1
|
Hung EC, Hodzic E, Tan ZC, Meyer AS. Censored Least Squares for Imputing Missing Values in PARAFAC Tensor Factorization. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.07.05.602272. [PMID: 39026852 PMCID: PMC11257416 DOI: 10.1101/2024.07.05.602272] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 07/20/2024]
Abstract
Tensor factorization is a dimensionality reduction method applied to multidimensional arrays. These methods are useful for identifying patterns within a variety of biomedical datasets due to their ability to preserve the organizational structure of experiments and therefore aid in generating meaningful insights. However, missing data in the datasets being analyzed can impose challenges. Tensor factorization can be performed with some level of missing data and reconstruct a complete tensor. However, while tensor methods may impute these missing values, the choice of fitting algorithm may influence the fidelity of these imputations. Previous approaches, based on alternating least squares with prefilled values or direct optimization, suffer from introduced bias or slow computational performance. In this study, we propose that censored least squares can better handle missing values with data structured in tensor form. We ran censored least squares on four different biological datasets and compared its performance against alternating least squares with prefilled values and direct optimization. We used the error of imputation and the ability to infer masked values to benchmark their missing data performance. Censored least squares appeared best suited for the analysis of high-dimensional biological data by accuracy and convergence metrics across several studies.
Collapse
Affiliation(s)
- Ethan C Hung
- Computational and Systems Biology, University of California, Los Angeles (UCLA), USA
| | | | | | - Aaron S Meyer
- Department of Bioengineering, UCLA, USA
- Jonsson Comprehensive Cancer Center, UCLA, USA
- Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research, UCLA, USA
| |
Collapse
|
2
|
Jiang S, Gai X, Treggiari MM, Stead WW, Zhao Y, Page CD, Zhang AR. Soft phenotyping for sepsis via EHR time-aware soft clustering. J Biomed Inform 2024; 152:104615. [PMID: 38423266 PMCID: PMC11073833 DOI: 10.1016/j.jbi.2024.104615] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2023] [Revised: 01/25/2024] [Accepted: 02/20/2024] [Indexed: 03/02/2024]
Abstract
OBJECTIVE Sepsis is one of the most serious hospital conditions associated with high mortality. Sepsis is the result of a dysregulated immune response to infection that can lead to multiple organ dysfunction and death. Due to the wide variability in the causes of sepsis, clinical presentation, and the recovery trajectories, identifying sepsis sub-phenotypes is crucial to advance our understanding of sepsis characterization, to choose targeted treatments and optimal timing of interventions, and to improve prognostication. Prior studies have described different sub-phenotypes of sepsis using organ-specific characteristics. These studies applied clustering algorithms to electronic health records (EHRs) to identify disease sub-phenotypes. However, prior approaches did not capture temporal information and made uncertain assumptions about the relationships among the sub-phenotypes for clustering procedures. METHODS We developed a time-aware soft clustering algorithm guided by clinical variables to identify sepsis sub-phenotypes using data available in the EHR. RESULTS We identified six novel sepsis hybrid sub-phenotypes and evaluated them for medical plausibility. In addition, we built an early-warning sepsis prediction model using logistic regression. CONCLUSION Our results suggest that these novel sepsis hybrid sub-phenotypes are promising to provide more accurate information on sepsis-related organ dysfunction and sepsis recovery trajectories which can be important to inform management decisions and sepsis prognosis.
Collapse
Affiliation(s)
- Shiyi Jiang
- Department of Electrical & Computer Engineering, Duke University, Durham, 27708, NC, USA
| | - Xin Gai
- Department of Statistical Science, Duke University, Durham, 27708, NC, USA
| | | | - William W Stead
- Department of Biomedical Informatics, Vanderbilt University, Nashville, 37235, TN, USA
| | - Yuankang Zhao
- Department of Biostatistics & Bioinformatics, Duke University, Durham, 27708, NC, USA
| | - C David Page
- Department of Biostatistics & Bioinformatics, Duke University, Durham, 27708, NC, USA
| | - Anru R Zhang
- Department of Biostatistics & Bioinformatics, Duke University, Durham, 27708, NC, USA; Department of Computer Science, Duke University, Durham, 27708, NC, USA.
| |
Collapse
|
3
|
Wang Y, Stroh JN, Hripcsak G, Low Wang CC, Bennett TD, Wrobel J, Der Nigoghossian C, Mueller SW, Claassen J, Albers DJ. A methodology of phenotyping ICU patients from EHR data: High-fidelity, personalized, and interpretable phenotypes estimation. J Biomed Inform 2023; 148:104547. [PMID: 37984547 PMCID: PMC10802138 DOI: 10.1016/j.jbi.2023.104547] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/24/2023] [Revised: 11/13/2023] [Accepted: 11/16/2023] [Indexed: 11/22/2023]
Abstract
OBJECTIVE Computing phenotypes that provide high-fidelity, time-dependent characterizations and yield personalized interpretations is challenging, especially given the complexity of physiological and healthcare systems and clinical data quality. This paper develops a methodological pipeline to estimate unmeasured physiological parameters and produce high-fidelity, personalized phenotypes anchored to physiological mechanics from electronic health record (EHR). METHODS A methodological phenotyping pipeline is developed that computes new phenotypes defined with unmeasurable computational biomarkers quantifying specific physiological properties in real time. Working within the inverse problem framework, this pipeline is applied to the glucose-insulin system for ICU patients using data assimilation to estimate an established mathematical physiological model with stochastic optimization. This produces physiological model parameter vectors of clinically unmeasured endocrine properties, here insulin secretion, clearance, and resistance, estimated for individual patient. These physiological parameter vectors are used as inputs to unsupervised machine learning methods to produce phenotypic labels and discrete physiological phenotypes. These phenotypes are inherently interpretable because they are based on parametric physiological descriptors. To establish potential clinical utility, the computed phenotypes are evaluated with external EHR data for consistency and reliability and with clinician face validation. RESULTS The phenotype computation was performed on a cohort of 109 ICU patients who received no or short-acting insulin therapy, rendering continuous and discrete physiological phenotypes as specific computational biomarkers of unmeasured insulin secretion, clearance, and resistance on time windows of three days. Six, six, and five discrete phenotypes were found in the first, middle, and last three-day periods of ICU stays, respectively. Computed phenotypic labels were predictive with an average accuracy of 89%. External validation of discrete phenotypes showed coherence and consistency in clinically observable differences based on laboratory measurements and ICD 9/10 codes and clinical concordance from face validity. A particularly clinically impactful parameter, insulin secretion, had a concordance accuracy of 83%±27%. CONCLUSION The new physiological phenotypes computed with individual patient ICU data and defined by estimates of mechanistic model parameters have high physiological fidelity, are continuous, time-specific, personalized, interpretable, and predictive. This methodology is generalizable to other clinical and physiological settings and opens the door for discovering deeper physiological information to personalize medical care.
Collapse
Affiliation(s)
- Yanran Wang
- Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, 13001 East 17th Place, 3rd Floor, Mail Stop B119, Aurora, CO 80045, United States of America; Department of Biomedical Informatics, University of Colorado School of Medicine, Anschutz Health Sciences Building, 1890 N. Revere Court, Mailstop F600, Aurora, CO 80045, United States of America.
| | - J N Stroh
- Department of Biomedical Informatics, University of Colorado School of Medicine, Anschutz Health Sciences Building, 1890 N. Revere Court, Mailstop F600, Aurora, CO 80045, United States of America; Department of Biomedical Engineering, University of Colorado, 12705 East Montview Boulevard, Suite 100, Aurora, CO 80045, United States of America
| | - George Hripcsak
- Biomedical Informatics, Columbia University, 622 W. 168th Street, PH20, New York, NY 10032, United States of America
| | - Cecilia C Low Wang
- Division of Endocrinology, Metabolism and Diabetes, Department of Medicine, University of Colorado School of Medicine, 12801 East 17th Avenue, 7103, Aurora, CO 80045, United States of America
| | - Tellen D Bennett
- Department of Biomedical Informatics, University of Colorado School of Medicine, Anschutz Health Sciences Building, 1890 N. Revere Court, Mailstop F600, Aurora, CO 80045, United States of America
| | - Julia Wrobel
- Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, 1518 Clifton Rd, NE Atlanta, GA 30322, United States of America
| | - Caroline Der Nigoghossian
- Columbia University School of Nursing, 560 West 168th Street, New York, NY 10032, United States of America
| | - Scott W Mueller
- Skaggs School of Pharmacy and Pharmaceutical Sciences, University of Colorado Anschutz Medical Campus, 12850 East Montview Boulevard, Aurora, CO 80045, United States of America
| | - Jan Claassen
- The Neurological Institute of New York, Columbia University Irving Medical Center, 710 West 168th Street, New York NY 10032, United States of America
| | - D J Albers
- Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, 13001 East 17th Place, 3rd Floor, Mail Stop B119, Aurora, CO 80045, United States of America; Department of Biomedical Informatics, University of Colorado School of Medicine, Anschutz Health Sciences Building, 1890 N. Revere Court, Mailstop F600, Aurora, CO 80045, United States of America; Department of Biomedical Engineering, University of Colorado, 12705 East Montview Boulevard, Suite 100, Aurora, CO 80045, United States of America; Biomedical Informatics, Columbia University, 622 W. 168th Street, PH20, New York, NY 10032, United States of America
| |
Collapse
|
4
|
Tsuyuzaki K, Yoshida N, Ishikawa T, Goshima Y, Kawakami E. Non-negative tensor factorization workflow for time series biomedical data. STAR Protoc 2023; 4:102318. [PMID: 37421614 PMCID: PMC10511860 DOI: 10.1016/j.xpro.2023.102318] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/10/2023] [Revised: 03/28/2023] [Accepted: 04/26/2023] [Indexed: 07/10/2023] Open
Abstract
Non-negative tensor factorization (NTF) enables the extraction of a small number of latent components from high-dimensional biomedical data. However, NTF requires many steps, which is a hurdle to implementation. Here, we provide a protocol for TensorLyCV, an easy to run and reproducible NTF analysis pipeline using Snakemake workflow management system and Docker container. Using vaccine adverse reaction data as an example, we describe steps for data processing, tensor decomposition, optimal rank parameter estimation, and visualization of factor matrices. For complete details on the use and execution of this protocol, please refer to Kei Ikeda et al.1.
Collapse
Affiliation(s)
- Koki Tsuyuzaki
- Laboratory for Bioinformatics Research, RIKEN Center for Biosystems Dynamics Research, Wako, Saitama 351-0198, Japan; Japan Science and Technology Agency, PRESTO, 7 Gobancho, Chiyoda-ku, Tokyo 102-0076, Japan.
| | - Naoki Yoshida
- Department of Artificial Intelligence Medicine, Graduate School of Medicine, Chiba University, Chiba 260-8670, Japan
| | - Tetsuo Ishikawa
- Department of Artificial Intelligence Medicine, Graduate School of Medicine, Chiba University, Chiba 260-8670, Japan; Advanced Data Science Project (ADSP), RIKEN Information R&D and Strategy Headquarters, Yokohama, Kanagawa 230-0045, Japan; Department of Extended Intelligence for Medicine, The Ishii-Ishibashi Laboratory, Keio University School of Medicine, Shinjuku-ku, Tokyo 160-8582, Japan
| | - Yuki Goshima
- Advanced Data Science Project (ADSP), RIKEN Information R&D and Strategy Headquarters, Yokohama, Kanagawa 230-0045, Japan
| | - Eiryo Kawakami
- Department of Artificial Intelligence Medicine, Graduate School of Medicine, Chiba University, Chiba 260-8670, Japan; Advanced Data Science Project (ADSP), RIKEN Information R&D and Strategy Headquarters, Yokohama, Kanagawa 230-0045, Japan; NEXT-Ganken Program, Japanese Foundation for Cancer Research (JFCR), Koto Ward, Tokyo 135-8550, Japan; Institute for Advanced Academic Research (IAAR), Chiba University, Chiba 260-8670, Japan.
| |
Collapse
|
5
|
Wang Y, Stroh JN, Hripcsak G, Low Wang CC, Bennett TD, Wrobel J, Der Nigoghossian C, Mueller S, Claassen J, Albers DJ. A methodology of phenotyping ICU patients from EHR data: high-fidelity, personalized, and interpretable phenotypes estimation. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2023:2023.03.15.23287315. [PMID: 37662404 PMCID: PMC10473766 DOI: 10.1101/2023.03.15.23287315] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 09/05/2023]
Abstract
Objective Computing phenotypes that provide high-fidelity, time-dependent characterizations and yield personalized interpretations is challenging, especially given the complexity of physiological and healthcare systems and clinical data quality. This paper develops a methodological pipeline to estimate unmeasured physiological parameters and produce high-fidelity, personalized phenotypes anchored to physiological mechanics from electronic health record (EHR). Methods A methodological phenotyping pipeline is developed that computes new phenotypes defined with unmeasurable computational biomarkers quantifying specific physiological properties in real time. Working within the inverse problem framework, this pipeline is applied to the glucose-insulin system for ICU patients using data assimilation to estimate an established mathematical physiological model with stochastic optimization. This produces physiological model parameter vectors of clinically unmeasured endocrine properties, here insulin secretion, clearance, and resistance, estimated for individual patient. These physiological parameter vectors are used as inputs to unsupervised machine learning methods to produce phenotypic labels and discrete physiological phenotypes. These phenotypes are inherently interpretable because they are based on parametric physiological descriptors. To establish potential clinical utility, the computed phenotypes are evaluated with external EHR data for consistency and reliability and with clinician face validation. Results The phenotype computation was performed on a cohort of 109 ICU patients who received no or short-acting insulin therapy, rendering continuous and discrete physiological phenotypes as specific computational biomarkers of unmeasured insulin secretion, clearance, and resistance on time windows of three days. Six, six, and five discrete phenotypes were found in the first, middle, and last three-day periods of ICU stays, respectively. Computed phenotypic labels were predictive with an average accuracy of 89%. External validation of discrete phenotypes showed coherence and consistency in clinically observable differences based on laboratory measurements and ICD 9/10 codes and clinical concordance from face validity. A particularly clinically impactful parameter, insulin secretion, had a concordance accuracy of 83% ± 27%. Conclusion The new physiological phenotypes computed with individual patient ICU data and defined by estimates of mechanistic model parameters have high physiological fidelity, are continuous, time-specific, personalized, interpretable, and predictive. This methodology is generalizable to other clinical and physiological settings and opens the door for discovering deeper physiological information to personalize medical care.
Collapse
|
6
|
Mohsen F, Al-Saadi B, Abdi N, Khan S, Shah Z. Artificial Intelligence-Based Methods for Precision Cardiovascular Medicine. J Pers Med 2023; 13:1268. [PMID: 37623518 PMCID: PMC10455092 DOI: 10.3390/jpm13081268] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/12/2023] [Revised: 02/27/2023] [Accepted: 03/04/2023] [Indexed: 08/26/2023] Open
Abstract
Precision medicine has the potential to revolutionize the way cardiovascular diseases are diagnosed, predicted, and treated by tailoring treatment strategies to the individual characteristics of each patient. Artificial intelligence (AI) has recently emerged as a promising tool for improving the accuracy and efficiency of precision cardiovascular medicine. In this scoping review, we aimed to identify and summarize the current state of the literature on the use of AI in precision cardiovascular medicine. A comprehensive search of electronic databases, including Scopes, Google Scholar, and PubMed, was conducted to identify relevant studies. After applying inclusion and exclusion criteria, a total of 28 studies were included in the review. We found that AI is being increasingly applied in various areas of cardiovascular medicine, including the diagnosis, prognosis of cardiovascular diseases, risk prediction and stratification, and treatment planning. As a result, most of these studies focused on prediction (50%), followed by diagnosis (21%), phenotyping (14%), and risk stratification (14%). A variety of machine learning models were utilized in these studies, with logistic regression being the most used (36%), followed by random forest (32%), support vector machine (25%), and deep learning models such as neural networks (18%). Other models, such as hierarchical clustering (11%), Cox regression (11%), and natural language processing (4%), were also utilized. The data sources used in these studies included electronic health records (79%), imaging data (43%), and omics data (4%). We found that AI is being increasingly applied in various areas of cardiovascular medicine, including the diagnosis, prognosis of cardiovascular diseases, risk prediction and stratification, and treatment planning. The results of the review showed that AI has the potential to improve the performance of cardiovascular disease diagnosis and prognosis, as well as to identify individuals at high risk of developing cardiovascular diseases. However, further research is needed to fully evaluate the clinical utility and effectiveness of AI-based approaches in precision cardiovascular medicine. Overall, our review provided a comprehensive overview of the current state of knowledge in the field of AI-based methods for precision cardiovascular medicine and offered new insights for researchers interested in this research area.
Collapse
Affiliation(s)
| | | | | | | | - Zubair Shah
- College of Science and Engineering, Hamad Bin Khalifa University, Qatar Foundation, Doha 34110, Qatar
| |
Collapse
|
7
|
Khodadadi A, Ghanbari Bousejin N, Molaei S, Kumar Chauhan V, Zhu T, Clifton DA. Improving Diagnostics with Deep Forest Applied to Electronic Health Records. SENSORS (BASEL, SWITZERLAND) 2023; 23:6571. [PMID: 37514865 PMCID: PMC10384165 DOI: 10.3390/s23146571] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/30/2023] [Revised: 07/08/2023] [Accepted: 07/14/2023] [Indexed: 07/30/2023]
Abstract
An electronic health record (EHR) is a vital high-dimensional part of medical concepts. Discovering implicit correlations in the information of this data set and the research and informative aspects can improve the treatment and management process. The challenge of concern is the data sources' limitations in finding a stable model to relate medical concepts and use these existing connections. This paper presents Patient Forest, a novel end-to-end approach for learning patient representations from tree-structured data for readmission and mortality prediction tasks. By leveraging statistical features, the proposed model is able to provide an accurate and reliable classifier for predicting readmission and mortality. Experiments on MIMIC-III and eICU datasets demonstrate Patient Forest outperforms existing machine learning models, especially when the training data are limited. Additionally, a qualitative evaluation of Patient Forest is conducted by visualising the learnt representations in 2D space using the t-SNE, which further confirms the effectiveness of the proposed model in learning EHR representations.
Collapse
Affiliation(s)
- Atieh Khodadadi
- Institute of Applied Informatics and Formal Description Methods, Karlsruhe Institute of Technology, 76133 Karlsruhe, Germany
| | | | - Soheila Molaei
- Department of Engineering Science, University of Oxford, Oxford OX1 3AZ, UK; (V.K.C.); (T.Z.); (D.A.C.)
| | - Vinod Kumar Chauhan
- Department of Engineering Science, University of Oxford, Oxford OX1 3AZ, UK; (V.K.C.); (T.Z.); (D.A.C.)
| | - Tingting Zhu
- Department of Engineering Science, University of Oxford, Oxford OX1 3AZ, UK; (V.K.C.); (T.Z.); (D.A.C.)
| | - David A. Clifton
- Department of Engineering Science, University of Oxford, Oxford OX1 3AZ, UK; (V.K.C.); (T.Z.); (D.A.C.)
- Oxford-Suzhou Centre for Advanced Research (OSCAR), Suzhou 215123, China
| |
Collapse
|
8
|
Levis M, Levy J, Dufort V, Russ CJ, Shiner B. Dynamic suicide topic modelling: Deriving population-specific, psychosocial and time-sensitive suicide risk variables from Electronic Health Record psychotherapy notes. Clin Psychol Psychother 2023; 30:795-810. [PMID: 36797651 PMCID: PMC11172400 DOI: 10.1002/cpp.2842] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/13/2023] [Accepted: 02/14/2023] [Indexed: 02/18/2023]
Abstract
In the machine learning subfield of natural language processing, a topic model is a type of unsupervised method that is used to uncover abstract topics within a corpus of text. Dynamic topic modelling (DTM) is used for capturing change in these topics over time. The study deploys DTM on corpus of electronic health record psychotherapy notes. This retrospective study examines whether DTM helps distinguish closely matched patients that did and did not die by suicide. Cohort consists of United States Department of Veterans Affairs (VA) patients diagnosed with Posttraumatic Stress Disorder (PTSD) between 2004 and 2013. Each case (those who died by suicide during the year following diagnosis) was matched with five controls (those who remained alive) that shared psychotherapists and had similar suicide risk based on VA's suicide prediction algorithm. Cohort was restricted to patients who received psychotherapy for 9+ months after initial PTSD diagnoses (cases = 77; controls = 362). For cases, psychotherapy notes from diagnosis until death were examined. For controls, psychotherapy notes from diagnosis until matched case's death date were examined. A Python-based DTM algorithm was utilized. Derived topics identified population-specific themes, including PTSD, psychotherapy, medication, communication and relationships. Control topics changed significantly more over time than case topics. Topic differences highlighted engagement, expressivity and therapeutic alliance. This study strengthens groundwork for deriving population-specific, psychosocial and time-sensitive suicide risk variables.
Collapse
Affiliation(s)
- Maxwell Levis
- White River Junction VA Medical Center, Hartford, Vermont, USA
- Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Joshua Levy
- Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Vincent Dufort
- White River Junction VA Medical Center, Hartford, Vermont, USA
| | - Carey J. Russ
- White River Junction VA Medical Center, Hartford, Vermont, USA
- Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Brian Shiner
- White River Junction VA Medical Center, Hartford, Vermont, USA
- Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
- National Center for PTSD Executive Division, Hartford, Vermont, USA
| |
Collapse
|
9
|
Zhou Y, Shi J, Stein R, Liu X, Baldassano RN, Forrest CB, Chen Y, Huang J. Missing data matter: an empirical evaluation of the impacts of missing EHR data in comparative effectiveness research. J Am Med Inform Assoc 2023; 30:1246-1256. [PMID: 37337922 PMCID: PMC10280351 DOI: 10.1093/jamia/ocad066] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2022] [Revised: 03/20/2023] [Accepted: 04/08/2023] [Indexed: 06/21/2023] Open
Abstract
OBJECTIVES The impacts of missing data in comparative effectiveness research (CER) using electronic health records (EHRs) may vary depending on the type and pattern of missing data. In this study, we aimed to quantify these impacts and compare the performance of different imputation methods. MATERIALS AND METHODS We conducted an empirical (simulation) study to quantify the bias and power loss in estimating treatment effects in CER using EHR data. We considered various missing scenarios and used the propensity scores to control for confounding. We compared the performance of the multiple imputation and spline smoothing methods to handle missing data. RESULTS When missing data depended on the stochastic progression of disease and medical practice patterns, the spline smoothing method produced results that were close to those obtained when there were no missing data. Compared to multiple imputation, the spline smoothing generally performed similarly or better, with smaller estimation bias and less power loss. The multiple imputation can still reduce study bias and power loss in some restrictive scenarios, eg, when missing data did not depend on the stochastic process of disease progression. DISCUSSION AND CONCLUSION Missing data in EHRs could lead to biased estimates of treatment effects and false negative findings in CER even after missing data were imputed. It is important to leverage the temporal information of disease trajectory to impute missing values when using EHRs as a data resource for CER and to consider the missing rate and the effect size when choosing an imputation method.
Collapse
Affiliation(s)
- Yizhao Zhou
- Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
- Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
| | - Jiasheng Shi
- Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
- Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
| | - Ronen Stein
- Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
- Department of Pediatrics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Xiaokang Liu
- Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Robert N Baldassano
- Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
- Department of Pediatrics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Christopher B Forrest
- Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
- Department of Pediatrics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Yong Chen
- Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Jing Huang
- Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
- Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
| |
Collapse
|
10
|
Kim DH, Jensen A, Jones K, Raghavan S, Phillips LS, Hung A, Sun YV, Li G, Reaven P, Zhou H, Zhou JJ. A platform for phenotyping disease progression and associated longitudinal risk factors in large-scale EHRs, with application to incident diabetes complications in the UK Biobank. JAMIA Open 2023; 6:ooad006. [PMID: 36789288 PMCID: PMC9912368 DOI: 10.1093/jamiaopen/ooad006] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/06/2022] [Revised: 01/19/2023] [Accepted: 01/31/2023] [Indexed: 02/12/2023] Open
Abstract
Objective Modern healthcare data reflect massive multi-level and multi-scale information collected over many years. The majority of the existing phenotyping algorithms use case-control definitions of disease. This paper aims to study the time to disease onset and progression and identify the time-varying risk factors that drive them. Materials and Methods We developed an algorithmic approach to phenotyping the incidence of diseases by consolidating data sources from the UK Biobank (UKB), including primary care electronic health records (EHRs). We focused on defining events, event dates, and their censoring time, including relevant terms and existing phenotypes, excluding generic, rare, or semantically distant terms, forward-mapping terminology terms, and expert review. We applied our approach to phenotyping diabetes complications, including a composite cardiovascular disease (CVD) outcome, diabetic kidney disease (DKD), and diabetic retinopathy (DR), in the UKB study. Results We identified 49 049 participants with diabetes. Among them, 1023 had type 1 diabetes (T1D), and 40 193 had type 2 diabetes (T2D). A total of 23 833 diabetes subjects had linked primary care records. There were 3237, 3113, and 4922 patients with CVD, DKD, and DR events, respectively. The risk prediction performance for each outcome was assessed, and our results are consistent with the prediction area under the ROC (receiver operating characteristic) curve (AUC) of standard risk prediction models using cohort studies. Discussion and Conclusion Our publicly available pipeline and platform enable streamlined curation of incidence events, identification of time-varying risk factors underlying disease progression, and the definition of a relevant cohort for time-to-event analyses. These important steps need to be considered simultaneously to study disease progression.
Collapse
Affiliation(s)
- Do Hyun Kim
- Department of Biostatistics, University of California, Los Angeles, California, USA
| | - Aubrey Jensen
- Department of Biostatistics, University of California, Los Angeles, California, USA
| | - Kelly Jones
- Department of Computer Science, Columbia University, New York, New York, USA
| | - Sridharan Raghavan
- Division of Hospital Medicine, University of Colorado School of Medicine, Aurora, Colorado, USA
- Rocky Mountain Regional VA Medical Center, Aurora, Colorado, USA
| | - Lawrence S Phillips
- Division of Endocrinology, Emory University School of Medicine, Atlanta, Georgia, USA
- Atlanta VA Medical Center, Decatur, Georgia, USA
| | - Adriana Hung
- VA Tennessee Valley Healthcare System, Nashville, Tennessee, USA
- Vanderbilt University, Nashville, Tennessee, USA
| | - Yan V Sun
- Department of Epidemiology, Emory University, Atlanta, Georgia, USA
| | - Gang Li
- Department of Biostatistics, University of California, Los Angeles, California, USA
- Department of Computational Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, USA
| | - Peter Reaven
- Phoenix VA Health Care System, Phoenix, Arizona, USA
| | - Hua Zhou
- Department of Biostatistics, University of California, Los Angeles, California, USA
- Department of Computational Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, USA
| | - Jin J Zhou
- Department of Biostatistics, University of California, Los Angeles, California, USA
- Phoenix VA Health Care System, Phoenix, Arizona, USA
- Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, California, USA
| |
Collapse
|
11
|
Pham TD, Ravi V, Fan C, Luo B, Sun XF. Tensor Decomposition of Largest Convolutional Eigenvalues Reveals Pathologic Predictive Power of RhoB in Rectal Cancer Biopsy. THE AMERICAN JOURNAL OF PATHOLOGY 2023; 193:579-590. [PMID: 36740183 DOI: 10.1016/j.ajpath.2023.01.007] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/10/2022] [Revised: 12/29/2022] [Accepted: 01/06/2023] [Indexed: 02/05/2023]
Abstract
RhoB protein belongs to the Rho GTPase family, which plays an important role in governing cell signaling and tissue morphology. RhoB expression is known to have implications in pathologic processes of diseases. Investigation in the regulation and communication of this protein, detected by immunohistochemical staining on the microscope, is worth exploring to gain insightful information that may lead to identifying optimal disease treatment options. In particular, the role of RhoB in rectal cancer is not well discovered. Here, we report that methods of deep learning-based image analysis and the decomposition of multiway arrays discover the predictive factor of RhoB in two cohorts of patients with rectal cancer having survival rates of <5 and >5 years. The analysis results show distinctions between the tensor decomposition factors of the two cohorts.
Collapse
Affiliation(s)
- Tuan D Pham
- Center for Artificial Intelligence, Prince Mohammad Bin Fahd University, Khobar, Saudi Arabia.
| | - Vinayakumar Ravi
- Center for Artificial Intelligence, Prince Mohammad Bin Fahd University, Khobar, Saudi Arabia
| | - Chuanwen Fan
- Department of Clinical and Experimental Medicine, Linkoping University, Linkoping, Sweden
| | - Bin Luo
- Department of Clinical and Experimental Medicine, Linkoping University, Linkoping, Sweden; Department of Gastrointestinal Surgery, Sichuan Provincial People's Hospital, Chengdu, China
| | - Xiao-Feng Sun
- Department of Clinical and Experimental Medicine, Linkoping University, Linkoping, Sweden
| |
Collapse
|
12
|
Kerchberger VE, Peterson JF, Wei WQ. Scanning the medical phenome to identify new diagnoses after recovery from COVID-19 in a US cohort. J Am Med Inform Assoc 2023; 30:233-244. [PMID: 36005898 PMCID: PMC9452157 DOI: 10.1093/jamia/ocac159] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/07/2022] [Revised: 06/29/2022] [Accepted: 08/23/2022] [Indexed: 01/20/2023] Open
Abstract
OBJECTIVE COVID-19 survivors are at risk for long-term health effects, but assessing the sequelae of COVID-19 at large scales is challenging. High-throughput methods to efficiently identify new medical problems arising after acute medical events using the electronic health record (EHR) could improve surveillance for long-term consequences of acute medical problems like COVID-19. MATERIALS AND METHODS We augmented an existing high-throughput phenotyping method (PheWAS) to identify new diagnoses occurring after an acute temporal event in the EHR. We then used the temporal-informed phenotypes to assess development of new medical problems among COVID-19 survivors enrolled in an EHR cohort of adults tested for COVID-19 at Vanderbilt University Medical Center. RESULTS The study cohort included 186 105 adults tested for COVID-19 from March 5, 2020 to November 1, 2021; of which 30 088 (16.2%) tested positive. Median follow-up after testing was 412 days (IQR 274-528). Our temporal-informed phenotyping was able to distinguish phenotype chapters based on chronicity of their constituent diagnoses. PheWAS with temporal-informed phenotypes identified increased risk for 43 diagnoses among COVID-19 survivors during outpatient follow-up, including multiple new respiratory, cardiovascular, neurological, and pregnancy-related conditions. Findings were robust to sensitivity analyses, and several phenotypic associations were supported by changes in outpatient vital signs or laboratory tests from the pretesting to postrecovery period. CONCLUSION Temporal-informed PheWAS identified new diagnoses affecting multiple organ systems among COVID-19 survivors. These findings can inform future efforts to enable longitudinal health surveillance for survivors of COVID-19 and other acute medical conditions using the EHR.
Collapse
Affiliation(s)
- Vern Eric Kerchberger
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
- Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Josh F Peterson
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
- Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| | - Wei-Qi Wei
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee, USA
| |
Collapse
|
13
|
Yang S, Varghese P, Stephenson E, Tu K, Gronsbell J. Machine learning approaches for electronic health records phenotyping: a methodical review. J Am Med Inform Assoc 2023; 30:367-381. [PMID: 36413056 PMCID: PMC9846699 DOI: 10.1093/jamia/ocac216] [Citation(s) in RCA: 23] [Impact Index Per Article: 23.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/28/2022] [Revised: 09/27/2022] [Accepted: 10/27/2022] [Indexed: 11/23/2022] Open
Abstract
OBJECTIVE Accurate and rapid phenotyping is a prerequisite to leveraging electronic health records for biomedical research. While early phenotyping relied on rule-based algorithms curated by experts, machine learning (ML) approaches have emerged as an alternative to improve scalability across phenotypes and healthcare settings. This study evaluates ML-based phenotyping with respect to (1) the data sources used, (2) the phenotypes considered, (3) the methods applied, and (4) the reporting and evaluation methods used. MATERIALS AND METHODS We searched PubMed and Web of Science for articles published between 2018 and 2022. After screening 850 articles, we recorded 37 variables on 100 studies. RESULTS Most studies utilized data from a single institution and included information in clinical notes. Although chronic conditions were most commonly considered, ML also enabled the characterization of nuanced phenotypes such as social determinants of health. Supervised deep learning was the most popular ML paradigm, while semi-supervised and weakly supervised learning were applied to expedite algorithm development and unsupervised learning to facilitate phenotype discovery. ML approaches did not uniformly outperform rule-based algorithms, but deep learning offered a marginal improvement over traditional ML for many conditions. DISCUSSION Despite the progress in ML-based phenotyping, most articles focused on binary phenotypes and few articles evaluated external validity or used multi-institution data. Study settings were infrequently reported and analytic code was rarely released. CONCLUSION Continued research in ML-based phenotyping is warranted, with emphasis on characterizing nuanced phenotypes, establishing reporting and evaluation standards, and developing methods to accommodate misclassified phenotypes due to algorithm errors in downstream applications.
Collapse
Affiliation(s)
- Siyue Yang
- Department of Statistical Sciences, University of Toronto, Toronto, Ontario, Canada
| | | | - Ellen Stephenson
- Department of Family & Community Medicine, University of Toronto, Toronto, Ontario, Canada
| | - Karen Tu
- Department of Family & Community Medicine, University of Toronto, Toronto, Ontario, Canada
| | - Jessica Gronsbell
- Department of Statistical Sciences, University of Toronto, Toronto, Ontario, Canada
- Department of Family & Community Medicine, University of Toronto, Toronto, Ontario, Canada
- Department of Computer Science, University of Toronto, Toronto, Ontario, Canada
| |
Collapse
|
14
|
Liang Y, Guo C. Heart failure disease prediction and stratification with temporal electronic health records data using patient representation. Biocybern Biomed Eng 2023. [DOI: 10.1016/j.bbe.2022.12.008] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/03/2023]
|
15
|
Integration of Omics and Phenotypic Data for Precision Medicine. METHODS IN MOLECULAR BIOLOGY (CLIFTON, N.J.) 2022; 2486:19-35. [PMID: 35437716 DOI: 10.1007/978-1-0716-2265-0_2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Subscribe] [Scholar Register] [Indexed: 10/18/2022]
Abstract
Over the past two decades, biomedical research is moving toward a big-data-driven approach. The underlying causes of this transition include the ability to gather genetic or molecular profiles of humans faster, the increasing adoption of electronic health record (EHR) system, and the growing interest in linking omics and phenotypic data for analysis. The integration of individual's biology data (e.g., genomics, proteomics, metabolomics), and health-care data has created unprecedented opportunities for precision medicine, that is, a medical model that uses a patient's unique information, mainly genetic, to prevent, diagnose, or treat disease. This chapter reviewed the research opportunities and applications of integrating omics and phenotypic data for precision medicine, such as understanding the relationship between genotype and phenotype, disease subtyping, and diagnosis or prediction of adverse outcomes. We reviewed the recent advanced methods, particularly the machine learning and deep learning-based approaches used for harnessing and harmonizing the multiomics and phenotypic data to address these applications. We finally discussed the challenges and future directions.
Collapse
|
16
|
Artificial Intelligence: A Shifting Paradigm in Cardio-Cerebrovascular Medicine. J Clin Med 2021; 10:jcm10235710. [PMID: 34884412 PMCID: PMC8658222 DOI: 10.3390/jcm10235710] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/15/2021] [Accepted: 12/02/2021] [Indexed: 12/21/2022] Open
Abstract
The future of healthcare is an organic blend of technology, innovation, and human connection. As artificial intelligence (AI) is gradually becoming a go-to technology in healthcare to improve efficiency and outcomes, we must understand our limitations. We should realize that our goal is not only to provide faster and more efficient care, but also to deliver an integrated solution to ensure that the care is fair and not biased to a group of sub-population. In this context, the field of cardio-cerebrovascular diseases, which encompasses a wide range of conditions-from heart failure to stroke-has made some advances to provide assistive tools to care providers. This article aimed to provide an overall thematic review of recent development focusing on various AI applications in cardio-cerebrovascular diseases to identify gaps and potential areas of improvement. If well designed, technological engines have the potential to improve healthcare access and equitability while reducing overall costs, diagnostic errors, and disparity in a system that affects patients and providers and strives for efficiency.
Collapse
|
17
|
Stevens LM, de Lemos JA, Das SR, Rutan C, Alger HM, Elkind MSV, Zhao J, Iyer K, Figueroa CA, Hall JL. American Heart Association Precision Medicine Platform Addresses Challenges in Data Sharing. Circ Cardiovasc Qual Outcomes 2021; 14:e007949. [PMID: 34517729 PMCID: PMC8452247 DOI: 10.1161/circoutcomes.121.007949] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Affiliation(s)
- Laura M Stevens
- Department of Cardiology and Computational Bioscience Program, University of Colorado, Aurora (L.M.S.).,Quality, Outcomes Research and Analytics, American Heart Association, Dallas, TX (L.M.S., C.R., H.A., J.L.H.)
| | | | - Sandeep R Das
- UT Southwestern Medical Center, Dallas, TX (J.A.d.L., S.R.D.).,Center for Innovation and Value at Parkland, Dallas, TX (S.R.D.)
| | - Christine Rutan
- Quality, Outcomes Research and Analytics, American Heart Association, Dallas, TX (L.M.S., C.R., H.A., J.L.H.)
| | - Heather M Alger
- Quality, Outcomes Research and Analytics, American Heart Association, Dallas, TX (L.M.S., C.R., H.A., J.L.H.)
| | - Mitchell S V Elkind
- Vagelos College of Physicians and Surgeons and Mailman School of Public Health, Columbia University, NewYork-Presbyterian Hospital (M.S.V.E.)
| | - Juan Zhao
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN (J.Z.)
| | - Kritika Iyer
- Departments of Biomedical Engineering (K.I. and C.A.F.), University of Michigan, Ann Arbor
| | - C Alberto Figueroa
- Departments of Biomedical Engineering (K.I. and C.A.F.), University of Michigan, Ann Arbor.,Surgery (C.A.F.), University of Michigan, Ann Arbor
| | - Jennifer L Hall
- Quality, Outcomes Research and Analytics, American Heart Association, Dallas, TX (L.M.S., C.R., H.A., J.L.H.).,Department of Medicine, Lillehei Heart Institute, University of Minnesota, Minneapolis (J.L.H.)
| |
Collapse
|
18
|
Linder JE, Bastarache L, Hughey JJ, Peterson JF. The Role of Electronic Health Records in Advancing Genomic Medicine. Annu Rev Genomics Hum Genet 2021; 22:219-238. [PMID: 34038146 PMCID: PMC9297710 DOI: 10.1146/annurev-genom-121120-125204] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
Abstract
Recent advances in genomic technology and widespread adoption of electronic health records (EHRs) have accelerated the development of genomic medicine, bringing promising research findings from genome science into clinical practice. Genomic and phenomic data, accrued across large populations through biobanks linked to EHRs, have enabled the study of genetic variation at a phenome-wide scale. Through new quantitative techniques, pleiotropy can be explored with phenome-wide association studies, the occurrence of common complex diseases can be predicted using the cumulative influence of many genetic variants (polygenic risk scores), and undiagnosed Mendelian syndromes can be identified using EHR-based phenotypic signatures (phenotype risk scores). In this review, we trace the role of EHRs from the development of genome-wide analytic techniques to translational efforts to test these new interventions to the clinic. Throughout, we describe the challenges that remain when combining EHRs with genetics to improve clinical care.
Collapse
Affiliation(s)
- Jodell E Linder
- Vanderbilt Institute for Clinical and Translational Research, Vanderbilt University Medical Center, Nashville, Tennessee 37203, USA;
| | - Lisa Bastarache
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee 37203, USA; , ,
| | - Jacob J Hughey
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee 37203, USA; , ,
| | - Josh F Peterson
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, Tennessee 37203, USA; , ,
- Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee 37203, USA
| |
Collapse
|
19
|
Majnarić LT, Babič F, O’Sullivan S, Holzinger A. AI and Big Data in Healthcare: Towards a More Comprehensive Research Framework for Multimorbidity. J Clin Med 2021; 10:jcm10040766. [PMID: 33672914 PMCID: PMC7918668 DOI: 10.3390/jcm10040766] [Citation(s) in RCA: 24] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/28/2020] [Revised: 02/02/2021] [Accepted: 02/11/2021] [Indexed: 12/11/2022] Open
Abstract
Multimorbidity refers to the coexistence of two or more chronic diseases in one person. Therefore, patients with multimorbidity have multiple and special care needs. However, in practice it is difficult to meet these needs because the organizational processes of current healthcare systems tend to be tailored to a single disease. To improve clinical decision making and patient care in multimorbidity, a radical change in the problem-solving approach to medical research and treatment is needed. In addition to the traditional reductionist approach, we propose interactive research supported by artificial intelligence (AI) and advanced big data analytics. Such research approach, when applied to data routinely collected in healthcare settings, provides an integrated platform for research tasks related to multimorbidity. This may include, for example, prediction, correlation, and classification problems based on multiple interaction factors. However, to realize the idea of this paradigm shift in multimorbidity research, the optimization, standardization, and most importantly, the integration of electronic health data into a common national and international research infrastructure is needed. Ultimately, there is a need for the integration and implementation of efficient AI approaches, particularly deep learning, into clinical routine directly within the workflows of the medical professionals.
Collapse
Affiliation(s)
- Ljiljana Trtica Majnarić
- Department of Internal Medicine, Family Medicine and the History of Medicine, Faculty of Medicine, University Josip Juraj Strossmayer, 31000 Osijek, Croatia;
- Department of Public Health, Faculty of Dental Medicine, University Josip Juraj Strossmayer, 31000 Osijek, Croatia
| | - František Babič
- Department of Cybernetics and Artificial Intelligence, Faculty of Electrical Engineering and Informatics, Technical University of Košice, 066 01 Košice, Slovakia
- Correspondence: ; Tel.: +421-55-602-4220
| | - Shane O’Sullivan
- Department of Pathology, Faculdade de Medicina, Universidade de São Paulo, 05508-220 São Paulo, Brazil;
| | - Andreas Holzinger
- Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, 8036 Graz, Austria;
| |
Collapse
|
20
|
Hassaine A, Canoy D, Solares JRA, Zhu Y, Rao S, Li Y, Zottoli M, Rahimi K, Salimi-Khorshidi G. Learning multimorbidity patterns from electronic health records using Non-negative Matrix Factorisation. J Biomed Inform 2020; 112:103606. [PMID: 33127447 DOI: 10.1016/j.jbi.2020.103606] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/18/2019] [Revised: 08/01/2020] [Accepted: 10/16/2020] [Indexed: 11/29/2022]
Abstract
Multimorbidity, or the presence of several medical conditions in the same individual, has been increasing in the population - both in absolute and relative terms. Nevertheless, multimorbidity remains poorly understood, and the evidence from existing research to describe its burden, determinants and consequences has been limited. Previous studies attempting to understand multimorbidity patterns are often cross-sectional and do not explicitly account for multimorbidity patterns' evolution over time; some of them are based on small datasets and/or use arbitrary and narrow age ranges; and those that employed advanced models, usually lack appropriate benchmarking and validations. In this study, we (1) introduce a novel approach for using Non-negative Matrix Factorisation (NMF) for temporal phenotyping (i.e., simultaneously mining disease clusters and their trajectories); (2) provide quantitative metrics for the evaluation of these clusters and trajectories; and (3) demonstrate how the temporal characteristics of the disease clusters that result from our model can help mine multimorbidity networks and generate new hypotheses for the emergence of various multimorbidity patterns over time. We trained and evaluated our models on one of the world's largest electronic health records (EHR) datasets, containing more than 7 million patients, from which over 2 million where relevant to, and hence included in this study.
Collapse
Affiliation(s)
- Abdelaali Hassaine
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom
| | - Dexter Canoy
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom
| | - Jose Roberto Ayala Solares
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom
| | - Yajie Zhu
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom
| | - Shishir Rao
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom
| | - Yikuan Li
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom
| | - Mariagrazia Zottoli
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom
| | - Kazem Rahimi
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom.
| | | |
Collapse
|
21
|
Hassaine A, Salimi-Khorshidi G, Canoy D, Rahimi K. Untangling the complexity of multimorbidity with machine learning. Mech Ageing Dev 2020; 190:111325. [PMID: 32768443 PMCID: PMC7493712 DOI: 10.1016/j.mad.2020.111325] [Citation(s) in RCA: 19] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/19/2020] [Revised: 07/28/2020] [Accepted: 07/30/2020] [Indexed: 12/20/2022]
Abstract
The prevalence of multimorbidity has been increasing in recent years, posing a major burden for health care delivery and service. Understanding its determinants and impact is proving to be a challenge yet it offers new opportunities for research to go beyond the study of diseases in isolation. In this paper, we review how the field of machine learning provides many tools for addressing research challenges in multimorbidity. We highlight recent advances in promising methods such as matrix factorisation, deep learning, and topological data analysis and how these can take multimorbidity research beyond cross-sectional, expert-driven or confirmatory approaches to gain a better understanding of evolving patterns of multimorbidity. We discuss the challenges and opportunities of machine learning to identify likely causal links between previously poorly understood disease associations while giving an estimate of the uncertainty on such associations. We finally summarise some of the challenges for wider clinical adoption of machine learning research tools and propose some solutions.
Collapse
Affiliation(s)
- Abdelaali Hassaine
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom; Nuffield Department of Women's and Reproductive Health, University of Oxford, Oxford, United Kingdom
| | - Gholamreza Salimi-Khorshidi
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; Nuffield Department of Women's and Reproductive Health, University of Oxford, Oxford, United Kingdom
| | - Dexter Canoy
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom; Nuffield Department of Women's and Reproductive Health, University of Oxford, Oxford, United Kingdom
| | - Kazem Rahimi
- Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom; NIHR Oxford Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom; Nuffield Department of Women's and Reproductive Health, University of Oxford, Oxford, United Kingdom.
| |
Collapse
|
22
|
Li Y, Nair P, Lu XH, Wen Z, Wang Y, Dehaghi AAK, Miao Y, Liu W, Ordog T, Biernacka JM, Ryu E, Olson JE, Frye MA, Liu A, Guo L, Marelli A, Ahuja Y, Davila-Velderrain J, Kellis M. Inferring multimodal latent topics from electronic health records. Nat Commun 2020; 11:2536. [PMID: 32439869 PMCID: PMC7242436 DOI: 10.1038/s41467-020-16378-3] [Citation(s) in RCA: 26] [Impact Index Per Article: 6.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/23/2019] [Accepted: 04/23/2020] [Indexed: 11/10/2022] Open
Abstract
Electronic health records (EHR) are rich heterogeneous collections of patient health information, whose broad adoption provides clinicians and researchers unprecedented opportunities for health informatics, disease-risk prediction, actionable clinical recommendations, and precision medicine. However, EHRs present several modeling challenges, including highly sparse data matrices, noisy irregular clinical notes, arbitrary biases in billing code assignment, diagnosis-driven lab tests, and heterogeneous data types. To address these challenges, we present MixEHR, a multi-view Bayesian topic model. We demonstrate MixEHR on MIMIC-III, Mayo Clinic Bipolar Disorder, and Quebec Congenital Heart Disease EHR datasets. Qualitatively, MixEHR disease topics reveal meaningful combinations of clinical features across heterogeneous data types. Quantitatively, we observe superior prediction accuracy of diagnostic codes and lab test imputations compared to the state-of-art methods. We leverage the inferred patient topic mixtures to classify target diseases and predict mortality of patients in critical conditions. In all comparison, MixEHR confers competitive performance and reveals meaningful disease-related topics.
Collapse
Affiliation(s)
- Yue Li
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada.
| | - Pratheeksha Nair
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada
| | - Xing Han Lu
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada
| | - Zhi Wen
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada
| | - Yuening Wang
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada
| | | | - Yan Miao
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada
| | - Weiqi Liu
- School of Computer Science and McGill Centre for Bioinformatics, McGill University, Montreal, Quebec, H3A0E9, Canada
| | - Tamas Ordog
- Department of Physiology and Biomedical Engineering and Division of Gastroenterology and Hepatology, Department of Medicine, and Center for Individualized Medicine, Mayo Clinic, Rochester, MN, USA
| | - Joanna M Biernacka
- Department of Health Sciences Research, Mayo Clinic, Rochester, MN, USA
- Department of Psychiatry and Psychology, Mayo Clinic, Rochester, MN, USA
| | - Euijung Ryu
- Department of Health Sciences Research, Mayo Clinic, Rochester, MN, USA
| | - Janet E Olson
- Department of Health Sciences Research, Mayo Clinic, Rochester, MN, USA
| | - Mark A Frye
- Department of Psychiatry and Psychology, Mayo Clinic, Rochester, MN, USA
| | - Aihua Liu
- McGill Adult Unit for Congenital Heart Disease Excellence (MAUDE Unit), Montreal, QC H4A 3J1, Quebec, Canada
| | - Liming Guo
- McGill Adult Unit for Congenital Heart Disease Excellence (MAUDE Unit), Montreal, QC H4A 3J1, Quebec, Canada
| | - Ariane Marelli
- McGill Adult Unit for Congenital Heart Disease Excellence (MAUDE Unit), Montreal, QC H4A 3J1, Quebec, Canada
| | - Yuri Ahuja
- Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, 32 Vassar St, Cambridge, MA, 02139, USA
| | - Jose Davila-Velderrain
- Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, 32 Vassar St, Cambridge, MA, 02139, USA
| | - Manolis Kellis
- Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, 32 Vassar St, Cambridge, MA, 02139, USA.
- The Broad Institute of Harvard and MIT, 415 Main Street, Cambridge, MA, 02142, USA.
| |
Collapse
|
23
|
Weng C, Shah NH, Hripcsak G. Deep phenotyping: Embracing complexity and temporality-Towards scalability, portability, and interoperability. J Biomed Inform 2020; 105:103433. [PMID: 32335224 PMCID: PMC7179504 DOI: 10.1016/j.jbi.2020.103433] [Citation(s) in RCA: 37] [Impact Index Per Article: 9.3] [Reference Citation Analysis] [Grants] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/18/2020] [Accepted: 04/20/2020] [Indexed: 01/07/2023]
Affiliation(s)
- Chunhua Weng
- Department of Biomedical Informatics, Columbia University, New York, NY, USA.
| | - Nigam H Shah
- Medicine - Biomedical Informatics Research, Stanford University, Stanford, CA, USA.
| | - George Hripcsak
- Department of Biomedical Informatics, Columbia University, New York, NY, USA.
| |
Collapse
|
24
|
Afshar A, Perros I, Park H, deFilippi C, Yan X, Stewart W, Ho J, Sun J. TASTE: Temporal and Static Tensor Factorization for Phenotyping Electronic Health Records. PROCEEDINGS OF THE ACM CONFERENCE ON HEALTH, INFERENCE, AND LEARNING 2020; 2020:193-203. [PMID: 33659966 PMCID: PMC7924914 DOI: 10.1145/3368555.3384464] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 01/16/2023]
Abstract
Phenotyping electronic health records (EHR) focuses on defining meaningful patient groups (e.g., heart failure group and diabetes group) and identifying the temporal evolution of patients in those groups. Tensor factorization has been an effective tool for phenotyping. Most of the existing works assume either a static patient representation with aggregate data or only model temporal data. However, real EHR data contain both temporal (e.g., longitudinal clinical visits) and static information (e.g., patient demographics), which are difficult to model simultaneously. In this paper, we propose Temporal And Static TEnsor factorization (TASTE) that jointly models both static and temporal information to extract phenotypes. TASTE combines the PARAFAC2 model with non-negative matrix factorization to model a temporal and a static tensor. To fit the proposed model, we transform the original problem into simpler ones which are optimally solved in an alternating fashion. For each of the sub-problems, our proposed mathematical re-formulations lead to efficient sub-problem solvers. Comprehensive experiments on large EHR data from a heart failure (HF) study confirmed that TASTE is up to 14× faster than several baselines and the resulting phenotypes were confirmed to be clinically meaningful by a cardiologist. Using 60 phenotypes extracted by TASTE, a simple logistic regression can achieve the same level of area under the curve (AUC) for HF prediction compared to a deep learning model using recurrent neural networks (RNN) with 345 features.
Collapse
|