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

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 X光醫學影像的辨識與統計分析
A Statistical Approach of X-ray Medical Image Recognition
作者 陳品華
Chen, Pin-Hua
貢獻者 余清祥
陳品華
Chen, Pin-Hua
關鍵詞 影像辨識
醫學影像
統計分析
維度縮減
機器學習
Image Recognition
Medical Image
Statistical Analysis
Dimensionality Reduction
Machine Learning
日期 2025
上傳時間 1-Sep-2025 14:49:06 (UTC+8)
摘要 影像辨識是人工智慧的重要應用之一，辨識手寫郵遞區號是知名範例，透過光學掃描再以電腦判讀，可大幅節省郵務的工作時間。近年電腦科技進步使得影像辨識的準確度及效率顯著提升，因此應用層面更為廣泛，像是機場掃描人臉通關、車牌辨識都是眾所皆知的應用。本文以醫學影像的病例分類為研究目標，引進統計分析的概念分辨肺炎感染者，並標示出可能異常位置，可供醫師判讀病患狀況。由於肺炎患者之肺紋會因密度增加造成毛玻璃現象，紋理變化可藉由方向梯度直方圖（Histogram of Oriented Gradients, HOG）等方法萃取其特性，本文先提取 HOG 在 12 個方向上的梯度變化累計，再透過統計、機器學習模型進行分類，可獲得較佳的準確率。此外，本文也考量分類結果與方向數量、切割區塊的關係，因為解析度扮演重要角色。結果顯示本文使用36 個解釋變數即可達到不錯的準確率，並兼顧維度縮減及運算效率，而且挑選出之解釋變數可作為醫師臨床診斷的參考。
Image recognition is one of the key applications of artificial intelligence. A well-known example is the recognition of handwritten postal codes, where optical scanning combined with computer interpretation can significantly reduce the workload of postal services. In recent years, advancements in computer technology have greatly improved the accuracy and efficiency of image recognition, leading to wider applications such as airport facial recognition and license plate recognition. This study focuses on classifying medical images of clinical cases, introducing statistical analysis to distinguish patients with pneumonia and highlight potential abnormal regions for physicians to interpret patient conditions. Due to increased lung texture density in pneumonia patients causing ground-glass opacity, texture variations can be captured using methods such as Histogram of Oriented Gradients (HOG). In this study, HOG features in 12 directions are extracted and accumulated, followed by classification using statistical and machine learning models to achieve better accuracy. Moreover, this study examines the relationship between classification performance and the number of gradient directions as well as image partitioning, since resolution plays a crucial role. The results show that using only 36 explanatory variables can achieve excellent accuracy while balancing dimensionality reduction and computational efficiency. The selected explanatory variables may also serve as references for clinical diagnosis by physicians.
參考文獻 一、中文文獻 [1] 陳慧霜（2023），「影像分析與深偽影片的偵測」，國立政治大學統計學系學位論文。 [2] 黃政嘉（2024），「Deepfake 與 GAN真偽圖像統計分析」，國立政治大學統計學系學位論文。 [3] 張倉銓（2022），「深度學習訓練效率的模糊驗證模式—以田口優化的卷積神經網路在辨識肺癌醫學影像為例」。國立臺中科技大學資訊工程系碩士論文。 [4] 郭立杰、許嘉汀（2022），「基於人工智慧的皮膚病理影像辨識」。收錄於《南臺灣醫學雜誌》，18（1），17‑27。 二、英文文獻 [1] Akl, A. A., Hosny, K. M., Fouda, M. M., & Salah, A. (2023).A hybrid CNN and ensemble model for COVID‑19 lung infection detection on chest CT scans. PLoS One, 18(3), e0282608. [2] Alshmrani, G. M. M., Ni, Q., Jiang, R., Pervaiz, H., & Elshennawy, N. M. (2023). A deep learning architecture for multi‑class lung diseases classification using chest X‑ray (CXR) images. Alexandria Engineering Journal, 64(10), 923– 935. [3] Apostolopoulos, I. D., & Mpesiana, T. A. (2020). COVID-19: Automatic detection from X-ray images utilizing transfer learning with convolutional neural networks. Physical and Engineering Sciences in Medicine, 43(2), 635-640. [4] Bhatele, K. R., Jha, A., Tiwari, D., Bhatele, M., Sharma, S., Mithora, M. R., & Singhal, S. (2022). COVID‑19 Detection: A systematic review of machine and deep learning‑based approaches utilizing chest X‑rays and CT scans. Cognitive Computation, 16, 1889–1926. [5] Chowdhury, M. E. H., Rahman, T., Khandakar, A., Mazhar, R., Kadir, M. A.,Mahbub, Z. B., … & Reaz, M. B. I. (2020).Can AI help in screening viral and COVID-19 pneumonia?IEEE Access, 8, 132665–132676. [6] Dalal, N., & Triggs, B. (2005). Histograms of oriented gradients for human detection.In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Vol. 1, pp. 886–893). [7] Gil, D., Díaz‑Chito, K., Sánchez, C., & Hernández‑Sabaté, A. (2020). Early screening of SARS‑CoV‑2 by intelligent analysis of X‑ray images. [8] Haralick, R. M., Shanmugam, K., & Dinstein, I. (1973). Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3(6), 610–621. [9] Ozturk, T., Talo, M., Yildirim, E. A., Baloglu, U. B., Yildirim, O., & Acharya, U.R. (2020). Automated detection of COVID-19 cases using deep neural networks with X-ray images. Computers in Biology and Medicine, 121, 103792. [10] Prabhakar, S. K., & Rajagopalan, S. P. (2019).GLCM based feature extraction and medical X-RAY image classification using machine learning techniques.International Journal of Engineering and Advanced Technology (IJEAT), 8(6), 1163–1168. [11] Rajpurkar, P., Irvin, J., Zhu, K., Yang, B., Mehta, H., Duan, T., Ding, D., Bagul, A., Langlotz, C., Shpanskaya, K., Lungren, M. P., & Ng, A. Y. (2017). CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv preprint arXiv:1711.05225. [12] Raza, R., Zulfiqar, F., Khan, M. O., Arif, M., Alvi, A., Iftikhar, M. A., & Alam, T. (2023).Lung‑EffNet: Lung cancer classification using EfficientNet from CT‑scan images.Engineering Applications of Artificial Intelligence, 126, 106902. [13] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham. [14] Vats, S., Sharma, V., Singh, K., Katti, A., Mohd Ariffin, M. M., Nazir Ahmad, M., Ahmadian, A., & Salahshour, S. (2024).Incremental learning‑based cascaded model for detection and localization of tuberculosis from chest X‑ray images. Expert Systems with Applications, 238, Article 122129. [15] Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M., & Summers, R. M. (2017). ChestX-ray8: Hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 3462-3471. [16] Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M., & Summers, R. M. (2017). ChestX‑ray8: Hospital‑scale chest X‑ray database and benchmarks on weakly‑supervised classification and localization of common thorax diseases. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 3462–3471).
描述 碩士
國立政治大學
統計學系
112354010
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0112354010
資料類型 thesis
dc.contributor.advisor 余清祥zh_TW
dc.contributor.author (Authors) 陳品華zh_TW
dc.contributor.author (Authors) Chen, Pin-Huaen_US
dc.creator (作者) 陳品華zh_TW
dc.creator (作者) Chen, Pin-Huaen_US
dc.date (日期) 2025en_US
dc.date.accessioned 1-Sep-2025 14:49:06 (UTC+8)-
dc.date.available 1-Sep-2025 14:49:06 (UTC+8)-
dc.date.issued (上傳時間) 1-Sep-2025 14:49:06 (UTC+8)-
dc.identifier (Other Identifiers) G0112354010en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/159037-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 統計學系zh_TW
dc.description (描述) 112354010zh_TW
dc.description.abstract (摘要) 影像辨識是人工智慧的重要應用之一，辨識手寫郵遞區號是知名範例，透過光學掃描再以電腦判讀，可大幅節省郵務的工作時間。近年電腦科技進步使得影像辨識的準確度及效率顯著提升，因此應用層面更為廣泛，像是機場掃描人臉通關、車牌辨識都是眾所皆知的應用。本文以醫學影像的病例分類為研究目標，引進統計分析的概念分辨肺炎感染者，並標示出可能異常位置，可供醫師判讀病患狀況。由於肺炎患者之肺紋會因密度增加造成毛玻璃現象，紋理變化可藉由方向梯度直方圖（Histogram of Oriented Gradients, HOG）等方法萃取其特性，本文先提取 HOG 在 12 個方向上的梯度變化累計，再透過統計、機器學習模型進行分類，可獲得較佳的準確率。此外，本文也考量分類結果與方向數量、切割區塊的關係，因為解析度扮演重要角色。結果顯示本文使用36 個解釋變數即可達到不錯的準確率，並兼顧維度縮減及運算效率，而且挑選出之解釋變數可作為醫師臨床診斷的參考。zh_TW
dc.description.abstract (摘要) Image recognition is one of the key applications of artificial intelligence. A well-known example is the recognition of handwritten postal codes, where optical scanning combined with computer interpretation can significantly reduce the workload of postal services. In recent years, advancements in computer technology have greatly improved the accuracy and efficiency of image recognition, leading to wider applications such as airport facial recognition and license plate recognition. This study focuses on classifying medical images of clinical cases, introducing statistical analysis to distinguish patients with pneumonia and highlight potential abnormal regions for physicians to interpret patient conditions. Due to increased lung texture density in pneumonia patients causing ground-glass opacity, texture variations can be captured using methods such as Histogram of Oriented Gradients (HOG). In this study, HOG features in 12 directions are extracted and accumulated, followed by classification using statistical and machine learning models to achieve better accuracy. Moreover, this study examines the relationship between classification performance and the number of gradient directions as well as image partitioning, since resolution plays a crucial role. The results show that using only 36 explanatory variables can achieve excellent accuracy while balancing dimensionality reduction and computational efficiency. The selected explanatory variables may also serve as references for clinical diagnosis by physicians.en_US
dc.description.tableofcontents 第一章 緒論 7 第一節 研究動機 7 第二節 研究目的 9 第二章 文獻回顧與資料介紹 10 第一節 文獻回顧 10 第二節 資料介紹 12 第三章 研究方法 14 第一節 色彩模型 15 第二節 資料結構化 16 第三節 資料抽樣與平衡 20 第四節 維度縮減 22 第五節 分類模型介紹 23 第四章 醫學影像分析 27 第一節 實驗流程 27 第二節 探索性資料分析 28 第三節 最佳分割數與方向數 34 第四節 變數選取 36 第五節 位置分析 38 第五章 統計與深度學習方法比較 40 第一節、不同結構化方法比較 40 第二節、深度學習模型比較 41 第六章 結論與建議 45 第一節 結論 45 第二節 研究建議與限制 46 參考文獻 47 附錄一、探索性資料分析 50 附錄二、切割數與方向數在不同模型準確率交叉表 51zh_TW
dc.format.extent 5106491 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0112354010en_US
dc.subject (關鍵詞) 影像辨識zh_TW
dc.subject (關鍵詞) 醫學影像zh_TW
dc.subject (關鍵詞) 統計分析zh_TW
dc.subject (關鍵詞) 維度縮減zh_TW
dc.subject (關鍵詞) 機器學習zh_TW
dc.subject (關鍵詞) Image Recognitionen_US
dc.subject (關鍵詞) Medical Imageen_US
dc.subject (關鍵詞) Statistical Analysisen_US
dc.subject (關鍵詞) Dimensionality Reductionen_US
dc.subject (關鍵詞) Machine Learningen_US
dc.title (題名) X光醫學影像的辨識與統計分析zh_TW
dc.title (題名) A Statistical Approach of X-ray Medical Image Recognitionen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) 一、中文文獻 [1] 陳慧霜（2023），「影像分析與深偽影片的偵測」，國立政治大學統計學系學位論文。 [2] 黃政嘉（2024），「Deepfake 與 GAN真偽圖像統計分析」，國立政治大學統計學系學位論文。 [3] 張倉銓（2022），「深度學習訓練效率的模糊驗證模式—以田口優化的卷積神經網路在辨識肺癌醫學影像為例」。國立臺中科技大學資訊工程系碩士論文。 [4] 郭立杰、許嘉汀（2022），「基於人工智慧的皮膚病理影像辨識」。收錄於《南臺灣醫學雜誌》，18（1），17‑27。 二、英文文獻 [1] Akl, A. A., Hosny, K. M., Fouda, M. M., & Salah, A. (2023).A hybrid CNN and ensemble model for COVID‑19 lung infection detection on chest CT scans. PLoS One, 18(3), e0282608. [2] Alshmrani, G. M. M., Ni, Q., Jiang, R., Pervaiz, H., & Elshennawy, N. M. (2023). A deep learning architecture for multi‑class lung diseases classification using chest X‑ray (CXR) images. Alexandria Engineering Journal, 64(10), 923– 935. [3] Apostolopoulos, I. D., & Mpesiana, T. A. (2020). COVID-19: Automatic detection from X-ray images utilizing transfer learning with convolutional neural networks. Physical and Engineering Sciences in Medicine, 43(2), 635-640. [4] Bhatele, K. R., Jha, A., Tiwari, D., Bhatele, M., Sharma, S., Mithora, M. R., & Singhal, S. (2022). COVID‑19 Detection: A systematic review of machine and deep learning‑based approaches utilizing chest X‑rays and CT scans. Cognitive Computation, 16, 1889–1926. [5] Chowdhury, M. E. H., Rahman, T., Khandakar, A., Mazhar, R., Kadir, M. A.,Mahbub, Z. B., … & Reaz, M. B. I. (2020).Can AI help in screening viral and COVID-19 pneumonia?IEEE Access, 8, 132665–132676. [6] Dalal, N., & Triggs, B. (2005). Histograms of oriented gradients for human detection.In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Vol. 1, pp. 886–893). [7] Gil, D., Díaz‑Chito, K., Sánchez, C., & Hernández‑Sabaté, A. (2020). Early screening of SARS‑CoV‑2 by intelligent analysis of X‑ray images. [8] Haralick, R. M., Shanmugam, K., & Dinstein, I. (1973). Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, SMC-3(6), 610–621. [9] Ozturk, T., Talo, M., Yildirim, E. A., Baloglu, U. B., Yildirim, O., & Acharya, U.R. (2020). Automated detection of COVID-19 cases using deep neural networks with X-ray images. Computers in Biology and Medicine, 121, 103792. [10] Prabhakar, S. K., & Rajagopalan, S. P. (2019).GLCM based feature extraction and medical X-RAY image classification using machine learning techniques.International Journal of Engineering and Advanced Technology (IJEAT), 8(6), 1163–1168. [11] Rajpurkar, P., Irvin, J., Zhu, K., Yang, B., Mehta, H., Duan, T., Ding, D., Bagul, A., Langlotz, C., Shpanskaya, K., Lungren, M. P., & Ng, A. Y. (2017). CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv preprint arXiv:1711.05225. [12] Raza, R., Zulfiqar, F., Khan, M. O., Arif, M., Alvi, A., Iftikhar, M. A., & Alam, T. (2023).Lung‑EffNet: Lung cancer classification using EfficientNet from CT‑scan images.Engineering Applications of Artificial Intelligence, 126, 106902. [13] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham. [14] Vats, S., Sharma, V., Singh, K., Katti, A., Mohd Ariffin, M. M., Nazir Ahmad, M., Ahmadian, A., & Salahshour, S. (2024).Incremental learning‑based cascaded model for detection and localization of tuberculosis from chest X‑ray images. Expert Systems with Applications, 238, Article 122129. [15] Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M., & Summers, R. M. (2017). ChestX-ray8: Hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 3462-3471. [16] Wang, X., Peng, Y., Lu, L., Lu, Z., Bagheri, M., & Summers, R. M. (2017). ChestX‑ray8: Hospital‑scale chest X‑ray database and benchmarks on weakly‑supervised classification and localization of common thorax diseases. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 3462–3471).zh_TW