跳到主要內容

臺灣博碩士論文加值系統

(3.236.23.193) 您好!臺灣時間:2021/07/26 07:13
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:劉致良
研究生(外文):LIU,ZHI-LIANG
論文名稱:機器學習應用於肺結核處方合理性之檢核與建議
論文名稱(外文):Abnormal Anti-tuberculosis Drugs Detection Using Machine Learning
指導教授:李麗惠李麗惠引用關係
指導教授(外文):LI,LI-HUI
口試委員:楊祥麟索任
口試委員(外文):YANG,XIANG-LINSUO,REN
口試日期:2020-07-29
學位類別:碩士
校院名稱:國立臺北護理健康大學
系所名稱:健康事業管理研究所
學門:商業及管理學門
學類:醫管學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:80
中文關鍵詞:臨床指引機器學習結核病處方資深醫師極限梯度提升樹
外文關鍵詞:clinical guidelinemachine learningtuberculosisprescriptionsenior doctorExtreme Gradient Boosting
相關次數:
  • 被引用被引用:0
  • 點閱點閱:38
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
背景:結核病(tuberculosis,TB)是高度傳染性的疾病,世界衛生組織(World Health Organization,WHO)於1997年發行臨床指引供醫師於治療TB病人時參考,然而臨床指引無法針對所有不同病情的TB病人給予治療建議,且對於治療TB病人經驗相較不足的醫師(資淺醫師)來說,較難評估病人各面向的治療風險,通常需要請教資深醫師,但隨著TB的個案數逐年下降,醫師診治TB病人與開立TB處方的機會逐漸減,於此同時,資深醫師也逐漸退休,診治TB病人的資淺醫師變多而可請教的對象變少,但2018年全球仍約有1,000萬名新發生個案,同年臺灣亦仍有9,127名新發生個案。
目的: 針對全國性資料,應用機器學習於建立不合理TB處方檢核模型,輔助資淺醫師快速針對各類型TB病人開立合理處方,並針對不合理處方給予提醒以及適當的處方開立建議,避免開立不合理處方。
方法: 使用結核病全國資料庫,取得2017年1月至2019年9月期間之TB病人通報資料,一共有561,819筆紀錄資料(即39,160位TB病人),包含去識別化之病人基本資料、檢驗結果、處方與劑量、機構、醫師及是否為不合理處方共59個特徵,本研究的合理處方定義是研究期間之開藥量前百大醫師或者開藥量前80%醫師所開立之處方,前者為本研究之模型一,後者為本研究之模型二,意即治療經驗相較多之醫師所開立之「專家處方」視為合理處方,兩個模型都利用隨機森林(Random Forest,RF)、極限梯度提升樹(eXtreme Gradient Boosting,XGBoost)、類神經網路(Artificial Neural Network,ANN)、及支持向量機(Support Vector Machine,SVM)四種常見且適用的演算法進行建模,使用80%之記錄作為訓練集進行模型訓練,20%之記錄作為測試集進行模型測試,測試指標包含準確率、精確率、敏感度、特異度、F1-Score、ROC曲線、及PR曲線,綜合評估指標值以找出最佳模型。進一步針對樹狀模型(RF及XGBoost)之基本資料與檢驗結果特徵進行重要性分數分析。以及隨機取得至多五萬筆之原始紀錄,分別使用本研究之最佳模型及逐筆比對相對應的專家處方以檢核處方是否不合理,並比較運算效能。最後針對模型預測為不合理的處方提出建議。
結果:兩個模型的最佳模型都是使用XGBoost演算法,但模型一比模型二的表現佳,其準確率為0.988、精確率為0.988、敏感度為0.987、特異度為0.988、F1-Score為0.988、ROC曲線為0.998、PR曲線0.997。影響模型預測能力之重要性分數較大的特徵為年齡(RF:0.534;XGBoost:0.017)及有肺外結核(RF:0.038; XGBoost:0.447)。運算效能的部分,若輸入五萬筆紀錄進行不合理處方檢核,僅需費時約5秒,但若逐條比對專家處方則需花約6分鐘又23秒。針對不合理的處方,本研究亦提供相同病人特質下之專家處方供參及TB診治指引提供之共病不建議開立處方作為學習與提醒。
結論:利用機器學習能夠快速將全國台灣治療TB的資深醫師之處方紀錄建立模型,不僅可用以檢核任一醫師所開立之處方是否與專家處方不一致,若不一致則給予專家處方開立方式及共病不適合用藥資訊供參,亦可作為TB醫學教育使用以傳承專業經驗。未來可持續匯入最新TB處方紀錄及更多特徵以強化模型檢核不合理處方之能力。
關鍵字: 臨床指引、機器學習、結核病、處方、資深醫師、極限梯度提升樹

Background: Tuberculosis is a highly infectious disease. The World Health Organization published clinical guidelines in 1997 for physicians in treating patients with TB. However, clinical guidelines could not provide treatment advice for all TB patients with different conditions. Physicians with relatively inexperienced in treating TB patients, i.e., junior physicians, have difficulties in assessing the treatment risks of patients in various aspects. They were usually needed to consult a senior doctor. However, as the number of TB cases decreased yearly, the chances of physicians to treat and prescribe for TB patients were gradually decreasing. At the same time, senior physicians were gradually retiring. That means the number of junior physicians is increased, and the number of senior physicians who can be consulted is decreased. However, there were still about ten million new cases in the world in 2018, and there were still 9,127 new cases in Taiwan in the same year.
Objective: For assisting junior physicians to provide prescriptions typically for TB patients, this study uses a national dataset and apply machine learning to establish an abnormal anti-tuberculosis drugs detection model. Also, to avoid abnormal drugs prescribed, we design response messages with advice and the prescriptions provided by senior physicians for the abnormal drug.
Methods: Using a national TB patient reporting database that contained records from January 2017 to September 2019. A total of 544,925 records (39,160 TB patients), including 59 features of de-identified patient information, examination results, prescription and dosage, institution, physician, and whether it is an abnormal drug. The study defines typically prescription as prescriptions prescribed by the top 100 physicians or the top 80% of physicians who prescribed the drug during the study period. The former is the first model of the study, and the latter is the second model of the study. This means that "expert prescriptions" are deemed as typical prescriptions. Both models use Random Forest (RF), Extreme Gradient Boosting (XGBoost), Artificial Neural Network (ANN), and Support Vector Machine (SVM) four applicable and commonly used algorithms for modeling. We use 80% of the records as a training set for model training, and 20% of the records as a testing set for model testing. The validation indices include accuracy, precision, sensitivity, specificity, F1-Score, ROC curve, and PR curve, and comprehensively evaluate values to find the best model. Further, analyze the importance scores based on the patient information and examination results to build tree-based models (RF and XGBoost). Moreover, randomly retrieve at most 50,000 records to check whether the prescriptions are abnormal drugs by using the best model of the study and the corresponding expert prescriptions and further to compare their computational efficiency. Finally, giving advice when the model predicts abnormal drugs.
Results: The best models of the two models are with the XGBoost algorithm, but model one performs better than model two. Model one’s accuracy is 0.988, precision is 0.988, sensitivity is 0.987, specificity is 0.989, and F1-Score is 0.988, ROC curve is 0.998, and PR curve is 0.997. The features with higher importance scores that affect the prediction ability of the model are age (RF: 0.534; XGBoost: 0.017) and extrapulmonary tuberculosis (RF: 0.038; XGBoost: 0.447). For computing performance, if 50,000 records are entered for checking abnormal drugs, it only takes about 5 seconds, but expert prescriptions take about 6 minutes and 23 seconds. This study also provides expert prescriptions under the same patient characteristics for reference, and the comorbidities are not advised to prescribe as learning and reminders in the TB diagnosis and treatment guidelines.
Conclusion: Using machine learning can quickly establish a model based on the prescription records of senior physicians in Taiwan. It is not only used to check whether the prescription prescribed by any physician is abnormal drugs and give appropriate TB prescription advice but also can be used as TB medical education to pass on professional experience. In the future, the model should be continued to import new prescription records and more features to strengthen the model's ability to check abnormal drugs.
Keywords: clinical guideline, machine learning, tuberculosis, prescription, senior doctor; Extreme Gradient Boosting

表目錄 viii
圖目錄 ix

第一章 緒論 1
第一節 研究背景 1
第二節 研究動機 2
第三節 研究目的 5
第二章 文獻探討 6
第一節 臨床指引的缺點與機器學習帶給資淺醫師的優勢 6
第二節 機器學習於處方及其他疾病之依變項定義應用 10
第三節 機器學習預測不合理處方之方法 15
3-1 隨機森林(Random Forest)15
3-2 極限梯度提升樹(eXtreme Gradient Boosting,XGBoost)16
3-3 類神經網路(Artificial Neural Network,ANN)17
3-4 支持向量機(Support Vector Machine,SVM)22
第三章 研究方法 24
第一節 資料來源與資料前處理 24
第二節 研究對象 28
第三節 研究架構 30
第四節 操作型定義 32
第五節 TB個案之處方紀錄描述性統計 34
第六節 模型參數設定選擇 38
第七節 分析工具及模型評估指標 39
第八節 處方不合理建議方式 43
第四章 研究結果 44
第一節 模型特徵選擇 44
第二節 模型評估結果 46
第三節 影響模型正確率之特徵重要性評估 50
第四節 以模型與專家處方進行TB處方不合理檢核之運算效能比較 51
第五節 處方不合理建議結果 52
第五章 討論與結論 53
第一節 研究結果討論 53
5-1模型評估結果 53
5-2以模型與專家處方進行TB處方不合理檢核之運算效能比較結果 54
5-3處方不合理性建議方法 55
5-4影響模型正確性結果探討 55
第二節 研究結論與建議 56
第三節 研究限制與未來展望 57
附件一 模型特徵調參結果 59
參考文獻 62


參考文獻
英文文獻
Agrawal, A., Agarwal, S. K., Kaleekal, T., & Gupta, Y. K. (2016). Rifampicin and anti-hypertensive drugs in chronic kidney disease: Pharmacokinetic interactions and their clinical impact. Indian journal of nephrology, 26(5), 322.
Ahuja, A. S. (2019). The impact of artificial intelligence in medicine on the future role of the physician. PeerJ, 7, e7702.
Anakal, S., & Sandhya, P. (2017). Clinical decision support system for chronic obstructive pulmonary disease using machine learning techniques. 2017 International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT), 1-5.
Avenir, A. J. E., Villota, J. D., & Rosas, M. F. (2015). Machine learning: A means of diagnosing and prescribing treatments for tuberculosis patients. International Journal of Computer Theory and Engineering, 7(3), 201.
Azlah, M. A. F., Chua, L. S., Rahmad, F. R., Abdullah, F. I., & Wan Alwi, S. R. (2019). Review on Techniques for Plant Leaf Classification and Recognition. Computers, 8(4), 77.
Basheer, I. A., & Hajmeer, M. (2000). Artificial neural networks: Fundamentals, computing, design, and application. Journal of microbiological methods, 43(1), 3-31.
Bloss, E., Chan, P. C., Cheng, N. W., Wang, K. F., Yang, S. L., & Cegielski, P. (2012). Increasing directly observed therapy related to improved tuberculosis treatment outcomes in Taiwan. The International journal of tuberculosis and lung disease, 16(4), 462-467.
Boudoulas, K. D., Leier, C. V., Geleris, P., & Boudoulas, H. (2015). The shortcomings of clinical practice guidelines. Cardiology, 130(3), 187-200.
Busse, J. W., Juurlink, D., & Guyatt, G. H. (2016). Addressing the limitations of the CDC guideline for prescribing opioids for chronic noncancer pain. CMAJ, 188(17-18), 1210-1211.
Cabitza, F., Rasoini, R., & Gensini, G. F. (2017). Unintended consequences of machine learning in medicine. Jama, 318(6), 517-518.
Chang, K. C., & Yew, W. W. (2013). ManAgement of difficult multidrug‐resistant tuberculosis and extensively drug‐resistant tuberculosis: update 2012. Respirology, 18(1), 8-21.
Chang, W., Liu, Y., Xiao, Y., Yuan, X., Xu, X., Zhang, S., & Zhou, S. (2019). A
Machine-Learning-Based Prediction Method for Hypertension Outcomes Based on Medical Data. Diagnostics, 9(4), 178.
Chen, C. C. F., Tsai, W. C., & Chou, W. T. (2004). The Application of Evidence Base Medicine on Health Care. Journal of the Formosan Medical Association, 8(2), 235-240. doi:10.6320/FJM.2004.8(2).12
Chen, C. (2010). Applying a Guideline-based Decision Support System for Pharmacological Treatment to Improve the Management of Hyperlipidemia. Retrieved from https://hdl.handle.net/11296/53y435
Chen, C. C., Chiang, C. Y., Pan, S. C., Wang, J. Y., & Lin, H. H. (2015). Health system delay among patients with tuberculosis in Taiwan: 2003–2010. BMC infectious diseases, 15(1), 491.
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree Boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining ,785-794.
Chung-Delgado, K., Revilla-Montag, A., Guillen-Bravo, S., Velez-Segovia, E., Soria-Montoya, A., Nuñez-Garbin, A., ... & Bernabe-Ortiz, A. (2011). Factors associated with anti-tuberculosis medication adverse effects: A case-control study in Lima, Peru. PloS one, 6(11) , e27610.
Deng, Y., Sun, Y., Zhu, Y., Xu, Y., Yang, Q., Zhang, S., ... & Yuan, K. (2019). A new framework to reduce doctor's workload for medical imAge Annotation. IEEE Access, 7, 107097-107104.
Dessì, D., Recupero, D. R., Fenu, G., & Consoli, S. (2019). A recommender system of medical reports leveraging cognitive computing and frame semantics. Machine Learning Paradigms, 7-30.
Dooley, K. E., & Chaisson, R. E. (2009). Tuberculosis and diabetes mellitus: convergence of two epidemics. The Lancet infectious diseases, 9(12), 737-746.
Fang Kuangnan, & Jianbin, W. (2011). A review of random forest method research. Statistics and Information Forum, 26(3), 32-38.
Feng, J. Y., Huang, S. F., Ting, W. Y., Chen, Y. C., Lin, Y. Y., Huang, R. M., ... & Yu, K. W. (2012). Gender differences in treatment outcomes of tuberculosis patients in Taiwan:A prospective observational study. Clinical Microbiology and Infection, 18(9), 331-337.
Geleris, P., & Boudoulas, H. (2011). Problems related to the application of guidelines in clinical practice:A critical analysis. Hellenic J Cardiol, 52, 97-102.
Golas, S. B., Shibahara, T., Agboola, S., Otaki, H., Sato, J., Nakae, T., ... & Kvedar, J. (2018). A machine learning model to predict the risk of 30-day readmissions in patients with heart failure: A retrospective analysis of electronic medical records data. BMC medical informatics and decision making, 18(1), 44.
Horne, D. J., Royce, S. E., Gooze, L., Narita, M., Hopewell, P. C., Nahid, P., & Steingart, K. R. (2010). Sputum monitoring during tuberculosis treatment for predicting outcome: Systematic review and meta-analysis. The Lancet infectious diseases, 10(6), 387-394.
Hsieh, M. J., Huang, Y. W., Tsai, Y. H., Feng, J. Y., & Su, W. J. (2020). Improved appropriateness of anti-tuberculosis prescription by the expert prescription review program in Taiwan. Journal of the Formosan Medical Association.
Huang, Z.-S. (2018). Machine learning: Bagging, Boosting and AdaBoost of Ensemble learning. Retrieved from https://medium.com/@chih.sheng.huang821/%E6%A9%9F%E5%99%A8%E5%AD%B8%E7%BF%92-ensemble-learning%E4%B9%8Bbagging-boosting%E5%92%8Cadaboost-af031229ebc3
Hujun, Y., David, C., Paulo, N.,& Antonio J.(2018). Machine learning for drugs prescription. International Conference on Intelligent Data Engineering and Automated Learning ,548-555.
Junli, C., & Licheng, J. (2000). Classification mechanism of support vector machines. In WCC 2000-ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000 (Vol. 3, pp. 1556-1559).
Kalhori, S. R. N., & Zeng, X. J. (2013). Evaluation and comparison of different machine learning methods to predict outcome of tuberculosis treatment course. Journal of Intelligent Learning Systems and Applications,184-193
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T. Y. (2017). Lightgbm: A highly efficient gradient Boosting decision tree. Advances in neural information processing systems, 3146-3154.
Kelly, C. J., Karthikesalingam, A., Suleyman, M., Corrado, G., & King, D. (2019). Key challenges for delivering clinical impact with artificial intelligence. BMC medicine, 17(1), 195.
Khan, M. T., Kaushik, A. C., Malik, S. I., Ali, S., & Wei, D. (2019). Artificial neural networks for prediction of tuberculosis disease. Frontiers in microbiology, 10, 395.
Kikvidze, M., & Ikiashvili, L. (2014). Comorbidities and MDR-TB treatment outcomes in Georgia-2009-11 cohort. European Respiratory Journal, 44(58), 1444.
Kong, X., Ai, B., Kong, Y., Su, L., Ning, Y., Howard, N., ... & Wang, J. (2019). Artificial intelligence: A key to relieve China's insufficient and unequally-distributed medical resources. American Journal of Translational Research, 11(5), 2632.
Krittanawong, C. (2018). The rise of artificial intelligence and the uncertain future for physicians. European journal of internal medicine, 48, 13-14.
Laing, R. O., & McGoldrick, K. M. (2000). Tuberculosis drug issues: prices, fixed-dose combination products and second-line drugs. The International Journal of Tuberculosis and Lung Disease, 4(12), 194-207.
Lao, Y. R., Gu, Q. Z., Liang, Z. H., Tan, D. Y., Fan, Y. P., Li, S. C., ... & Li, Y. (2011, November). A data mining research method based on the concept of evidence based TCM inheritance in famous veteran TCM doctors' personal medical records. In 2011 IEEE International Conference on Bioinformatics and Biomedicine Workshops (BIBMW) ,746-748.
Li, C. Y., Chen, H. C., Cheng, H. Y., Chian, C. F., Chang, F. Y., Chen, H. I., ... & Lin, J. C. (2015). Role of QuantiFERON-TB-Gold in tube assay for active and latent tuberculosis infection in investigation of tuberculosis outbreak in a university. Journal of Microbiology, Immunology and Infection, 48(3), 263-268.
Li, S., & Zhang, X. (2019). Research on orthopedic auxiliary classification and prediction model based on XGBoost algorithm. Neural Computing and Applications, 1-9.
Li, Z.-Y., Yang, X.-L., Wu, W.-Z., & Liu, D.-P. (2014). Taiwan tuberculosis treatment policy evolution and implementation monitoring mechanism. TAIWAN EPIDEMIOLOGY BULLETIN, 30(5), 92-100. doi:10.6524/EB.201403_30(5).0002
Liang, Z., Zhang, G., Li, Z., Yin, J., & Fu, W. (2012, October). Deep learning for acupuncture point selection patterns based on veteran doctor experience of chinese medicine. In 2012 IEEE International Conference on Bioinformatics and Biomedicine Workshops,396-401.
Lin, Y. C., Huang, W. T., Ou, S. C., Hung, H. H., Cheng, W. Z., Lin, S. S., ... & Huang, S. T. (2019). Neural network analysis of Chinese herbal medicine prescriptions for patients with colorectal cancer. Complementary therapies in medicine, 42, 279-285.
Liu, Y.-L. (2009). A Decision Support System for Screening Prescription Errors in Treatment for Pulmonary Tuberculosis. Retrieved from https://hdl.handle.net/11296/f6m2wf
Liu, C., Liu, X., Wu, F., Xie, M., Feng, Y., & Hu, C. (2018). Using artificial intelligence (Watson for Oncology) for treatment recommendations amongst Chinese patients with lung cancer: feasibility study. Journal of medical Internet research, 20(9), e11087.
Lo, H. Y., Suo, J., Chang, H. J., Yang, S. L., & Chou, P. (2015). Risk factors associated with death in a 12-month cohort analysis of tuberculosis patients: 12-month follow-up after registration. Asia Pacific Journal of Public Health, 27(2), NP758-NP768.
Lugtenberg, M., Burgers, J. S., Clancy, C., Westert, G. P., & Schneider, E. C. (2011). Current guidelines have limited applicability to patients with comorbid conditions: a systematic analysis of evidence-based guidelines. PloS one, 6(10), e25987.
Ma, B., Meng, F., Yan, G., Yan, H., Chai, B., & Song, F. (2020). Diagnostic
classification of cancers using extreme gradient boosting algorithm and multi-omics data. Computers in Biology and Medicine, 103761.
Ma, L., & Fan, S. (2017). CURE-SMOTE algorithm and hybrid algorithm for feature selection and parameter optimization based on random forests. BMC bioinformatics, 18(1), 169.
Magoulas, G. D., & Prentza, A. (1999). Machine learning in medical applications. Advanced course on artificial intelligence, 300-307.
Magrabi, F., Ammenwerth, E., McNair, J. B., De Keizer, N. F., Hyppönen, H., Nykänen, P., ... & Georgiou, A. (2019). Artificial Intelligence in clinical decision support: Challenges for evaluating AI and practical implications. Yearbook of medical informatics, 28(01), 128-134.
Mahmoodi, D., Marvi, H., Taghizadeh, M., Soleimani, A., Razzazi, F., & Mahmoodi, M. (2011, July). Age estimation based on speech features and support vector machine. In 2011 3rd Computer Science and Electronic Engineering Conference (CEEC) (pp. 60-64). IEEE.
Petersen, E., Rao, M., Ippolito, G., Gualano, G., Chakaya, J., Ntoumi, F., ... & Hergens, M. P. (2019). World Tuberculosis Day March 24th 2019 Theme: "It's TIME"—International Journal of Infectious Diseases Tuberculosis Theme Series. International Journal of Infectious Diseases, 80, 1-5.
Rough, K., Dai, A. M., Zhang, K., Xue, Y., Vardoulakis, L. M., Cui, C., ... & Rajkomar, A. (2020). Predicting inpatient medication orders from electronic health record data. Clinical Pharmacology & Therapeutics.
Rozenblum, R., Rodriguez-Monguio, R., Volk, L. A., Forsythe, K. J., Myers, S., McGurrin, M., ... & Seoane-Vazquez, E. (2020). Using a machine learning system to identify and prevent medication prescribing errors: A clinical and cost analysis evaluation. The Joint Commission Journal on Quality and Patient Safety, 46(1), 3-10.
Ruan, S.-Y. (2015). Clinical manifestations and diagnosis of tuberculous pleural effusion. Antituberculosis Magazine. Retrieved from https://www.tb.org.tw/publication_D.php?id=80
Rubin, H. R., Pronovost, P., & Diette, G. B. (2001). The advantAges and disadvantAges of process‐based measures of health care quality. International Journal for Quality in Health Care, 13(6), 469-474.
Shekelle, P. G., Woolf, S. H., Eccles, M., & Grimshaw, J. (1999). Developing guidelines. Bmj, 318(7183), 593-596.
Shuhua, H., Jianbang, X., Guifeng, W., Xianglin, Y., Yanfang, H., & Changxun, C. (2014). My country's tuberculosis treatment policy evolution and implementation monitoring mechanism. TAIWAN EPIDEMIOLOGY BULLETIN, 30(5), 92-100. doi:10.6524/EB.201403_30(5).0002
Singh, N., Singh, P., & Bhagat, D. (2019). A rule extraction approach from support
vector machines for diagnosing hypertension among diabetics. Expert Systems with Applications, 130, 188-205.
Sklearn.(2019).RandomForest. Retrieved from
https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html
Somashekhar, S. P., Sepúlveda, M. J., Puglielli, S., Norden, A. D., Shortliffe, E. H., Rohit Kumar, C., ... & Ramya, Y. (2018). Watson for Oncology and breast cancer treatment recommendations: agreement with an expert multidisciplinary tumor board. Annals of Oncology, 29(2), 418-423.
Sotgiu, G., Centis, R., D'ambrosio, L., & Migliori, G. B. (2015). Tuberculosis treatment and drug regimens. Cold Spring Harbor perspectives in medicine, 5(5), a017822.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1), 1929-1958.
Taiwan Centers for Disease Control. (2017). Guidelines for Diagnosis and Treatment of Tuberculosis (Sixth Edition): Taiwan Centers for Disease Control.
Taiwan Centers for Disease Control. (2019). The epidemic trend of tuberculosis in Taiwan and current prevention and treatment policies. Retrieved from https://wwwap.cdc.gov.tw/Uploads/55244fff-ec3e-433d-b507-409bf6ddb1dd.pdf
Ting, S. L., Kwok, S. K., Tsang, A. H., & Lee, W. B. (2011). A hybrid knowledge-based approach to supporting the medical prescription for general practitioners: Real case in a Hong Kong medical center. Knowledge-Based Systems, 24(3), 444-456.
Toussi, M., Lamy, J. B., Le Toumelin, P., & Venot, A. (2009). Using data mining techniques to explore physicians' therapeutic decisions when clinical guidelines do not provide recommendations: methods and example for type 2 diabetes. BMC medical informatics and decision making, 9(1), 28.
Tian, Y., Liu, X., Wang, Z., Cao, S., Liu, Z., Ji, Q., ... & Zhou, Y. (2020). Concordance Between Watson for Oncology and a Multidisciplinary Clinical Decision-Making Team for Gastric Cancer and the Prognostic Implications: Retrospective study. Journal of Medical Internet Research, 22(2), e14122.
Verikas, A., Gelzinis, A., & Bacauskiene, M. (2011). Mining data with random forests: A survey and results of new tests. Pattern recognition, 44(2), 330-349.
Wang, Z.-Y. (2017). Do elderly people need latent tuberculosis infection? Antituberculosis Magazine. Retrieved from https://www.tb.org.tw/publication_D.php?id=117
Wang, M. C., & Cervantes, J. (2019). Glycemic control in tuberculosis: Lessons learned from Taiwan. Asian Pacific Journal of Tropical Medicine, 12(10), 438.
Wang, Y., Li, P.-F., Tian, Y., Ren, J.-J., & Li, J.-S. (2016). A shared decision-making system for diabetes medication choice utilizing electronic health record data. 21(5), 1280-1287.
World Health Organization. (2018). Tuberculosis. Retrieved from
https://www.who.int/health-topics/tuberculosis#tab=tab_1
Yao, L., Cai, M., Chen, Y., Shen, C., Shi, L., & Guo, Y. (2019). Prediction of antiepileptic drug treatment outcomes of patients with newly diagnosed epilepsy by machine learning. Epilepsy & Behavior, 96, 92-97.
Ye, Y., Wu, Q., Huang, J. Z., Ng, M. K., & Li, X. (2013). Stratified sampling for feature subspace selection in random forests for high dimensional data. Pattern Recognition, 46(3), 769-787.
Zhang, Y.-J., & Chen, D.-J. (2018). Rifampin applied to the drug interaction between cancer patients and HIV-HCV co-infection patients. Tuberculosis Bulletin. Retrieved from http://www.tstld.org/information_list.php?page=3


電子全文 電子全文(網際網路公開日期:20250827)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top