Machine Learning for Predicting Chronic Renal Disease Progression in COVID-19 Patients with Acute Renal Injury: A Feasibility Study
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Type and Design
2.2. Population, Setting, and Period of Study
2.3. Definitions
- Asymptomatic was characterized by the absence of symptoms.
- Mild disease was defined as symptomatic patients who met the case definition for COVID-19, without evidence of viral pneumonia or hypoxia.
- Moderate disease involved clinical signs of pneumonia, including fever, cough, dyspnea, and fast breathing, but there were no signs of severe pneumonia, with an SpO2 level of 90% or higher while breathing room air.
- Severe disease involved clinical signs of pneumonia, along with one of the following: a respiratory rate of over 30 breaths per minute, severe respiratory distress, or an SpO2 level below 90% while breathing room air.
- Critical disease was characterized by the presence of acute respiratory distress syndrome (ARDS).
- Stage 1: creatinine increase ≥ 0.3 mg/dL or ≥150–200% of the baseline.
- Stage 2: creatinine increase ≥ 200–300% of the baseline.
- Stage 3: creatinine increases of ≥300% of the baseline or a serum creatinine level > 4 mg/dL with a sudden increase of at least 0.5 mg/dL [12].
2.4. Data Source of AKI and CKD Progression
2.5. Prediction Variables
2.6. Data Preprocessing
2.7. Data Analysis
2.8. Metrics for Evaluating Model Performance
3. Results
3.1. Population, Setting, and Period of Study
3.2. Clinical and Epidemiological Characteristics of the Cohort
3.3. Development and Internal Validation of the Predictive Models
4. Discussion
5. Conclusions
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Hassanein, H.; Radhakrishnan, Y.; Sedor, J.; Vachharajani, T.; Vachharajani, V.T.; Augustine, J.; Demirjian, S.; Thomas, G. COVID-19 and the kidney. Clevel. Clin. J. Med. 2020, 87, 619–631. [Google Scholar] [CrossRef] [PubMed]
- Farouk, S.S.; Fiaccadori, E.; Cravedi, P.; Campbell, K.N. COVID-19 and the kidney: What we think we know so far and what we don’t. J. Nephrol. 2020, 33, 1213–1218. [Google Scholar] [CrossRef] [PubMed]
- Coca, S.G.; Singanamala, S.; Parikh, C.R. Chronic kidney disease after acute kidney injury: A systematic review and meta-analysis. Kidney Int. 2012, 81, 442–448. [Google Scholar] [CrossRef] [PubMed]
- Hsu, C.Y.; Chinchilli, V.M.; Coca, S.; Devarajan, P.; Ghahramani, N.; Go, A.S.; Hsu, R.K.; Ikizler, T.A.; Kaufman, J.; Liu, K.D.; et al. Post-acute kidney injury proteinuria and subsequent kidney disease progression: The assessment, serial evaluation, and subsequent sequelae in acute kidney injury (ASSESS-AKI) study. JAMA Intern. Med. 2020, 180, 402–410. [Google Scholar] [CrossRef] [PubMed]
- James, M.T.; Pannu, N.; Hemmelgarn, B.R.; Austin, P.C.; Tan, Z.; McArthur, E.; Manns, B.J.; Tonelli, M.; Wald, R.; Quinn, R.R.; et al. Derivation and external validation of prediction models for advanced chronic kidney disease following acute kidney injury. JAMA 2017, 318, 1787–1797. [Google Scholar] [CrossRef]
- Brereton, R.G.; Lloyd, G.R. Support vector machines for classification and regression. Analyst 2010, 135, 230–267. [Google Scholar] [CrossRef]
- Swapna, G.; Vinayakumar, R.; Soman, K.P. Diabetes detection using deep learning algorithms. ICT Express 2018, 2, 243–246. [Google Scholar]
- Talavera, J.O.; Roy-García, I.; Palacios-Cruz, L.; Rivas-Ruiz, R.; Hoyo, I.; Pérez-Rodríguez, M. Back to the clinic. Methods I. Research designs. Higher quality of information, more certainty to the answer. Gac. Médica México 2023, 155, 3484. [Google Scholar] [CrossRef]
- Wang, C.; Liu, F. (Eds.) Textbook of Clinical Epidemiology: For Medical Students; Springer Nature Singapore: Singapore, 2023. [Google Scholar]
- Living Guidance for Clinical Management of COVID-19. Available online: https://www.who.int/publications-detail-redirect/WHO-2019-nCoV-clinical-2021-2 (accessed on 19 May 2024).
- Inker, L.A.; Eneanya, N.D.; Coresh, J.; Tighiouart, H.; Wang, D.; Sang, Y.; Crews, D.C.; Doria, A.; Estrella, M.M.; Froissart, M.; et al. New creatinine- and cystatin C–based equations to estimate GFR without race. N. Engl. J. Med. 2021, 385, 1737–1749. [Google Scholar] [CrossRef]
- Kellum, J.A.; Lameire, N.; Aspelin, P.; Barsoum, R.S.; Burdmann, E.A.; Goldstein, S.L.; Herzog, C.A.; Joannidis, M.; Kribben, A.; Levey, A.S.; et al. Kidney disease: Improving global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int. Suppl. 2012, 2, 1–138. [Google Scholar]
- Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int. Suppl. 2013, 3, 1–150. [Google Scholar]
- Batista, G.E.A.P.A.; Prati, R.C.; Monard, M.C. A study of the behavior of several methods for balancing machine learning training data. ACM SIGKDD Explor. Newsl. 2004, 6, 20–29. [Google Scholar] [CrossRef]
- Misra, R.; Shyamasundar, R.K.; Chaturvedi, A.; Omer, R. (Eds.) Machine Learning and Big Data Analytics (Proceedings of International Conference on Machine Learning and Big Data Analytics (ICMLBDA) 2021); Lecture Notes in Networks and Systems; Springer International Publishing: Cham, Switzerland, 2022; Volume 256. [Google Scholar]
- Sanguansat, P. Principal Component Analysis—Multidisciplinary Applications; IntechOpen: London, UK, 2012. [Google Scholar] [CrossRef]
- Williams, C.K.I. Learning with Kernels: Support vector machines, regularization, optimization, and beyond. J. Am. Stat. Assoc. 2003, 98, 489. [Google Scholar] [CrossRef]
- Nguyen, J.M.; Jézéquel, P.; Gillois, P.; Silva, L.; Ben Azzouz, F.; Lambert-Lacroix, S.; Juin, P.; Campone, M.; Gaultier, A.; Moreau-Gaudry, A.; et al. Random forest of perfect trees: Concept, performance, applications and perspectives. Bioinformatics 2021, 37, 2165–2174. [Google Scholar] [CrossRef] [PubMed]
- Dreiseitl, S.; Ohno-Machado, L. Logistic regression and artificial neural network classification models: A methodology review. J. Biomed. Inform. 2002, 35, 352–359. [Google Scholar] [CrossRef] [PubMed]
- Schapire, R.E.; Freund, Y. Boosting: Foundations and Algorithms, 1st ed.; The MIT Press: Cambridge, MA, USA, 2012; pp. 23–52. [Google Scholar]
- Cook, N.R. Statistical evaluation of prognostic versus diagnostic models: Beyond the ROC curve. Clin. Chem. 2008, 54, 17–23. [Google Scholar] [CrossRef] [PubMed]
- Silver, S.A.; Beaubien-Souligny, W.; Shah, P.S.; Harel, S.; Blum, D.; Kishibe, T.; Meraz-Munoz, A.; Wald, R.; Harel, Z. The prevalence of acute kidney injury in patients hospitalized with COVID-19 infection: A systematic review and meta-analysis. Kidney Med. 2021, 3, 83–98. [Google Scholar] [CrossRef] [PubMed]
- Van Son, J.; Oussaada, S.M.; Şekercan, A.; Beudel, M.; Dongelmans, D.A.; Van Assen, S.; Eland, I.A.; Moeniralam, H.S.; Dormans, T.P.J.; Van Kalkeren, C.A.J.; et al. Overweight and obesity are associated with acute kidney injury and acute respiratory distress syndrome, but not with increased mortality in hospitalized COVID-19 patients: A retrospective cohort study. Front. Endocrinol. 2021, 12, 747732. [Google Scholar] [CrossRef]
- Romagnani, P.; Remuzzi, G.; Glassock, R.; Levin, A.; Jager, K.J.; Tonelli, M.; Massy, Z.; Wanner, C.; Anders, H.-J. Chronic Kidney Disease. Nat. Rev. Dis. Primers 2017, 3, 17088. [Google Scholar] [CrossRef]
- Helal, I.; Fick-Brosnahan, G.M.; Reed-Gitomer, B.; Schrier, R.W. Glomerular Hyperfiltration: Definitions, Mechanisms and Clinical Implications. Nat. Rev. Nephrol. 2012, 8, 293–300. [Google Scholar] [CrossRef]
- Deicher, R.; Hörl, W.H. Anaemia as a Risk Factor for the Progression of Chronic Kidney Disease. Curr. Opin. Nephrol. Hypertens. 2003, 12, 139. [Google Scholar] [CrossRef] [PubMed]
- Wang, B.; Li, Z.-L.; Zhang, Y.-L.; Wen, Y.; Gao, Y.-M.; Liu, B.-C. Hypoxia and Chronic Kidney Disease. EBioMedicine 2022, 77, 103942. [Google Scholar] [CrossRef] [PubMed]
- Iseki, K.; Kohagura, K. Anemia as a Risk Factor for Chronic Kidney Disease. Kidney Int. 2007, 72, S4–S9. [Google Scholar] [CrossRef] [PubMed]
- Agarwal, R.; Light, R.P. Patterns and Prognostic Value of Total and Differential Leukocyte Count in Chronic Kidney Disease. Clin. J. Am. Soc. Nephrol. 2011, 6, 1393–1399. [Google Scholar] [CrossRef] [PubMed]
- Litjens, N.H.R.; Van Druningen, C.J.; Betjes, M.G.H. Progressive Loss of Renal Function Is Associated with Activation and Depletion of Naive T Lymphocytes. Clin. Immunol. 2006, 118, 83–91. [Google Scholar] [CrossRef] [PubMed]
- Anders, H.-J.; Huber, T.B.; Isermann, B.; Schiffer, M. CKD in Diabetes: Diabetic Kidney Disease versus Nondiabetic Kidney Disease. Nat. Rev. Nephrol. 2018, 14, 361–377. [Google Scholar] [CrossRef] [PubMed]
- Dessie, Z.G.; Zewotir, T. Mortality-related risk factors of COVID-19: A systematic review and meta-analysis of 42 studies and 423,117 patients. BMC Infect. Dis. 2021, 21, 855. [Google Scholar] [CrossRef] [PubMed]
- Li, Y.; Qin, Y.; Wang, L.; Chen, J.; Ma, S. Grouped variable selection using area under the ROC with imbalanced data. Commun. Stat. Simul. Comput. 2016, 45, 1268–1280. [Google Scholar] [CrossRef]
- Cortes, C.; Vapnik, V.; Saitta, L. Support-vector networks. Mach. Learn. 1995, 20, 273–297. [Google Scholar] [CrossRef]
- Piccialli, V.; Sciandrone, M. Nonlinear optimization and support vector machines. 4OR-Q. J. Oper. Res. 2022, 16, 111–149. [Google Scholar] [CrossRef]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef]
- Ferreira, J.A. Models under which random forests perform badly; consequences for applications. Comput. Stat. 2022, 37, 1839–1854. [Google Scholar] [CrossRef]
- Aggarwal, C.C. Outlier Analysis, 2nd ed.; Springer: Cham, Switzerland; New York, NY, USA, 2017; pp. 219–247. [Google Scholar]
- Koyner, J.L.; Carey, K.A.; Edelson, D.P.; Churpek, M.M. The development of a machine learning inpatient acute kidney injury prediction model. Crit. Care Med. 2018, 46, 1070–1077. [Google Scholar] [CrossRef] [PubMed]
- Churpek, M.M.; Carey, K.A.; Edelson, D.P.; Singh, T.; Astor, B.C.; Gilbert, E.R.; Winslow, C.; Shah, N.; Afshar, M.; Koyner, J.L. Internal and external validation of a machine learning risk score for acute kidney injury. JAMA Netw. Open 2020, 3, e2012892. [Google Scholar] [CrossRef] [PubMed]
- Hsu, C.N.; Liu, C.L.; Tain, Y.L.; Kuo, C.Y.; Lin, Y.C. Machine learning model for risk prediction of community-acquired acute kidney injury hospitalization from electronic health records: Development and validation study. J. Med. Internet Res. 2020, 22, e16903. [Google Scholar] [CrossRef]
- Tomašev, N.; Glorot, X.; Rae, J.W.; Zielinski, M.; Askham, H.; Saraiva, A.; Mottram, A.; Meyer, C.; Ravuri, S.; Protsyuk, I.; et al. A clinically applicable approach to continuous prediction of future acute kidney injury. Nature 2019, 572, 116–119. [Google Scholar] [CrossRef]
- Simonov, M.; Ugwuowo, U.; Moreira, E.; Yamamoto, Y.; Biswas, A.; Martin, M.; Testani, J.; Wilson, F.P. A simple real-time model for predicting acute kidney injury in hospitalized patients in the US: A descriptive modeling study. PLoS Med. 2019, 16, e1002861. [Google Scholar] [CrossRef]
- Malhotra, R.; Kashani, K.B.; Macedo, E.; Kim, J.; Bouchard, J.; Wynn, S.; Li, G.; Ohno-Machado, L.; Mehta, R. A risk prediction score for acute kidney injury in the intensive care unit. Nephrol. Dial. Transplant. 2017, 32, 814–822. [Google Scholar] [CrossRef]
- Flechet, M.; Güiza, F.; Schetz, M.; Wouters, P.; Vanhorebeek, I.; Derese, I.; Gunst, J.; Spriet, I.; Casaer, M.; Van den Berghe, G.; et al. AKIpredictor, an online prognostic calculator for acute kidney injury in adult critically ill patients: Development, validation and comparison to serum neutrophil gelatinase-associated lipocalin. Intensive Care Med. 2017, 43, 764–773. [Google Scholar] [CrossRef]
- Gross, C.; Miao Jonasson, J.; Buchebner, D.; Agvall, B. Prognosis and mortality within 90 days in community-acquired acute kidney injury in the Southwest of Sweden. BMC Nephrol. 2023, 24, 171. [Google Scholar] [CrossRef]
- Sawhney, S.; Marks, A.; Fluck, N.; Levin, A.; Prescott, G.; Black, C. Intermediate and long-term outcomes of survivors of acute kidney injury episodes: A large population-based cohort study. Am. J. Kidney Dis. 2017, 69, 18–28. [Google Scholar] [CrossRef] [PubMed]
- Bagshaw, S.M.; Wald, R. Indications and timing of continuous renal replacement therapy application. Contrib. Nephrol. 2018, 194, 25–37. [Google Scholar] [PubMed]
- Schiffl, H. Discontinuation of renal replacement therapy in critically ill patients with severe acute kidney injury: Predictive factors of renal function recovery. Int. Urol. Nephrol. 2018, 50, 1845–1851. [Google Scholar] [CrossRef] [PubMed]
- Jones, J.; Holmen, J.; De Graauw, J.; Jovanovich, A.; Thornton, S.; Chonchol, M. Association of complete recovery from acute kidney injury with incident CKD stage 3 and all-cause mortality. Am. J. Kidney Dis. 2012, 60, 402–408. [Google Scholar] [CrossRef]
- Flammia, R.S.; Tufano, A.; Proietti, F.; Gerolimetto, C.; De Nunzio, C.; Franco, G.; Leonardo, C. Renal Surgery for Kidney Cancer: Is Preoperative Proteinuria a Predictor of Functional and Survival Outcomes after Surgery? A Systematic Review of the Literature. Minerva Urol. Nephrol. 2022, 74, 255–264. [Google Scholar] [CrossRef]
- Siew, E.D.; Liu, K.D.; Bonn, J.; Chinchilli, V.; Dember, L.M.; Girard, T.D.; Greene, T.; Hernandez, A.F.; Alp Ikizler, T.; James, M.T.; et al. Improving care for patients after hospitalization with AKI. J. Am. Soc. Nephrol. 2020, 31, 2237–2241. [Google Scholar] [CrossRef] [PubMed]
- Kurzhagen, J.T.; Dellepiane, S.; Cantaluppi, V.; Rabb, H. AKI: An increasingly recognized risk factor for CKD development and progression. J. Nephrol. 2020, 33, 1171–1187. [Google Scholar] [CrossRef] [PubMed]
- Ruiz-Ortega, M.; Rayego-Mateos, S.; Lamas, S.; Ortiz, A.; Rodrigues-Diez, R.R. Targeting the progression of chronic kidney disease. Nat. Rev. Nephrol. 2020, 16, 269–288. [Google Scholar] [CrossRef]
Clinical and Laboratory Variables | Total, AKI (n = 131) | No Progression to CKD (n = 91) | Progression to CKD (n = 40) | p |
---|---|---|---|---|
Age (years), M (P25–P75) | 61.0 (51.0–70.0) | 59.0 (47.0–70.0) | 63.0 (55.0–69.5) | 0.167 |
Male, n (%) | 81 (61.8) | 56 (61.5) | 25 (62.5) | 0.917 |
BMI (kg/m2), M (P25–P75) | 31.6 (27.4–35.1) | 32.0 (27.6–35.3) | 30.2 (26.6–35.0) | 0.162 |
Comorbidities, n (%) | ||||
Diabetes | 53 (40.5) | 30 (33.0) | 23 (57.5) | 0.008 |
Hypertension | 74 (56.5) | 49 (53.8) | 25 (62.5) | 0.358 |
Obesity | 85 (64.9) | 64 (70.3) | 21 (52.5) | 0.049 |
History of kidney disease | 45 (34.4) | 25 (27.5) | 20 (50.0) | 0.012 |
COVID-19 metrics | ||||
COVID-19 severity, n (%) | 0.343 | |||
Asymptomatic | 10 (7.6) | 7 (7.7) | 3 (7.5) | |
Mild disease | 11 (8.4) | 7 (7.7) | 4 (10.0) | |
Moderate disease | 30 (22.9) | 21 (23.3) | 9 (22.5) | |
Severe disease | 71 (54.2) | 47 (51.6) | 24 (60.0) | |
Critical disease | 9 (6.9) | 9 (9.9) | 0 | |
Kidney disease severity (Stage 3), n (%) | 0.033 | |||
1 | 82 (62.6) | 61 (67.0) | 21 (52.5) | |
2 | 36 (27.5) | 25 (27.5) | 11 (27.5) | |
3 | 13 (9.9) | 5 (5.5) | 8 (20.0) | |
Cardiovascular event, n (%) | 9 (6.9) | 7 (7.7) | 2 (5.0) | 0.575 |
Number of days on oxygen, M (P25–P75) | 9.0 (3.0–15.0) | 9.0 (3.0–14.0) | 10 (3–15) | 0.672 |
Number of days in hospital, M (P25–P75) | 15.48 (13.47–17.49) | 13.0 (8.0–21.0) | 13 (9–17) | 0.785 |
Laboratory variables admission, M (P25–P75) | ||||
Hemoglobin (g/dL) | 13.2 (12.0–14.7) | 13.8 (12.3–15.0) | 12.7 (11.0–13.8) | 0.008 |
Hematocrit (%) | 40.0 (36.0–45.0) | 41.0 (37.0–45.0) | 39.0 (32.2–42.0) | 0.008 |
Creatinine (mg/dL) | 1.2 (0.9–1.8) | 1.1 (0.9–1.5) | 1.5 (1.1–2.1) | 0.012 |
Clearance of creatinine (ml/min) | 61.0 (36.0–82.5) | 63.0 (39.0–85.8) | 46.6 (29.0–68.9) | 0.019 |
Laboratory variables discharge, M (P25–P75) | ||||
Neutrophils (%) | 74.0 (65.0–80.0) | 72.0 (64.0–67.0) | 78.0 (72.0–81.0) | 0.012 |
Lymphocytes (%) | 17.0 (12.0–24.0) | 18.0 (13.0–25.0) | 13.0 (11.0–19.2) | 0.003 |
Glucose (mg/dL) | 107.0 (89.0–137.0) | 103 (87.0–123.0) | 120.5 (94.0–183.0) | 0.025 |
Creatinine (mg/dL) | 0.9 (0.7–1.1) | 0.8 (0.6–1.0) | 1.0 (0.8–1.2) | 0.006 |
Creatinine clearance (ml/min) | 90.5 (65.8–101.3) | 94.0 (74.0–104.0) | 78 (55.9–97.4) | 0.010 |
Validation | |||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Accuracy | Precision | Recall | F1-Score | Sensitivity | Specificity | AUC | |||||||||
All variables (44) | SVM | 81.62 | 1.71 | 83.56 | 4.78 | 80.53 | 4.33 | 81.81 | 1.51 | 80.53 | 4.33 | 82.78 | 6.11 | 90.15 | 2.70 |
RF | 79.46 | 3.42 | 76.69 | 3.03 | 86.32 | 5.66 | 81.14 | 3.49 | 86.32 | 5.66 | 72.22 | 4.54 | 88.06 | 1.53 | |
RL | 68.65 | 4.80 | 66.98 | 3.73 | 76.84 | 8.67 | 71.41 | 5.21 | 76.84 | 8.67 | 60.00 | 6.31 | 66.58 | 3.79 | |
Boosting | 72.70 | 5.62 | 69.61 | 5.93 | 84.74 | 6.77 | 76.17 | 4.23 | 84.74 | 6.77 | 60.00 | 11.94 | 77.65 | 4.73 | |
ROC (8) | SVM | 93.22 | 1.74 | 100.00 | 0.00 | 88.05 | 3.06 | 93.62 | 1.76 | 88.05 | 3.06 | 100.00 | 0.00 | 98.43 | 1.79 |
RF | 80.81 | 2.37 | 76.11 | 2.99 | 91.58 | 2.72 | 83.08 | 1.80 | 91.58 | 2.72 | 69.44 | 5.40 | 89.88 | 2.76 | |
RL | 65.95 | 2.28 | 65.21 | 1.47 | 72.11 | 5.58 | 68.41 | 3.10 | 72.11 | 5.58 | 59.44 | 2.68 | 72.11 | 0.50 | |
Boosting | 74.05 | 5.58 | 70.50 | 5.81 | 86.32 | 7.53 | 77.36 | 4.78 | 86.32 | 7.53 | 61.11 | 11.11 | 78.95 | 6.97 | |
SHAP (11) | SVM | 92.16 | 2.37 | 100.00 | 0.00 | 84.74 | 4.61 | 91.68 | 2.79 | 84.74 | 4.61 | 100.00 | 0.00 | 93.68 | 2.61 |
RF | 79.73 | 3.43 | 79.01 | 4.77 | 83.16 | 7.36 | 80.73 | 3.57 | 83.16 | 7.36 | 76.11 | 7.43 | 85.01 | 3.08 | |
RL | 70.54 | 0.85 | 65.93 | 1.14 | 88.42 | 3.33 | 75.49 | 0.98 | 88.42 | 3.33 | 51.67 | 3.75 | 73.30 | 0.76 | |
Boosting | 83.51 | 4.12 | 77.44 | 2.61 | 95.79 | 7.77 | 85.53 | 4.18 | 95.79 | 7.77 | 70.56 | 3.75 | 85.70 | 4.37 | |
PCA (24) | SVM | 85.14 | 1.42 | 100.00 | 0.00 | 71.05 | 2.77 | 83.05 | 1.90 | 71.05 | 2.77 | 100.00 | 0.00 | 89.04 | 2.42 |
RF | 77.84 | 4.19 | 78.36 | 3.12 | 78.42 | 7.21 | 78.29 | 4.87 | 78.42 | 7.21 | 77.22 | 3.15 | 86.54 | 3.75 | |
RL | 72.97 | 2.55 | 70.16 | 2.59 | 82.63 | 2.54 | 75.86 | 2.07 | 82.63 | 2.54 | 62.78 | 4.57 | 77.78 | 1.51 | |
Boosting | 77.57 | 5.26 | 75.20 | 6.43 | 85.26 | 5.44 | 79.69 | 4.12 | 85.26 | 5.44 | 69.44 | 10.88 | 80.45 | 5.28 | |
LR forward (10) | SVM | 92.43 | 2.79 | 100.00 | 0.00 | 85.26 | 5.44 | 91.96 | 3.27 | 85.26 | 5.44 | 100.00 | 0.00 | 93.99 | 2.02 |
RF | 78.11 | 3.70 | 75.00 | 4.99 | 86.84 | 2.77 | 80.36 | 2.69 | 86.84 | 2.77 | 68.89 | 8.36 | 86.14 | 3.03 | |
RL | 74.59 | 1.89 | 72.69 | 2.08 | 81.05 | 2.72 | 76.61 | 1.71 | 81.05 | 2.72 | 67.78 | 3.51 | 79.62 | 1.61 | |
Boosting | 71.89 | 3.65 | 67.91 | 3.62 | 86.32 | 5.08 | 75.92 | 3.05 | 86.32 | 5.08 | 56.67 | 6.83 | 82.95 | 3.43 | |
LR backward (14) | SVM | 90.54 | 1.42 | 100.00 | 0.00 | 81.58 | 2.77 | 89.83 | 1.68 | 81.58 | 2.77 | 100.00 | 0.00 | 90.86 | 1.76 |
RF | 77.03 | 2.92 | 75.28 | 2.85 | 82.63 | 7.46 | 78.58 | 3.46 | 82.63 | 7.46 | 71.11 | 5.74 | 85.64 | 3.30 | |
RL | 67.30 | 4.50 | 68.07 | 4.25 | 68.42 | 5.55 | 68.20 | 4.58 | 68.42 | 5.55 | 66.11 | 4.86 | 76.78 | 1.28 | |
Boosting | 81.08 | 4.41 | 75.85 | 4.06 | 93.16 | 8.25 | 83.41 | 4.22 | 93.16 | 8.25 | 68.33 | 7.88 | 85.77 | 4.79 | |
LR forward admission (9) | SVM | 80.54 | 3.99 | 81.52 | 4.18 | 80.53 | 6.10 | 80.89 | 4.18 | 80.53 | 6.10 | 80.56 | 5.40 | 86.41 | 2.88 |
RF | 74.59 | 3.17 | 74.06 | 3.24 | 77.89 | 4.15 | 75.88 | 3.06 | 77.89 | 4.15 | 71.11 | 4.38 | 85.26 | 2.35 | |
RL | 72.97 | 3.37 | 70.31 | 2.96 | 82.11 | 5.08 | 75.69 | 3.29 | 82.11 | 5.08 | 63.33 | 4.68 | 77.69 | 2.40 | |
Boosting | 74.59 | 5.58 | 71.14 | 5.84 | 86.32 | 9.35 | 77.66 | 5.15 | 86.32 | 9.35 | 62.22 | 11.65 | 78.82 | 6.01 | |
LR backward admission (12) | SVM | 87.03 | 3.07 | 100.00 | 0.00 | 74.74 | 5.98 | 85.42 | 4.04 | 74.74 | 5.98 | 100.00 | 0.00 | 89.02 | 3.85 |
RF | 73.51 | 5.81 | 74.59 | 7.17 | 74.74 | 10.76 | 74.15 | 6.12 | 74.74 | 10.76 | 72.22 | 10.48 | 82.06 | 4.44 | |
RL | 66.76 | 3.13 | 67.74 | 2.78 | 67.37 | 5.98 | 67.46 | 3.71 | 67.37 | 5.98 | 66.11 | 4.10 | 78.25 | 0.74 | |
Boosting | 74.05 | 5.28 | 68.46 | 5.27 | 93.16 | 7.46 | 78.67 | 4.04 | 93.16 | 7.46 | 53.89 | 10.81 | 83.17 | 4.08 |
Discrimination Performance | Calibration Performance | Sensitivity | Specificity | ||
---|---|---|---|---|---|
Dataset | AUC (95% CI) | Brier Score | Hosmer–Lemeshow Test | ||
Training cohort | 99.88 (0.11) | 0.011 | 0.751 | 97.66 (1.81) | 99.93 (0.46) |
Validation cohort | 98.43 (1.79) | 0.011 | 0.635 | 88.05 (3.06) | 100 (0) |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Gracida-Osorno, C.; Molina-Salinas, G.M.; Góngora-Hernández, R.; Brito-Loeza, C.; Uc-Cachón, A.H.; Paniagua-Sierra, J.R. Machine Learning for Predicting Chronic Renal Disease Progression in COVID-19 Patients with Acute Renal Injury: A Feasibility Study. Biomedicines 2024, 12, 1511. https://doi.org/10.3390/biomedicines12071511
Gracida-Osorno C, Molina-Salinas GM, Góngora-Hernández R, Brito-Loeza C, Uc-Cachón AH, Paniagua-Sierra JR. Machine Learning for Predicting Chronic Renal Disease Progression in COVID-19 Patients with Acute Renal Injury: A Feasibility Study. Biomedicines. 2024; 12(7):1511. https://doi.org/10.3390/biomedicines12071511
Chicago/Turabian StyleGracida-Osorno, Carlos, Gloria María Molina-Salinas, Roxana Góngora-Hernández, Carlos Brito-Loeza, Andrés Humberto Uc-Cachón, and José Ramón Paniagua-Sierra. 2024. "Machine Learning for Predicting Chronic Renal Disease Progression in COVID-19 Patients with Acute Renal Injury: A Feasibility Study" Biomedicines 12, no. 7: 1511. https://doi.org/10.3390/biomedicines12071511
APA StyleGracida-Osorno, C., Molina-Salinas, G. M., Góngora-Hernández, R., Brito-Loeza, C., Uc-Cachón, A. H., & Paniagua-Sierra, J. R. (2024). Machine Learning for Predicting Chronic Renal Disease Progression in COVID-19 Patients with Acute Renal Injury: A Feasibility Study. Biomedicines, 12(7), 1511. https://doi.org/10.3390/biomedicines12071511