近年來，機器學習已經成功被應用在許多醫學領域上，像是影像處理任務、自然語言處理、心衰竭預測等。在未來，為了提升醫院品質以及病人照護，能夠導入機器學習甚至是人工智慧成為了決定性的重要關鍵。機器學習在醫學照護裡在許多方面可能會有幫忙，包含：(1)疾病診斷及(2)預後預測。
因此本研究著重在探討與使用機器學習來嘗試輔助臨床人員照護品質，分別針對急性腎臟衰竭的死亡預測與慢性腎臟病人的高血鉀預測，來提供臨床照護的另外一個選擇，希望可以增加照護品質。在急性腎臟衰竭的死亡預測上，我們主要使用兩個開放式的加護病房資料庫，來增加資料多樣性，並使用不同機器學習模型來互相比較，藉此加強模型預測的效能，也希望之後有機會使用台灣資料來做驗證。另一方面，關於慢性腎臟病人的高血鉀預測，我們使用了台灣醫院的慢性腎臟病資料庫，同時也使用機器學習的視覺化模組，希望可以找到不同於傳統高血鉀的危險因子，同時也期待可以運用在臨床照護上，讓慢性腎臟病人可以得到更好的照護。
針對急性腎臟衰竭的死亡預測與慢性腎臟病人的高血鉀預測的任務上。在不同資料集的實驗結果中，我們發現即使在不同的資料庫上，機器學習仍然可以有不錯的預測效能。而在高血鉀的預測上，機器學習可以比臨床醫師有更好的判斷成效。

In recent years, machine learning has been successfully applied in various medical fields such as image processing tasks, natural language processing, and heart failure prediction. In the future, incorporating machine learning or even artificial intelligence will be a crucial factor in improving hospital quality and patient care. Machine learning can potentially assist in two aspects of medical care: (1) disease diagnosis and (2) prognosis prediction.
Therefore, this study focuses on exploring the use of machine learning to assist healthcare professionals in improving the quality of care. Specifically, it aims to provide an alternative option for clinical care by predicting mortality in acute kidney failure and predicting high potassium levels in chronic kidney disease patients, with the goal of enhancing the quality of care. In predicting mortality in acute kidney failure, we primarily utilized two open intensive care unit databases to increase data diversity and compared different machine learning models to improve prediction performance. We also hope to have the opportunity to validate our findings using data from Taiwan. On the other hand, for predicting high potassium levels in chronic kidney disease patients, we utilized a database specific to chronic kidney disease in Taiwan. Additionally, we employed a machine learning visualization module in hopes of identifying novel risk factors distinct from traditional indicators of high potassium levels.
Regarding the tasks of predicting mortality in acute kidney failure and high potassium levels in chronic kidney disease patients, our experimental results across different datasets demonstrated that machine learning can achieve favorable prediction performance even when applied to different databases. As shown by the experimental results, machine learning performed better than physicians in hyperkalemia prediction.

Bennett, T. D., Moffitt, R. A., Hajagos, J. G., Amor, B., Anand, A., Bissell, M. M., . . . Denham, A. (2021). Clinical characterization and prediction of clinical severity of SARS-CoV-2 infection among US adults using data from the US National COVID Cohort Collaborative. JAMA network open, 4(7), e2116901-e2116901.
Brajer, N., Cozzi, B., Gao, M., Nichols, M., Revoir, M., Balu, S., . . . Hernandez, A. (2020). Prospective and external evaluation of a machine learning model to predict in-hospital mortality of adults at time of admission. JAMA network open, 3(2), e1920733-e1920733.
Breiman, L. (2001). Random forests. Machine learning, 45, 5-32.
Cha, G.-W., Moon, H.-J., & Kim, Y.-C. (2021). Comparison of random forest and gradient boosting machine models for predicting demolition waste based on small datasets and categorical variables. International Journal of Environmental Research and Public Health, 18(16), 8530.
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.
Chawla, L. S., & Kimmel, P. L. (2012). Acute kidney injury and chronic kidney disease: an integrated clinical syndrome. Kidney International, 82(5), 516-524.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Paper presented at the Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining.
Chen, T. K., Knicely, D. H., & Grams, M. E. (2019). Chronic kidney disease diagnosis and management: a review. Jama, 322(13), 1294-1304.
Chen, T. T., Chung, K. P., Lin, I. C., & Lai, M. S. (2011). The unintended consequence of diabetes mellitus pay‐for‐performance (P4P) program in Taiwan: are patients with more comorbidities or more severe conditions likely to be excluded from the P4P program? Health services research, 46(1p1), 47-60.
Chou, Y. H., Chu, T. S., & Lin, S. L. (2018). Role of renin‐angiotensin system in acute kidney injury‐chronic kidney disease transition. Nephrology, 23, 121-125.
Christodoulou, E., Ma, J., Collins, G. S., Steyerberg, E. W., Verbakel, J. Y., & Van Calster, B. (2019). A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models. Journal of clinical epidemiology, 110, 12-22.
Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. G. (2015). Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) the TRIPOD statement. Circulation, 131(2), 211-219.
Czempik, P., Cieśla, D., Knapik, P., & Krzych, Ł. (2018). Mortality of patients with acute kidney injury requiring renal replacement therapy. Advances in Clinical and Experimental Medicine, 27(3), 327-333.
da Hora Passos, R., Ramos, J. G. R., Mendonça, E. J. B., Miranda, E. A., Dutra, F. R. D., Coelho, M. F. R., . . . Macedo, E. (2017). A clinical score to predict mortality in septic acute kidney injury patients requiring continuous renal replacement therapy: the HELENICC score. BMC anesthesiology, 17, 1-8.
de Carvalho, G. M. C., Leite, T. T., & Libório, A. B. (2018). Prediction of 60-day case fatality in critically ill patients receiving renal replacement therapy: External validation of a prediction model. Shock, 50(2), 156-161.
Demirjian, S., Chertow, G. M., Zhang, J. H., O'Connor, T. Z., Vitale, J., Paganini, E. P., . . . Network, V. N. A. R. F. T. (2011). Model to predict mortality in critically ill adults with acute kidney injury. Clinical Journal of the American Society of Nephrology: CJASN, 6(9), 2114.
Dewey, J., Mastenbrook, J., & Bauler, L. D. (2020). Differentiating pseudohyperkalemia from true hyperkalemia in a patient with chronic lymphocytic leukemia and diverticulitis. Cureus, 12(8).
Einhorn, L. M., Zhan, M., Walker, L. D., Moen, M. F., Seliger, S. L., Weir, M. R., & Fink, J. C. (2009). The frequency of hyperkalemia and its significance in chronic kidney disease. Archives of internal medicine, 169(12), 1156-1162.
Emmert-Streib, F., Yang, Z., Feng, H., Tripathi, S., & Dehmer, M. (2020). An introductory review of deep learning for prediction models with big data. Frontiers in Artificial Intelligence, 3, 4.
Falcão, A. L. E., Barros, A. G. d. A., Bezerra, A. A. M., Ferreira, N. L., Logato, C. M., Silva, F. P., . . . Moreno, R. (2019). The prognostic accuracy evaluation of SAPS 3, SOFA and APACHE II scores for mortality prediction in the surgical ICU: an external validation study and decision-making analysis. Annals of intensive care, 9(1), 1-10.
Group, K. D. I. G. O. C.-M. W. (2009). KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of chronic kidney disease-mineral and bone disorder (CKD-MBD). Kidney international. Supplement, 76(113), S1-130.
Guzzi, F., Cirillo, L., Roperto, R. M., Romagnani, P., & Lazzeri, E. (2019). Molecular mechanisms of the acute kidney injury to chronic kidney disease transition: an updated view. International Journal of Molecular Sciences, 20(19), 4941.
Hoste, E. A., Bagshaw, S. M., Bellomo, R., Cely, C. M., Colman, R., Cruz, D. N., . . . Govil, D. (2015). Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. Intensive care medicine, 41, 1411-1423.
Hsieh, H.-M., Lin, M.-Y., Chiu, Y.-W., Wu, P.-H., Cheng, L.-J., Jian, F.-S., . . . Hwang, S.-J. (2017). Economic evaluation of a pre-ESRD pay-for-performance programme in advanced chronic kidney disease patients. Nephrology Dialysis Transplantation, 32(7), 1184-1194.
Hsu, C.-y., Chertow, G. M., McCulloch, C. E., Fan, D., Ordonez, J. D., & Go, A. S. (2009). Nonrecovery of kidney function and death after acute on chronic renal failure. Clinical Journal of the American Society of Nephrology: CJASN, 4(5), 891.
Hu, C.-A., Chen, C.-M., Fang, Y.-C., Liang, S.-J., Wang, H.-C., Fang, W.-F., . . . Kao, K.-C. (2020). Using a machine learning approach to predict mortality in critically ill influenza patients: a cross-sectional retrospective multicentre study in Taiwan. BMJ open, 10(2), e033898.
HWANG, S. J., TSAI, J. C., & CHEN, H. C. (2010). Epidemiology, impact and preventive care of chronic kidney disease in Taiwan. Nephrology, 15, 3-9.
Ij, H. (2018). Statistics versus machine learning. Nat Methods, 15(4), 233.
James, C., Ranson, J. M., Everson, R., & Llewellyn, D. J. (2021). Performance of machine learning algorithms for predicting progression to dementia in memory clinic patients. JAMA network open, 4(12), e2136553-e2136553.
Johnson, A. E., Bulgarelli, L., Shen, L., Gayles, A., Shammout, A., Horng, S., . . . Lehman, L.-w. H. (2023). MIMIC-IV, a freely accessible electronic health record dataset. Scientific data, 10(1), 1.
Kang, M. W., Kim, J., Kim, D. K., Oh, K.-H., Joo, K. W., Kim, Y. S., & Han, S. S. (2020). Machine learning algorithm to predict mortality in patients undergoing continuous renal replacement therapy. Critical Care, 24(1), 1-9.
Kashihara, N., Kohsaka, S., Kanda, E., Okami, S., & Yajima, T. (2019). Hyperkalemia in real-world patients under continuous medical care in Japan. Kidney international reports, 4(9), 1248-1260.
Kellum, J. A., Romagnani, P., Ashuntantang, G., Ronco, C., Zarbock, A., & Anders, H.-J. (2021). Acute kidney injury. Nature reviews Disease primers, 7(1), 52.
Kerr, K. F., Brown, M. D., Zhu, K., & Janes, H. (2016). Assessing the clinical impact of risk prediction models with decision curves: guidance for correct interpretation and appropriate use. Journal of Clinical Oncology, 34(21), 2534.
Khwaja, A. (2012). KDIGO clinical practice guidelines for acute kidney injury. Nephron Clinical Practice, 120(4), c179-c184.
Kurzhagen, J., Dellepiane, S., Cantaluppi, V., & Rabb, H. (2020). AKI: an increasingly recognized risk factor for CKD development and progression. Journal of Nephrology, 33(6), 1171-1187.
Lee, T.-T., Cheng, S.-H., Chen, C.-C., & Lai, M.-S. (2010). A pay-for-performance program for diabetes care in Taiwan: a preliminary assessment. The American journal of managed care, 16(1), 65-69.
Li, D. H., Wald, R., Blum, D., McArthur, E., James, M. T., Burns, K. E., . . . Lebovic, G. (2020). Predicting mortality among critically ill patients with acute kidney injury treated with renal replacement therapy: Development and validation of new prediction models. Journal of Critical Care, 56, 113-119.
Lin, M.-Y., Chang, M.-Y., Wu, P.-Y., Wu, P.-H., Lin, M.-H., Hsu, C.-C., . . . Chiu, Y.-W. (2022). Multidisciplinary care program in pre-end-stage kidney disease from 2010 to 2018 in Taiwan. Journal of the Formosan Medical Association, 121, S64-S72.
London, A. J. (2019). Artificial intelligence and black‐box medical decisions: accuracy versus explainability. Hastings Center Report, 49(1), 15-21.
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, 30.
Luo, J., Brunelli, S. M., Jensen, D. E., & Yang, A. (2016). Association between serum potassium and outcomes in patients with reduced kidney function. Clinical Journal of the American Society of Nephrology: CJASN, 11(1), 90.
Lv, W., Booz, G. W., Wang, Y., Fan, F., & Roman, R. J. (2018). Inflammation and renal fibrosis: recent developments on key signaling molecules as potential therapeutic targets. European journal of pharmacology, 820, 65-76.
Maccariello, E., Valente, C., Nogueira, L., Bonomo, H., Ismael, M., Machado, J. E., . . . Rocha, E. (2010). SAPS 3 scores at the start of renal replacement therapy predict mortality in critically ill patients with acute kidney injury. Kidney International, 77(1), 51-56.
Makris, K., & Spanou, L. (2016). Acute kidney injury: definition, pathophysiology and clinical phenotypes. The clinical biochemist reviews, 37(2), 85.
Mansoor, F., Bai, P., Kaur, N., Sultan, S., Sharma, S., Dilip, A., . . . Rizwan, A. (2021). Evaluation of serum electrolyte levels in patients with anemia. Cureus, 13(10).
Moreno, R. P., Metnitz, P. G., Almeida, E., Jordan, B., Bauer, P., Campos, R. A., . . . Le Gall, J.-R. (2005). SAPS 3—From evaluation of the patient to evaluation of the intensive care unit. Part 2: Development of a prognostic model for hospital mortality at ICU admission. Intensive care medicine, 31, 1345-1355.
Nassar, A. P., Malbouisson, L. S., & Moreno, R. (2014). Evaluation of simplified acute physiology score 3 performance: a systematic review of external validation studies. Critical Care, 18, 1-9.
Palevsky, P. M., O'Connor, T., Zhang, J. H., Star, R. A., & Smith, M. W. (2005). Design of the VA/NIH Acute Renal Failure Trial Network (ATN) Study: intensive versus conventional renal support in acute renal failure. Clinical Trials, 2(5), 423-435.
Pollard, T. J., Johnson, A. E., Raffa, J. D., Celi, L. A., Mark, R. G., & Badawi, O. (2018). The eICU Collaborative Research Database, a freely available multi-center database for critical care research. Scientific data, 5(1), 1-13.
Price, W. N. (2018). Big data and black-box medical algorithms. Science translational medicine, 10(471), eaao5333.
Raebel, M. A. (2012). Hyperkalemia associated with use of angiotensin‐converting enzyme inhibitors and angiotensin receptor blockers. Cardiovascular therapeutics, 30(3), e156-e166.
Rana, A., Rawat, A. S., Bijalwan, A., & Bahuguna, H. (2018). Application of multi layer (perceptron) artificial neural network in the diagnosis system: a systematic review. Paper presented at the 2018 International conference on research in intelligent and computing in engineering (RICE).
Royston, P., & White, I. R. (2011). Multiple imputation by chained equations (MICE): implementation in Stata. Journal of statistical software, 45, 1-20.
Sarica, A., Cerasa, A., & Quattrone, A. (2017). Random forest algorithm for the classification of neuroimaging data in Alzheimer's disease: a systematic review. Frontiers in aging neuroscience, 9, 329.
Singbartl, K., & Kellum, J. A. (2012). AKI in the ICU: definition, epidemiology, risk stratification, and outcomes. Kidney International, 81(9), 819-825.
Soo, A., Zuege, D. J., Fick, G. H., Niven, D. J., Berthiaume, L. R., Stelfox, H. T., & Doig, C. J. (2019). Describing organ dysfunction in the intensive care unit: a cohort study of 20,000 patients. Critical Care, 23(1), 1-15.
Stevens, J. R., Suyundikov, A., & Slattery, M. L. (2016). Accounting for missing data in clinical research. Jama, 315(5), 517-518.
Thakar, C. V., Arrigain, S., Worley, S., Yared, J.-P., & Paganini, E. P. (2005). A clinical score to predict acute renal failure after cardiac surgery. Journal of the American Society of Nephrology, 16(1), 162-168.
Uchino, S., Kellum, J. A., Bellomo, R., Doig, G. S., Morimatsu, H., Morgera, S., . . . Macedo, E. (2005). Acute renal failure in critically ill patients: a multinational, multicenter study. Jama, 294(7), 813-818.
Vickers, A. J., & Elkin, E. B. (2006). Decision curve analysis: a novel method for evaluating prediction models. Medical Decision Making, 26(6), 565-574.
Vickers, A. J., & Holland, F. (2021). Decision curve analysis to evaluate the clinical benefit of prediction models. The Spine Journal, 21(10), 1643-1648.
Vincent, J.-L., Moreno, R., Takala, J., Willatts, S., De Mendonça, A., Bruining, H., . . . Thijs, L. G. (1996). The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure: On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine (see contributors to the project in the appendix). In: Springer-Verlag.
Wald, R., Adhikari, N. K., Smith, O. M., Weir, M. A., Pope, K., Cohen, A., . . . Soth, M. (2015). Comparison of standard and accelerated initiation of renal replacement therapy in acute kidney injury. Kidney International, 88(4), 897-904.
Wang, H. E., Muntner, P., Chertow, G. M., & Warnock, D. G. (2012). Acute kidney injury and mortality in hospitalized patients. American journal of nephrology, 35(4), 349-355.
Wang, J. S., Yen, F. S., Lin, K. D., Shin, S. J., Hsu, Y. H., Hsu, C. C., & China, D. K. D. R. C. o. t. D. A. o. t. R. o. (2021). Epidemiological characteristics of diabetic kidney disease in Taiwan. Journal of Diabetes Investigation, 12(12), 2112-2123.
Webster, A. C., Nagler, E. V., Morton, R. L., & Masson, P. (2017). Chronic kidney disease. The lancet, 389(10075), 1238-1252.
Weir, M. R., & Rolfe, M. (2010). Potassium homeostasis and renin-angiotensin-aldosterone system inhibitors. Clinical Journal of the American Society of Nephrology, 5(3), 531-548.
Zarbock, A., Kellum, J. A., Schmidt, C., Van Aken, H., Wempe, C., Pavenstädt, H., . . . Meersch, M. (2016). Effect of early vs delayed initiation of renal replacement therapy on mortality in critically ill patients with acute kidney injury: the ELAIN randomized clinical trial. Jama, 315(20), 2190-2199.
Zimmerman, J. E., Kramer, A. A., McNair, D. S., & Malila, F. M. (2006). Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today’s critically ill patients. Critical Care Medicine, 34(5), 1297-1310.