric inventory: Comprehensive assessment of psychopathology in dementia,”
Neurology, vol. 44, no. 12, pp. 2308–2308, 1994.
[20] “Welcome to antspynet＇s documentation!” (), [Online]. Available: https://antsx.github.
io/ANTsPyNet/docs/build/html/index.html.
[21] “Advanced normalization tools.” (), [Online]. Available: https://stnava.github.io/ANTs/.
[22] “大津演算法.”(),[Online].Available: https://zh.wikipedia.org/zh-tw/%E5%A4%A7%E6%
B4%A5%E7%AE%97%E6%B3%95.
[23] R.M.Haralick,S.R.Sternberg,andX.Zhuang,“Imageanalysisusingmathematicalmorphology,”
IEEE transactions on pattern analysis and machine intelligence, no. 4, pp. 532–550, 1987.