Please use this identifier to cite or link to this item: https://ah.nccu.edu.tw/handle/140.119/136832


Title: 使用機器學習方法對燒傷程度進行分類
A Machine Learning Approach for Classifying the Burn Degree
Authors: 謝承翰
Xie, Cheng-Han
Contributors: 洪英超
Hung, Ying-Chau
謝承翰
Xie, Cheng-Han
Keywords: 視頻血管造影術
機器學習
隨機森林
Indocyamime Green Video Angiography
Machine Learning
Random Forest
AdaBoost
Date: 2021
Issue Date: 2021-09-02 15:42:16 (UTC+8)
Abstract: 在如今資訊發達的時代,各項科技大幅進步,資料儲存越來越容易,不僅是成本降低,透過網路的便利性也迅速地提升了各項資料累積的速度,而藉助大量的資料,數據分析在各領域有不同的發展及應用,從交通、金融、醫療、體育到日常生活,都能依據不同的目的提供不一樣的幫助。在醫療上對於每次成功的燒傷治療來說,即早的判斷燒傷程度是非常重要的,因此燒傷程度的判定是決策過程中的關鍵步驟。本研究的目的是利用從Indocyamime Green(ICG)視頻血管造影術收集的數據,設計一種非侵入性和有效的方法來對患者的燒傷程度進行分類。研究上分兩階段,第一階段以2015年6月於台灣發生之八仙粉塵爆炸造成二度及三度燒傷的病患共21位於手術中應用Indocyamime Green(ICG)視頻血管造影術蒐集資料。第二階段採用隨機森林及AdaBoost兩種機器學習方法,根據從ICG視頻血管造影術收集的數據來對不同的燒傷程度進行分類。研究結果顯示,結合使用手術中應用ICG視頻血管造影收集的數據與所提出的機器學習方法,將對於醫師在燒傷程度的判定有極大的幫助。此項研究不但可以協助醫療資源不足時的燒傷程度判定及處理,未來也可用於開發醫療診斷的AI人工智能系統。
In today’s information-developed era, various technologies have greatly advanced, and data storage has become easier and easier. Not only is the cost reduced, the convenience of the Internet has also rapidly increased the speed of various data accumulation, and with the help of a large amount of data, data Analysis has different developments and applications in various fields, from transportation, finance, medical care, sports to daily life, and can provide different help according to different purposes. In the medical field, for each successful burn treatment, it is very important to judge the degree of burn as soon as possible. Therefore, the judgment of the degree of burn is a key step in the decision-making process. The purpose of this study is to use the data collected from Indocyamime Green(ICG) video angiography to design a non-invasive and effective method to classify the degree of burns in patients. The study is divided into two phases. In the first phase, 21 patients with second and third degree burns caused by the Baxian dust explosion in Taiwan in June 2015 were used Indocyamime Green(ICG) video angiography during surgery. In the second stage, the machine learning method Random Forest and AdaBoost are used to perform binary classification based on the data collected from ICG video angiography to improve the prediction accuracy of the degree of burn. The results of the study show that the use of data collected during surgery using ICG video angiography combined with the proposed machine learning method is of great help to physicians in judging the degree of burns. It can assist the judegement of burn degrees when the medical resource is not sufficient and can be further used in developing the artificial intelligence(AI) systems for medical diagnosis.
Reference: 1. Alander, J. T., Kaartinen, I., Laakso, A., Pätilä, T., Spillmann, T., Tuchin, V. V., ... & Välisuo, P. (2012). A review of indocyanine green fluorescent imaging in surgery. International journal of biomedical imaging, 2012.
2. Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
3. Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16, 321-357.
4. Dissanaike, S., Abdul-Hamed, S., & Griswold, J. A. (2014). Variations in burn perfusion over time as measured by portable ICG fluorescence: A case series. Burns & trauma, 2(4), 2321-3868.
5. Flower, R. W. (1973). Injection technique for indocyanine green and sodium fluorescein dye angiography of the eye. Investigative Ophthalmology & Visual Science, 12(12), 881-895.
6. Hajiran, A., Zekan, D., Trump, T., Dangerfield, D., & Luchey, A. (2019). Use of SPY Elite Fluorescence Imaging in Creation of a Continent Urinary Diversion. Case reports in urology, 2019.
7. Kamolz, L. P., Andel, H., Haslik, W., Donner, A., Winter, W., Meissl, G., & Frey, M. (2003). Indocyanine green video angiographies help to identify burns requiring operation. Burns, 29(8), 785-791.
8. Liaw, A., & Wiener, M. (2002). Classification and regression by randomForest. R news, 2(3), 18-22.
9. Loh, W. Y. (2011). Classification and regression trees. Wiley interdisciplinary reviews: data mining and knowledge discovery, 1(1), 14-23.
10. MacQueen, J. (1967, June). Some methods for classification and analysis of multivariate observations. In Proceedings of the fifth Berkeley symposium on mathematical statistics and probability (Vol. 1, No. 14, pp. 281-297).
11. McUmber, H., Dabek, R. J., Bojovic, B., & Driscoll, D. N. (2019). Burn depth analysis using indocyanine green fluorescence: a review. Journal of Burn Care & Research, 40(4), 513-516.
12. Oshiro, T. M., Perez, P. S., & Baranauskas, J. A. (2012, July). How many trees in a random forest?. In International workshop on machine learning and data mining in pattern recognition (pp. 154-168). Springer, Berlin, Heidelberg.
13. Samuel, A. L. (1959). Some studies in machine learning using the game of checkers. IBM Journal of research and development, 3(3), 210-229.
14. Schapire, R. E. (2013). Explaining adaboost. In Empirical inference (pp. 37-52). Springer, Berlin, Heidelberg.
15. Ying, C., Qi-Guang, M., Jia-Chen, L., & Lin, G. (2013). Advance and prospects of AdaBoost algorithm. Acta Automatica Sinica, 39(6), 745-758.
Description: 碩士
國立政治大學
統計學系
108354019
Source URI: http://thesis.lib.nccu.edu.tw/record/#G0108354019
Data Type: thesis
Appears in Collections:[統計學系] 學位論文

Files in This Item:

File Description SizeFormat
401901.pdf2287KbAdobe PDF0View/Open


All items in 學術集成 are protected by copyright, with all rights reserved.


社群 sharing