Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City

Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City View ORCID Profile Suraj K Jaladanki , View ORCID Profile Akhil Vaid , View ORCID Profile Ashwin S Sawant , View ORCID Profile Jie Xu , Kush Shah , View ORCID Profile Sergio Dellepiane , View ORCID Profile Ishan Paranjpe , View ORCID Profile Lili Chan , View ORCID Profile Patricia Kovatch , View ORCID Profile Alexander W Charney , View ORCID Profile Fei Wang , View ORCID Profile Benjamin S Glicksberg , View ORCID Profile Karandeep Singh , View ORCID Profile Girish N Nadkarni doi: https://doi.org/10.1101/2021.07.25.21261105 Suraj K Jaladanki 1 The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai , New York, New York, USA 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA BS Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Suraj K Jaladanki Akhil Vaid 1 The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai , New York, New York, USA 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA MD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Akhil Vaid Ashwin S Sawant 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA 3 The Division of Data-Driven and Digital Medicine (D3M), Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA MD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ashwin S Sawant Jie Xu 4 Department of Population Health Sciences, Weill Cornell Medicine , New York, New York, USA PhD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jie Xu Kush Shah 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA BA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sergio Dellepiane 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA 3 The Division of Data-Driven and Digital Medicine (D3M), Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA MD, PhD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Sergio Dellepiane Ishan Paranjpe 1 The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai , New York, New York, USA 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA BS Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ishan Paranjpe Lili Chan 3 The Division of Data-Driven and Digital Medicine (D3M), Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA 5 The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA 6 BioMe Phenomics Center, Icahn School of Medicine at Mount Sinai , New York, New York, USA 7 Division of Nephrology, Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA MD, MS Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Lili Chan Patricia Kovatch 8 Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai , New York, New York, USA BS Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Patricia Kovatch Alexander W Charney 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA 8 Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai , New York, New York, USA 9 The Pamela Sklar Division of Psychiatric Genomics, Icahn School of Medicine at Mount Sinai , New York, New York, USA 10 Department of Psychiatry, Icahn School of Medicine at Mount Sinai , New York, New York, USA MD, PhD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Alexander W Charney Fei Wang 4 Department of Population Health Sciences, Weill Cornell Medicine , New York, New York, USA PhD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Fei Wang Benjamin S Glicksberg 1 The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai , New York, New York, USA 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA 8 Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai , New York, New York, USA PhD Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Benjamin S Glicksberg Karandeep Singh 11 Department of Learning Health Sciences, University of Michigan Medical School , Ann Arbor, Michigan, USA MD, MS Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Karandeep Singh Girish N Nadkarni 1 The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai , New York, New York, USA 2 The Mount Sinai Clinical Intelligence Center (MSCIC) , New York, New York, USA 3 The Division of Data-Driven and Digital Medicine (D3M), Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA 5 The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA 7 Division of Nephrology, Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, New York, USA MD, MPH Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Girish N Nadkarni For correspondence: girish.nadkarni{at}mountsinai.org Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF ABSTRACT Federated learning is a technique for training predictive models without sharing patient-level data, thus maintaining data security while allowing inter-institutional collaboration. We used federated learning to predict acute kidney injury within three and seven days of admission, using demographics, comorbidities, vital signs, and laboratory values, in 4029 adults hospitalized with COVID-19 at five sociodemographically diverse New York City hospitals, between March-October 2020. Prediction performance of federated models was generally higher than single-hospital models and was comparable to pooled-data models. In the first use-case in kidney disease, federated learning improved prediction of a common complication of COVID-19, while preserving data privacy. Introduction Machine Learning (ML) models have been developed to aid in risk-stratification in patients with Coronavirus Disease-19 (COVID-19). 1 - 3 Due to concerns about data privacy and ownership, most of these models were trained on data from single institutions, which limits their generalizability. 4 , 5 However, it is critical that these models be trained on large, diverse datasets, reflecting the variability seen in clinical practice. Federated learning (FL), a technique that allows model training to occur without sharing confidential patient data, has gained significant interest during the pandemic. FL avoids the need to collect and store private data at a centralized location by allowing institutions to download a preliminary ML model, refine it locally, and then upload updated model parameters to an aggregator. 5 - 7 FL models have been utilized to identify COVID-19 through computed tomography scans and to predict clinical outcomes in COVID-19 patients — published reports indicate that FL models not only outperform those trained on single-institution data but also approach the performance of traditional ML models trained on data pooled from multiple institutions. 6 , 8 - 10 COVID-19 has diverse clinical manifestations, and a common complication in hospitalized COVID-19 patients is acute kidney injury (AKI). 11 - 14 Studies have reported AKI prevalence up to 46%, and mortality rates in AKI cohorts vary from 35%–71%. 15 , 16 Clinical models to pre-emptively identify patients with COVID-19 at increased risk for developing AKI can be valuable during a pandemic when resources may be constrained. 17 Using COVID-19 associated AKI as a use case, we aimed to evaluate how FL trained models compare to models trained at individual institutions and pooled multi-institutional data. We developed federated models to predict AKI within three and seven days of admission using electronic health records (EHR) in patients hospitalized with COVID-19 and compared their performance to local and pooled models. Short Methods The study population consisted of adult patients admitted to one of five Mount Sinai Health System hospitals in New York City with laboratory confirmed SARS-CoV-2 infection within 48 hours of admission between March 1 st and October 18 th , 2020. We excluded patients with a history of transplants, a diagnosis of kidney failure or admissions less than 48 hours. We defined AKI according to 2012 Kidney Disease Improving Global Outcomes (KDIGO) guidelines. 18 The primary outcomes of interest were incident AKI within three (AKI 3 ) and seven (AKI 7 ) days of admission. We included demographics, comorbidities, vital signs, laboratory values, and clinical outcomes extracted from an EHR database. For each parameter, we used the first available values obtained within 48 hours after admission. We used two classifiers: multilayer perceptron (MLP) to represent deep learning and Least Absolute Shrinkage and Selection Operator (LASSO) as a regression method. We used three training strategies for each classifier: local, federated, and pooled ( Figure 1A, 1B ). Local models for each hospital were trained and tested using only data from that hospital. Federated models used a FL framework to share model parameters with a central aggregator. Finally, pooled models combined data from all hospitals for model development and represented an optimal scenario. Model performance was assessed by average areas under the receiver operating characteristic curve (AUROCs) across 100 independent experiments with 70%/30% training/testing splits. SHapley Additive exPlanations (SHAP) scores were used to evaluate the importance of features contributing to model predictions. 19 Detailed inclusion and exclusion criteria, data processing, classifier parameters, and software are described in Detailed Methods (Supplementary Material). We used the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) guidelines and shared our code under the GNU General Public License v3 to enhance replicability (Supplementary Material). Download figure Open in new tab Figure 1. Study Design and Model Performance. (A) Overview of local and pooled models. Local models only utilize data from the site itself while pooled models incorporate data from all sites. Both local and pooled MLP, LR, and LASSO models were utilized. Based on https://medinform.jmir.org/2021/1/e24207/ (B) Overview of federated model. Parameters from a central aggregator are shared with each site, and sites do not have direct access to clinical data from others. After models are trained locally at a site, parameters are sent back to the central aggregator to update federated model parameters. Federated LASSO and MLP models were utilized. Based on https://medinform.jmir.org/2021/1/e24207/ (C) Performance of all models averaged across all sites as measured by area under the receiver-operating characteristic (AUROC) after 70-30 train-test split over 100 experiments with 95% confidence intervals for predicting AKI within three (top) and seven (bottom) days of admission. Results Cohort Characteristics Among 4029 patients, the prevalence of AKI 3 ranged from 23% to 37% across five hospitals, and 27% to 44% for AKI 7 . There were significant differences in demographics, comorbidities, laboratory measurements, outcome prevalence, and sample sizes between hospitals (Supplementary Table S3). For example, Mount Sinai West (MSW) had the largest class imbalance and smallest sample size of all hospitals. Learning Framework Comparisons We evaluated performance of three model strategies (local, pooled, and federated) at each hospital for predicting AKI 3 and AKI 7. Performance averaged across the five sites is shown in Figure 1C . All models had AUROCs in the range of 0.70–0.90 for predicting AKI 3 and AKI 7 . Pooled models consistently outperformed both local and federated models for AKI 3 and AKI 7 prediction. LASSO federated also outperformed LASSO local at all hospitals except one for both AKI 3 and AKI 7 prediction (Supplementary Table S5). MLP federated outperformed MLP local only at MSW for both AKI 3 and AKI 7 prediction (Supplementary Table S5). Additional performance metrics for all models are available in the supplementary material (Supplementary Table S4). MSW Performance and Feature Importance The largest improvements in performance between local and federated models were seen for MSW ( Figure 2A ), with an increase of at least 3% (0.03) in AUROC for all three classifiers for AKI 3 and AKI 7 prediction. SHAP plots for LASSO local and LASSO federated models for AKI 3 prediction at MSW demonstrate differences in feature importance ( Figure 2B ). For LASSO local , history of stroke, black race, and Hispanic/latino ethnicity were the three most important features for predicting AKI 3 . However, for the corresponding LASSO federated model, the three features with highest importance were blood urea nitrogen, age, and albumin. Download figure Open in new tab Figure 2. Model Performance and Feature Importance at MSW (A) Performance of all models at Mount Sinai West (MSW) (n=474) as measured by area under the receiver-operating characteristic (AUROC) after 70-30 train-test split over 100 experiments with 95% confidence intervals for predicting AKI within three (left) and seven (right) days of admission. (B) SHAP values were calculated for LASSO local (left) and LASSO federated (right) for predicting AKI within three days of admission at MSW and illustrated in summary plots where features are listed in decreasing order of importance. Discussion We trained models using federated learning to predict AKI at three and seven days in hospitalized patients with COVID-19. To our knowledge, this is the first use case of federated learning in kidney disease and to predict AKI over multiple time horizons. We found that models trained with FL outperformed those trained on locally available data, which was most apparent at the hospital with the smallest dataset. All models had AUROCs of at least 0.70, suggesting that early vital signs and laboratory measurements can predict kidney complications during the first week in patients hospitalized with COVID-19. A similar approach has been used to predict mortality in hospitalized patients with COVID-19, highlighting the utility of peri-admission measurements in the prediction of clinical outcomes in this cohort. 10 Model performance was consistently lower for AKI 7 than AKI 3 , likely due to the longer prediction interval in the former case. We found that LASSO federated models outperformed their respective LASSO local models at most hospitals. At MSW, the hospital with the smallest sample size and largest class imbalance, SHAP plots indicate that the LASSO federated model attached greater importance to features that have more biologically plausible links to renal outcomes than the corresponding LASSO local model, which was strongly influenced by race and a history of stroke. MSW was also the only site where MLP federated outperformed MLP local , indicating that the greatest benefits from using FL to train deep learning models may accrue to sites with small amounts of local data available for training. Models developed during this study may be implemented in clinical practice to predict onset of AKI in COVID-19 patients several days before it occurs, allowing clinicians to triage patients for better monitoring and use of intravenous fluids. Early prediction of AKI also allows for proactive allocation of resources, because published data indicate that these patients are hospitalized longer, are more likely to need intensive care, and more likely to require inpatient dialysis. 14 Our study has several limitations. First, our data was limited to a single hospital system within New York City, which may limit generalizability to other geographical regions. Similarly, as the standard of care for hospitalized patients with COVID-19 has evolved after our data was collected, model performance should be validated with newer cohorts. Our models did not use inputs such as imaging and echocardiograms, and their exclusion may have hindered performance. We deliberately limited hyperparameter tuning of models to simulate a scenario where a federated model has to be urgently deployed at resource-constrained hospitals, such as during the COVID-19 pandemic. Allowing further optimization of the MLP structure, and techniques such as transfer learning may improve MLP federated model performance. 20 We examined only two classifiers; other approaches such as random forests or support vector machines may have yielded better results. Lastly, performance with the FL approach was generally statistically inferior to pooled dataset performance, and it remains to be explored if other approaches exist which can help bridge this gap while retaining the privacy advantages of FL. In summary, we demonstrate the utility of FL to improve prediction of key outcomes while maintaining privacy and confidentiality. We hope this will encourage the development of generalizable clinical models which would otherwise be hindered by inability to share patient-level data across institutional boundaries. Data Availability This article is written following the TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) guidelines, which are further elaborated in Supplementary Table 3. Furthermore, we release all code used for building the classifier under the GPLv3 license in a public GitHub repository. https://github.com/HPIMS/COVID_Federated_AKI Footnotes Sources of support. This study was supported by the National Center for Advancing Translational Sciences, National Institutes of Health (U54 TR001433-05). GNN is also supported by R01DK127139, R01HL155915, and R56DK126930. FW would like to acknowledge the support from NSF 1750326 and NIH RF1AG072449. References 1. ↵ Mei X , Lee H-C , Diao K-y , et al. Artificial intelligence–enabled rapid diagnosis of patients with COVID-19 . Nature Medicine 2020 ; 26 : 1224 – 1228 . OpenUrl 2. Vaid A , Somani S , Russak AJ , et al. Machine Learning to Predict Mortality and Critical Events in a Cohort of Patients With COVID-19 in New York City: Model Development and Validation . J Med Internet Res 2020 ; 22 : e24018 . OpenUrl CrossRef 3. ↵ Wynants L , Van Calster B , Collins GS , et al. Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal . BMJ 2020 ; 369 : m1328 . OpenUrl Abstract / FREE Full Text 4. ↵ Cahan EM , Hernandez-Boussard T , Thadaney-Israni S , et al. Putting the data before the algorithm in big data addressing personalized healthcare . npj Digital Medicine 2019 ; 2 : 78 . OpenUrl 5. ↵ Song X , Yu ASL , Kellum JA , et al. Cross-site transportability of an explainable artificial intelligence model for acute kidney injury prediction . Nature Communications 2020 ; 11 : 5668 . OpenUrl 6. ↵ Flores M , Dayan I , Roth H , et al. Federated Learning used for predicting outcomes in SARS-COV-2 patients . Res Sq 2021 . 7. ↵ Xu J , Glicksberg BS , Su C , et al. Federated learning for healthcare informatics . arXiv preprint 2020 . 8. ↵ Kumar R , Khan AA , Zhang S , et al. Blockchain-Federated-Learning and Deep Learning Models for COVID-19 detection using CT Imaging . arXiv preprint 2020 . 9. Xu Y , Ma L , Yang F , et al. A collaborative online AI engine for CT-based COVID-19 diagnosis . medRxiv 2020 : 2020.2005.2010.20096073. 10. ↵ Vaid A , Jaladanki SK , Xu J , et al. Federated Learning of Electronic Health Records to Improve Mortality Prediction in Hospitalized Patients With COVID-19: Machine Learning Approach . JMIR Med Inform 2021 ; 9 : e24207 . OpenUrl 11. ↵ Wu Z , McGoogan JM . Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72□314 Cases From the Chinese Center for Disease Control and Prevention . JAMA 2020 ; 323 : 1239 – 1242 . OpenUrl CrossRef PubMed 12. Chung M , Bernheim A , Mei X , et al. CT Imaging Features of 2019 Novel Coronavirus (2019-nCoV) . Radiology 2020 ; 295 : 202 – 207 . OpenUrl CrossRef PubMed 13. Lala A , Johnson KW , Januzzi JL , et al. Prevalence and Impact of Myocardial Injury in Patients Hospitalized With COVID-19 Infection . J Am Coll Cardiol 2020 ; 76 : 533 – 546 . OpenUrl FREE Full Text 14. ↵ Chan L , Chaudhary K , Saha A , et al. AKI in Hospitalized Patients with COVID-19 . Journal of the American Society of Nephrology 2021 ; 32 : 151 – 160 . OpenUrl Abstract / FREE Full Text 15. ↵ Zahid U , Ramachandran P , Spitalewitz S , et al. Acute Kidney Injury in COVID-19 Patients: An Inner City Hospital Experience and Policy Implications . American Journal of Nephrology 2020 ; 51 : 786 – 796 . OpenUrl 16. ↵ Hirsch JS , Ng JH , Ross DW , et al. Acute kidney injury in patients hospitalized with COVID-19 . Kidney International 2020 ; 98 : 209 – 218 . OpenUrl CrossRef PubMed 17. ↵ Goldfarb DS , Benstein JA , Zhdanova O , et al. Impending Shortages of Kidney Replacement Therapy for COVID-19 Patients . Clinical Journal of the American Society of Nephrology 2020 ; 15 : 880 – 882 . OpenUrl FREE Full Text 18. ↵ Kidney Disease : Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guidelines for Acute Kidney Injury . Kidney inter., Suppl . 2012 ; 2 : 1 – 138 . OpenUrl CrossRef 19. ↵ Lundberg SM , Erion G , Chen H , et al. From Local Explanations to Global Understanding with Explainable AI for Trees . Nature machine intelligence 2020 ; 2 : 56 – 67 . OpenUrl 20. ↵ Ching T , Himmelstein DS , Beaulieu-Jones BK , et al. Opportunities and obstacles for deep learning in biology and medicine . J R Soc Interface 2018 ; 15 : 20170387 . OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted July 28, 2021. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City Suraj K Jaladanki , Akhil Vaid , Ashwin S Sawant , Jie Xu , Kush Shah , Sergio Dellepiane , Ishan Paranjpe , Lili Chan , Patricia Kovatch , Alexander W Charney , Fei Wang , Benjamin S Glicksberg , Karandeep Singh , Girish N Nadkarni medRxiv 2021.07.25.21261105; doi: https://doi.org/10.1101/2021.07.25.21261105 Share This Article: Copy Citation Tools Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City Suraj K Jaladanki , Akhil Vaid , Ashwin S Sawant , Jie Xu , Kush Shah , Sergio Dellepiane , Ishan Paranjpe , Lili Chan , Patricia Kovatch , Alexander W Charney , Fei Wang , Benjamin S Glicksberg , Karandeep Singh , Girish N Nadkarni medRxiv 2021.07.25.21261105; doi: https://doi.org/10.1101/2021.07.25.21261105 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Nephrology Subject Areas All Articles Addiction Medicine (578) Allergy and Immunology (869) Anesthesia (306) Cardiovascular Medicine (4491) Dentistry and Oral Medicine (450) Dermatology (385) Emergency Medicine (616) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1530) Epidemiology (15300) Forensic Medicine (31) Gastroenterology (1135) Genetic and Genomic Medicine (6657) Geriatric Medicine (673) Health Economics (1006) Health Informatics (4621) Health Policy (1379) Health Systems and Quality Improvement (1631) Hematology (545) HIV/AIDS (1278) Infectious Diseases (except HIV/AIDS) (15971) Intensive Care and Critical Care Medicine (1113) Medical Education (626) Medical Ethics (147) Nephrology (678) Neurology (6711) Nursing (346) Nutrition (1006) Obstetrics and Gynecology (1154) Occupational and Environmental Health (962) Oncology (3372) Ophthalmology (989) Orthopedics (370) Otolaryngology (422) Pain Medicine (438) Palliative Medicine (131) Pathology (671) Pediatrics (1705) Pharmacology and Therapeutics (700) Primary Care Research (718) Psychiatry and Clinical Psychology (5504) Public and Global Health (9297) Radiology and Imaging (2232) Rehabilitation Medicine and Physical Therapy (1375) Respiratory Medicine (1202) Rheumatology (598) Sexual and Reproductive Health (723) Sports Medicine (536) Surgery (724) Toxicology (100) Transplantation (291) Urology (268) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a08a8b33b8810f09',t:'MTc4MDk0OTQ2Ng=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();