Application of explainable ensemble artificial intelligence model to categorization of hemodialysis-patient and treatment using nationwide-real-world data in Japan.

Published on May 29, 2020in PLOS ONE2.74
· DOI :10.1371/JOURNAL.PONE.0233491
Eiichiro Kanda16
Estimated H-index: 16
(Kawasaki Medical School),
Bogdan I. Epureanu25
Estimated H-index: 25
(UM: University of Michigan)
+ 6 AuthorsKosaku Nitta61
Estimated H-index: 61
BACKGROUND Although dialysis patients are at a high risk of death, it is difficult for medical practitioners to simultaneously evaluate many inter-related risk factors. In this study, we evaluated the characteristics of hemodialysis patients using machine learning model, and its usefulness for screening hemodialysis patients at a high risk of one-year death using the nation-wide database of the Japanese Society for Dialysis Therapy. MATERIALS AND METHODS The patients were separated into two datasets (n = 39,930, 39,930, respectively). We categorized hemodialysis patients in Japan into new clusters generated by the K-means clustering method using the development dataset. The association between a cluster and the risk of death was evaluated using multivariate Cox proportional hazards models. Then, we developed an ensemble model composed of the clusters and support vector machine models in the model development phase, and compared the accuracy of the prediction of mortality between the machine learning models in the model validation phase. RESULTS Average age of the subjects was 65.7±12.2 years; 32.7% had diabetes mellitus. The five clusters clearly distinguished the groups on the basis of their characteristics: Cluster 1, young male, and chronic glomerulonephritis; Cluster 2, female, and chronic glomerulonephritis; Cluster 3, diabetes mellitus; Cluster 4, elderly and nephrosclerosis; Cluster 5, elderly and protein energy wasting. These clusters were associated with the risk of death; Cluster 5 compared with Cluster 1, hazard ratio 8.86 (95% CI 7.68, 10.21). The accuracy of the ensemble model for the prediction of 1-year death was 0.948 and higher than those of logistic regression model (0.938), support vector machine model (0.937), and deep learning model (0.936). CONCLUSIONS The clusters clearly categorized patient on their characteristics, and reflected their prognosis. Our real-world-data-based machine learning system is applicable to identifying high-risk hemodialysis patients in clinical settings, and has a strong potential to guide treatments and improve their prognosis.
Figures & Tables
📖 Papers frequently viewed together
2 Citations
6 Citations
#1Sabrina Mezzatesta (UNIME: University of Messina)H-Index: 1
#2Claudia Torino (National Research Council)H-Index: 19
Last. Antonio Vilasi (National Research Council)H-Index: 3
view all 5 authors...
Abstract Background and Objective: Patients with End- Stage Kidney Disease (ESKD) have a unique cardiovascular risk. This study aims at predicting, with a certain precision, death and cardiovascular diseases in dialysis patients. Methods: To achieve our aim, machine learning techniques have been used. Two datasets have been taken into consideration: the first is an Italian dataset obtained from the Istituto di Fisiologia Clinica of Consiglio Nazionale delle Ricerche of Reggio Calabria; the secon...
38 CitationsSource
#1Oguz Akbilgic (RMIT: RMIT University)H-Index: 13
#2Yoshitsugu Obi (UCI: University of California, Irvine)H-Index: 27
Last. Csaba P. Kovesdy (UTHSC: University of Tennessee Health Science Center)H-Index: 102
view all 11 authors...
1. Abstract Introduction Given the high mortality rate within the first year of dialysis initiation, an accurate estimation of post-dialysis mortality could help patients and clinicians in decision-making about initiation of dialysis. We aimed to utilize machine learning (ML) by incorporating complex information from electronic health records to predict patients at risk for post-dialysis short term mortality. Methods This study was carried out on a contemporary cohort of 27,615 US veterans with ...
15 CitationsSource
#1Eiichiro Kanda (Kawasaki Medical School)H-Index: 16
#2Akihiko Kato (Hamamatsu University)H-Index: 29
Last. Yoshihiko Kanno (TMU: Tokyo Medical University)H-Index: 34
view all 4 authors...
Background Protein energy wasting (PEW) is a risk factor for death. However, the cutoff vales for PEW are not optimized for early identification of hemodialysis patients with malnutrition. We evaluated the prognosis of Japanese maintenance hemodialysis patients using nutritional indices optimized for them. Materials and methods We analyzed data from a nation-wide prospective cohort study of the Japanese Society for Dialysis Therapy Renal Data Registry to develop and validate a nutritional risk i...
16 CitationsSource
#1Eiichiro Kanda (Kawasaki Medical School)H-Index: 16
#2Yuki TsurutaH-Index: 13
Last. Ikuto MasakaneH-Index: 31
view all 4 authors...
Because hypotension during hemodialysis (HD) makes continuation of HD difficult and is associated with mortality, pressor approaches are necessary for patients with hypotension. However, the relationships between the pressor approaches and the risk of death have not been clarified yet. We analyzed data from a nationwide prospective cohort study of the Japanese Society for Dialysis Therapy Renal Data Registry (n = 29,309). The outcome was all-cause one-year death. The association between the use ...
2 CitationsSource
#1Rajiv SaranH-Index: 61
#2Bruce G. RobinsonH-Index: 86
Last. Vahakn ShahinianH-Index: 15
view all 79 authors...
Author(s): Saran, Rajiv; Robinson, Bruce; Abbott, Kevin C; Agodoa, Lawrence YC; Bragg-Gresham, Jennifer; Balkrishnan, Rajesh; Bhave, Nicole; Dietrich, Xue; Ding, Zhechen; Eggers, Paul W; Gaipov, Abduzhappar; Gillen, Daniel; Gipson, Debbie; Gu, Haoyu; Guro, Paula; Haggerty, Diana; Han, Yun; He, Kevin; Herman, William; Heung, Michael; Hirth, Richard A; Hsiung, Jui-Ting; Hutton, David; Inoue, Aya; Jacobsen, Steven J; Jin, Yan; Kalantar-Zadeh, Kamyar; Kapke, Alissa; Kleine, Carola-Ellen; Kovesdy, Cs...
395 CitationsSource
#1Jianxing He (Guangzhou Medical University)H-Index: 52
#2Sally L. Baxter (UCSD: University of California, San Diego)H-Index: 8
Last. Kang ZhangH-Index: 89
view all 6 authors...
The development of artificial intelligence (AI)-based technologies in medicine is advancing rapidly, but real-world clinical implementation has not yet become a reality. Here we review some of the key practical issues surrounding the implementation of AI into existing clinical workflows, including data sharing and privacy, transparency of algorithms, data standardization, and interoperability across multiple platforms, and concern for patient safety. We summarize the current regulatory environme...
357 CitationsSource
#1Eric J. Topol (Scripps Health)H-Index: 223
The use of artificial intelligence, and the deep-learning subtype in particular, has been enabled by the use of labeled big data, along with markedly enhanced computing power and cloud storage, across all sectors. In medicine, this is beginning to have an impact at three levels: for clinicians, predominantly via rapid, accurate image interpretation; for health systems, by improving workflow and the potential for reducing medical errors; and for patients, by enabling them to process their own dat...
1,209 CitationsSource
#1Shi Yan (NUS: National University of Singapore)H-Index: 4
#2Yu Heng Kwan (NUS: National University of Singapore)H-Index: 13
Last. Lian Leng Low (SGH: Singapore General Hospital)H-Index: 16
view all 5 authors...
Data-driven population segmentation analysis utilizes data analytics to divide a heterogeneous population into parsimonious and relatively homogenous groups with similar healthcare characteristics. It is a promising patient-centric analysis that enables effective integrated healthcare interventions specific for each segment. Although widely applied, there is no systematic review on the clinical application of data-driven population segmentation analysis. We carried out a systematic literature se...
23 CitationsSource
#1Mario Cozzolino (University of Milan)H-Index: 47
#2Michela Mangano (University of Milan)H-Index: 5
Last. Andrea Galassi (University of Milan)H-Index: 18
view all 6 authors...
: Cardiovascular disease (CVD) is a highly common complication and the first cause of death in patients with end-stage renal disease (ESRD) on haemodialysis (HD). In this population, mortality due to CVD is 20 times higher than in the general population and the majority of maintenance HD patients have CVD. This is likely due to ventricular hypertrophy as well as non-traditional risk factors, such as chronic volume overload, anaemia, inflammation, oxidative stress, chronic kidney disease-mineral ...
117 CitationsSource
#1Lian Leng Low (SingHealth)H-Index: 16
#2Shi Yan (NUS: National University of Singapore)H-Index: 4
Last. Julian ThumbooH-Index: 53
view all 5 authors...
Background Segmentation of heterogeneous patient populations into parsimonious and relatively homogenous groups with similar healthcare needs can facilitate healthcare resource planning and development of effective integrated healthcare interventions for each segment. We aimed to apply a data-driven, healthcare utilization-based clustering analysis to segment a regional health system patient population and validate its discriminative ability on 4-year longitudinal healthcare utilization and mort...
19 CitationsSource
Cited By3
#1Covadonga Diez-Sanmartín (Complutense University of Madrid)H-Index: 1
#2Antonio Sarasa-Cabezuelo (Complutense University of Madrid)H-Index: 8
Last. Amado Andrés Belmonte (Complutense University of Madrid)
view all 3 authors...
Abstract In the field of medicine, decision-making has traditionally been carried out based on the best available scientific information and the experience of specialists using data found in analog formats such as radiographies, medical reports, and handwritten notes, among others. In this sense, the Big Data phenomenon is changing the world of medicine since the technologies that have been developed have made available to researchers and clinicians enormous amounts of data in digital formats th...
#1Luigi MancoH-Index: 1
#2N. MaffeiH-Index: 5
Last. Lidia StrigariH-Index: 32
view all 6 authors...
The manuscript aims at providing an overview of the published algorithms/automation tool for artificial intelligence applied to imaging for Healthcare. A PubMed search was performed using the query string to identify the proposed approaches (algorithms/automation tools) for artificial intelligence (machine and deep learning) in a 5-year period. The distribution of manuscript in the various disciplines and the investigated image types according to the AI approaches are presented. The limitation a...
3 CitationsSource
#2Carlo BarbieriH-Index: 11
view all 4 authors...