Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering

Charat Thongprayoon; Pradeep Vaitla; Voravech Nissaisorakarn; Michael A. Mao; Jose L.Zabala Genovez; Andrea G. Kattah; Pattharawin Pattharanitima; Saraschandra Vallabhajosyula; Mira T. Keddis; Fawad Qureshi; John J. Dillon; Vesna D. Garovic; Kianoush B. Kashani; Wisit Cheungpasitporn

doi:10.3390/medsci9040060

Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering

Charat Thongprayoon, Pradeep Vaitla, Voravech Nissaisorakarn, Michael A. Mao, Jose L.Zabala Genovez, Andrea G. Kattah, Pattharawin Pattharanitima, Saraschandra Vallabhajosyula, Mira T. Keddis, Fawad Qureshi, John J. Dillon, Vesna D. Garovic, Kianoush B. Kashani, Wisit Cheungpasitporn

Research output: Contribution to journal › Article › peer-review

Abstract

BACKGROUND: We aimed to cluster patients with acute kidney injury at hospital admission into clinically distinct subtypes using an unsupervised machine learning approach and assess the mortality risk among the distinct clusters. METHODS: We performed consensus clustering analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 4289 hospitalized adult patients with acute kidney injury at admission. The standardized difference of each variable was calculated to identify each cluster's key features. We assessed the association of each acute kidney injury cluster with hospital and one-year mortality. RESULTS: Consensus clustering analysis identified four distinct clusters. There were 1201 (28%) patients in cluster 1, 1396 (33%) patients in cluster 2, 1191 (28%) patients in cluster 3, and 501 (12%) patients in cluster 4. Cluster 1 patients were the youngest and had the least comorbidities. Cluster 2 and cluster 3 patients were older and had lower baseline kidney function. Cluster 2 patients had lower serum bicarbonate, strong ion difference, and hemoglobin, but higher serum chloride, whereas cluster 3 patients had lower serum chloride but higher serum bicarbonate and strong ion difference. Cluster 4 patients were younger and more likely to be admitted for genitourinary disease and infectious disease but less likely to be admitted for cardiovascular disease. Cluster 4 patients also had more severe acute kidney injury, lower serum sodium, serum chloride, and serum bicarbonate, but higher serum potassium and anion gap. Cluster 2, 3, and 4 patients had significantly higher hospital and one-year mortality than cluster 1 patients (p < 0.001). CONCLUSION: Our study demonstrated using machine learning consensus clustering analysis to characterize a heterogeneous cohort of patients with acute kidney injury on hospital admission into four clinically distinct clusters with different associated mortality risks.

Original language	English (US)
Journal	Medical sciences (Basel, Switzerland)
Volume	9
Issue number	4
DOIs	https://doi.org/10.3390/medsci9040060
State	Published - Sep 24 2021

Keywords

AKI
acute kidney injury
artificial intelligence
clustering
hospitalization
machine learning
mortality
nephrology

ASJC Scopus subject areas

General Medicine

Access to Document

10.3390/medsci9040060

Cite this

Thongprayoon, C., Vaitla, P., Nissaisorakarn, V., Mao, M. A., Genovez, J. L. Z., Kattah, A. G., Pattharanitima, P., Vallabhajosyula, S., Keddis, M. T., Qureshi, F., Dillon, J. J., Garovic, V. D., Kashani, K. B., & Cheungpasitporn, W. (2021). Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering. Medical sciences (Basel, Switzerland), 9(4). https://doi.org/10.3390/medsci9040060

Thongprayoon, C, Vaitla, P, Nissaisorakarn, V, Mao, MA, Genovez, JLZ, Kattah, AG, Pattharanitima, P, Vallabhajosyula, S, Keddis, MT, Qureshi, F, Dillon, JJ, Garovic, VD, Kashani, KB & Cheungpasitporn, W 2021, 'Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering', Medical sciences (Basel, Switzerland), vol. 9, no. 4. https://doi.org/10.3390/medsci9040060

@article{a8db05bb15b24982b5fd7786ddf29a27,

title = "Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering",

abstract = "BACKGROUND: We aimed to cluster patients with acute kidney injury at hospital admission into clinically distinct subtypes using an unsupervised machine learning approach and assess the mortality risk among the distinct clusters. METHODS: We performed consensus clustering analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 4289 hospitalized adult patients with acute kidney injury at admission. The standardized difference of each variable was calculated to identify each cluster's key features. We assessed the association of each acute kidney injury cluster with hospital and one-year mortality. RESULTS: Consensus clustering analysis identified four distinct clusters. There were 1201 (28%) patients in cluster 1, 1396 (33%) patients in cluster 2, 1191 (28%) patients in cluster 3, and 501 (12%) patients in cluster 4. Cluster 1 patients were the youngest and had the least comorbidities. Cluster 2 and cluster 3 patients were older and had lower baseline kidney function. Cluster 2 patients had lower serum bicarbonate, strong ion difference, and hemoglobin, but higher serum chloride, whereas cluster 3 patients had lower serum chloride but higher serum bicarbonate and strong ion difference. Cluster 4 patients were younger and more likely to be admitted for genitourinary disease and infectious disease but less likely to be admitted for cardiovascular disease. Cluster 4 patients also had more severe acute kidney injury, lower serum sodium, serum chloride, and serum bicarbonate, but higher serum potassium and anion gap. Cluster 2, 3, and 4 patients had significantly higher hospital and one-year mortality than cluster 1 patients (p < 0.001). CONCLUSION: Our study demonstrated using machine learning consensus clustering analysis to characterize a heterogeneous cohort of patients with acute kidney injury on hospital admission into four clinically distinct clusters with different associated mortality risks.",

keywords = "AKI, acute kidney injury, artificial intelligence, clustering, hospitalization, machine learning, mortality, nephrology",

author = "Charat Thongprayoon and Pradeep Vaitla and Voravech Nissaisorakarn and Mao, {Michael A.} and Genovez, {Jose L.Zabala} and Kattah, {Andrea G.} and Pattharawin Pattharanitima and Saraschandra Vallabhajosyula and Keddis, {Mira T.} and Fawad Qureshi and Dillon, {John J.} and Garovic, {Vesna D.} and Kashani, {Kianoush B.} and Wisit Cheungpasitporn",

year = "2021",

month = sep,

day = "24",

doi = "10.3390/medsci9040060",

language = "English (US)",

volume = "9",

journal = "Medical sciences (Basel, Switzerland)",

issn = "2076-3271",

publisher = "NLM (Medline)",

number = "4",

}

TY - JOUR

T1 - Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering

AU - Thongprayoon, Charat

AU - Vaitla, Pradeep

AU - Nissaisorakarn, Voravech

AU - Mao, Michael A.

AU - Genovez, Jose L.Zabala

AU - Kattah, Andrea G.

AU - Pattharanitima, Pattharawin

AU - Vallabhajosyula, Saraschandra

AU - Keddis, Mira T.

AU - Qureshi, Fawad

AU - Dillon, John J.

AU - Garovic, Vesna D.

AU - Kashani, Kianoush B.

AU - Cheungpasitporn, Wisit

PY - 2021/9/24

Y1 - 2021/9/24

N2 - BACKGROUND: We aimed to cluster patients with acute kidney injury at hospital admission into clinically distinct subtypes using an unsupervised machine learning approach and assess the mortality risk among the distinct clusters. METHODS: We performed consensus clustering analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 4289 hospitalized adult patients with acute kidney injury at admission. The standardized difference of each variable was calculated to identify each cluster's key features. We assessed the association of each acute kidney injury cluster with hospital and one-year mortality. RESULTS: Consensus clustering analysis identified four distinct clusters. There were 1201 (28%) patients in cluster 1, 1396 (33%) patients in cluster 2, 1191 (28%) patients in cluster 3, and 501 (12%) patients in cluster 4. Cluster 1 patients were the youngest and had the least comorbidities. Cluster 2 and cluster 3 patients were older and had lower baseline kidney function. Cluster 2 patients had lower serum bicarbonate, strong ion difference, and hemoglobin, but higher serum chloride, whereas cluster 3 patients had lower serum chloride but higher serum bicarbonate and strong ion difference. Cluster 4 patients were younger and more likely to be admitted for genitourinary disease and infectious disease but less likely to be admitted for cardiovascular disease. Cluster 4 patients also had more severe acute kidney injury, lower serum sodium, serum chloride, and serum bicarbonate, but higher serum potassium and anion gap. Cluster 2, 3, and 4 patients had significantly higher hospital and one-year mortality than cluster 1 patients (p < 0.001). CONCLUSION: Our study demonstrated using machine learning consensus clustering analysis to characterize a heterogeneous cohort of patients with acute kidney injury on hospital admission into four clinically distinct clusters with different associated mortality risks.

AB - BACKGROUND: We aimed to cluster patients with acute kidney injury at hospital admission into clinically distinct subtypes using an unsupervised machine learning approach and assess the mortality risk among the distinct clusters. METHODS: We performed consensus clustering analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 4289 hospitalized adult patients with acute kidney injury at admission. The standardized difference of each variable was calculated to identify each cluster's key features. We assessed the association of each acute kidney injury cluster with hospital and one-year mortality. RESULTS: Consensus clustering analysis identified four distinct clusters. There were 1201 (28%) patients in cluster 1, 1396 (33%) patients in cluster 2, 1191 (28%) patients in cluster 3, and 501 (12%) patients in cluster 4. Cluster 1 patients were the youngest and had the least comorbidities. Cluster 2 and cluster 3 patients were older and had lower baseline kidney function. Cluster 2 patients had lower serum bicarbonate, strong ion difference, and hemoglobin, but higher serum chloride, whereas cluster 3 patients had lower serum chloride but higher serum bicarbonate and strong ion difference. Cluster 4 patients were younger and more likely to be admitted for genitourinary disease and infectious disease but less likely to be admitted for cardiovascular disease. Cluster 4 patients also had more severe acute kidney injury, lower serum sodium, serum chloride, and serum bicarbonate, but higher serum potassium and anion gap. Cluster 2, 3, and 4 patients had significantly higher hospital and one-year mortality than cluster 1 patients (p < 0.001). CONCLUSION: Our study demonstrated using machine learning consensus clustering analysis to characterize a heterogeneous cohort of patients with acute kidney injury on hospital admission into four clinically distinct clusters with different associated mortality risks.

KW - AKI

KW - acute kidney injury

KW - artificial intelligence

KW - clustering

KW - hospitalization

KW - machine learning

KW - mortality

KW - nephrology

UR - http://www.scopus.com/inward/record.url?scp=85115775118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85115775118&partnerID=8YFLogxK

U2 - 10.3390/medsci9040060

DO - 10.3390/medsci9040060

M3 - Article

C2 - 34698185

AN - SCOPUS:85115775118

SN - 2076-3271

VL - 9

JO - Medical sciences (Basel, Switzerland)

JF - Medical sciences (Basel, Switzerland)

IS - 4

ER -

Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this