A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease
Background: This study aimed to develop and validate a machine learning (ML) model based on serum Klotho for predicting end-stage kidney disease (ESKD) and cardiovascular disease (CVD) in patients with chronic kidney disease (CKD). Methods: Five different ML models were trained to predict the risk o...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Karger Publishers
2024-03-01
|
| Series: | Kidney Diseases |
| Online Access: | https://beta.karger.com/Article/FullText/538510 |
| _version_ | 1797201811217580032 |
|---|---|
| author | Yating Wang Yu Shi Tangli Xiao Xianjin Bi Qingyu Huo Shaobo Wang Jiachuan Xiong Jinghong Zhao |
| author_facet | Yating Wang Yu Shi Tangli Xiao Xianjin Bi Qingyu Huo Shaobo Wang Jiachuan Xiong Jinghong Zhao |
| author_sort | Yating Wang |
| collection | DOAJ |
| description | Background: This study aimed to develop and validate a machine learning (ML) model based on serum Klotho for predicting end-stage kidney disease (ESKD) and cardiovascular disease (CVD) in patients with chronic kidney disease (CKD).
Methods: Five different ML models were trained to predict the risk of ESKD and CVD at three different time points (3, 5, and 8-year) using a cohort of 400 non-dialysis CKD patients. The dataset was divided into a training set (70%) and an internal validation set (30%). These models were informed by data comprising 47 clinical features, including serum Klotho. The best-performing model was selected and used to identify risk factors for each outcome. Model performance was assessed using various metrics.
Results: The findings showed that the Lasso regression model had the highest accuracy (C-index=0.71) in predicting ESKD. The features mainly included in this model were estimated glomerular filtration rate (eGFR), 24-hour urinary microalbumin, serum albumin, phosphate, parathyroid hormone, and serum Klotho, which achieved the highest area under the curve (AUC) of 0.930 (95% CI: 0.897-0.962). In addition, for the CVD risk prediction, the Random Survival Forest (RSF) model with the highest accuracy (C-index=0.66) was selected and achieved the highest AUC of 0.782 (95% CI: 0.633-0.930). The features mainly included in this model were age, history of primary hypertension, calcium, tumor necrosis factor-alpha, and serum Klotho.
Conclusion: We successfully developed and validated Klotho-based ML risk prediction models for CVD and ESKD in CKD patients with good performance, indicating their high clinical utility. |
| first_indexed | 2024-04-24T07:53:28Z |
| format | Article |
| id | doaj.art-934f529a88cf409ca3682e2957302dcd |
| institution | Directory Open Access Journal |
| issn | 2296-9357 |
| language | English |
| last_indexed | 2024-04-24T07:53:28Z |
| publishDate | 2024-03-01 |
| publisher | Karger Publishers |
| record_format | Article |
| series | Kidney Diseases |
| spelling | doaj.art-934f529a88cf409ca3682e2957302dcd2024-04-18T07:17:49ZengKarger PublishersKidney Diseases2296-93572024-03-011110.1159/000538510538510A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney DiseaseYating WangYu ShiTangli XiaoXianjin BiQingyu HuoShaobo WangJiachuan XiongJinghong ZhaoBackground: This study aimed to develop and validate a machine learning (ML) model based on serum Klotho for predicting end-stage kidney disease (ESKD) and cardiovascular disease (CVD) in patients with chronic kidney disease (CKD). Methods: Five different ML models were trained to predict the risk of ESKD and CVD at three different time points (3, 5, and 8-year) using a cohort of 400 non-dialysis CKD patients. The dataset was divided into a training set (70%) and an internal validation set (30%). These models were informed by data comprising 47 clinical features, including serum Klotho. The best-performing model was selected and used to identify risk factors for each outcome. Model performance was assessed using various metrics. Results: The findings showed that the Lasso regression model had the highest accuracy (C-index=0.71) in predicting ESKD. The features mainly included in this model were estimated glomerular filtration rate (eGFR), 24-hour urinary microalbumin, serum albumin, phosphate, parathyroid hormone, and serum Klotho, which achieved the highest area under the curve (AUC) of 0.930 (95% CI: 0.897-0.962). In addition, for the CVD risk prediction, the Random Survival Forest (RSF) model with the highest accuracy (C-index=0.66) was selected and achieved the highest AUC of 0.782 (95% CI: 0.633-0.930). The features mainly included in this model were age, history of primary hypertension, calcium, tumor necrosis factor-alpha, and serum Klotho. Conclusion: We successfully developed and validated Klotho-based ML risk prediction models for CVD and ESKD in CKD patients with good performance, indicating their high clinical utility.https://beta.karger.com/Article/FullText/538510 |
| spellingShingle | Yating Wang Yu Shi Tangli Xiao Xianjin Bi Qingyu Huo Shaobo Wang Jiachuan Xiong Jinghong Zhao A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease Kidney Diseases |
| title | A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease |
| title_full | A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease |
| title_fullStr | A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease |
| title_full_unstemmed | A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease |
| title_short | A Klotho-based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease |
| title_sort | klotho based machine learning model for prediction of both kidney and cardiovascular outcomes in chronic kidney disease |
| url | https://beta.karger.com/Article/FullText/538510 |
| work_keys_str_mv | AT yatingwang aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT yushi aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT tanglixiao aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT xianjinbi aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT qingyuhuo aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT shaobowang aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT jiachuanxiong aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT jinghongzhao aklothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT yatingwang klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT yushi klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT tanglixiao klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT xianjinbi klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT qingyuhuo klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT shaobowang klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT jiachuanxiong klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease AT jinghongzhao klothobasedmachinelearningmodelforpredictionofbothkidneyandcardiovascularoutcomesinchronickidneydisease |