1
|
Holdefer AA, Pizarro J, Saunders-Hastings P, Beers J, Sang A, Hettinger AZ, Blumenthal J, Martinez E, Jones LD, Deady M, Ezzeldin H, Anderson SA. Development of Interoperable Computable Phenotype Algorithms for Adverse Events of Special Interest to Be Used for Biologics Safety Surveillance: Validation Study. JMIR Public Health Surveill 2024; 10:e49811. [PMID: 39008361 PMCID: PMC11287092 DOI: 10.2196/49811] [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: 06/09/2023] [Revised: 02/24/2024] [Accepted: 05/26/2024] [Indexed: 07/16/2024] Open
Abstract
BACKGROUND Adverse events associated with vaccination have been evaluated by epidemiological studies and more recently have gained additional attention with the emergency use authorization of several COVID-19 vaccines. As part of its responsibility to conduct postmarket surveillance, the US Food and Drug Administration continues to monitor several adverse events of special interest (AESIs) to ensure vaccine safety, including for COVID-19. OBJECTIVE This study is part of the Biologics Effectiveness and Safety Initiative, which aims to improve the Food and Drug Administration's postmarket surveillance capabilities while minimizing public burden. This study aimed to enhance active surveillance efforts through a rules-based, computable phenotype algorithm to identify 5 AESIs being monitored by the Center for Disease Control and Prevention for COVID-19 or other vaccines: anaphylaxis, Guillain-Barré syndrome, myocarditis/pericarditis, thrombosis with thrombocytopenia syndrome, and febrile seizure. This study examined whether these phenotypes have sufficiently high positive predictive value (PPV) to ensure that the cases selected for surveillance are reasonably likely to be a postbiologic adverse event. This allows patient privacy, and security concerns for the data sharing of patients who had nonadverse events can be properly accounted for when evaluating the cost-benefit aspect of our approach. METHODS AESI phenotype algorithms were developed to apply to electronic health record data at health provider organizations across the country by querying for standard and interoperable codes. The codes queried in the rules represent symptoms, diagnoses, or treatments of the AESI sourced from published case definitions and input from clinicians. To validate the performance of the algorithms, we applied them to electronic health record data from a US academic health system and provided a sample of cases for clinicians to evaluate. Performance was assessed using PPV. RESULTS With a PPV of 93.3%, our anaphylaxis algorithm performed the best. The PPVs for our febrile seizure, myocarditis/pericarditis, thrombocytopenia syndrome, and Guillain-Barré syndrome algorithms were 89%, 83.5%, 70.2%, and 47.2%, respectively. CONCLUSIONS Given our algorithm design and performance, our results support continued research into using interoperable algorithms for widespread AESI postmarket detection.
Collapse
Affiliation(s)
| | | | | | | | | | - Aaron Zachary Hettinger
- Center for Biostatistics, Informatics and Data Science, MedStar Health Research Institute, Columbia, MD, United States
- Department of Emergency Medicine, Georgetown University School of Medicine, Washington, DC, United States
| | - Joseph Blumenthal
- Center for Biostatistics, Informatics and Data Science, MedStar Health Research Institute, Columbia, MD, United States
| | | | | | | | - Hussein Ezzeldin
- Center for Biologics Evaluation and Research, United States Food and Drug Administration, Silver Spring, MD, United States
| | - Steven A Anderson
- Center for Biologics Evaluation and Research, United States Food and Drug Administration, Silver Spring, MD, United States
| |
Collapse
|
2
|
Fu S, Jia H, Vassilaki M, Keloth VK, Dang Y, Zhou Y, Garg M, Petersen RC, St Sauver J, Moon S, Wang L, Wen A, Li F, Xu H, Tao C, Fan J, Liu H, Sohn S. FedFSA: Hybrid and federated framework for functional status ascertainment across institutions. J Biomed Inform 2024; 152:104623. [PMID: 38458578 PMCID: PMC11005095 DOI: 10.1016/j.jbi.2024.104623] [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: 10/12/2023] [Revised: 01/12/2024] [Accepted: 03/04/2024] [Indexed: 03/10/2024]
Abstract
INTRODUCTION Patients' functional status assesses their independence in performing activities of daily living, including basic ADLs (bADL), and more complex instrumental activities (iADL). Existing studies have discovered that patients' functional status is a strong predictor of health outcomes, particularly in older adults. Depite their usefulness, much of the functional status information is stored in electronic health records (EHRs) in either semi-structured or free text formats. This indicates the pressing need to leverage computational approaches such as natural language processing (NLP) to accelerate the curation of functional status information. In this study, we introduced FedFSA, a hybrid and federated NLP framework designed to extract functional status information from EHRs across multiple healthcare institutions. METHODS FedFSA consists of four major components: 1) individual sites (clients) with their private local data, 2) a rule-based information extraction (IE) framework for ADL extraction, 3) a BERT model for functional status impairment classification, and 4) a concept normalizer. The framework was implemented using the OHNLP Backbone for rule-based IE and open-source Flower and PyTorch library for federated BERT components. For gold standard data generation, we carried out corpus annotation to identify functional status-related expressions based on ICF definitions. Four healthcare institutions were included in the study. To assess FedFSA, we evaluated the performance of category- and institution-specific ADL extraction across different experimental designs. RESULTS ADL extraction performance ranges from an F1-score of 0.907 to 0.986 for bADL and 0.825 to 0.951 for iADL across the four healthcare sites. The performance for ADL extraction with impairment ranges from an F1-score of 0.722 to 0.954 for bADL and 0.674 to 0.813 for iADL across four healthcare sites. For category-specific ADL extraction, laundry and transferring yielded relatively high performance, while dressing, medication, bathing, and continence achieved moderate-high performance. Conversely, food preparation and toileting showed low performance. CONCLUSION NLP performance varied across ADL categories and healthcare sites. Federated learning using a FedFSA framework performed higher than non-federated learning for impaired ADL extraction at all healthcare sites. Our study demonstrated the potential of the federated learning framework in functional status extraction and impairment classification in EHRs, exemplifying the importance of a large-scale, multi-institutional collaborative development effort.
Collapse
Affiliation(s)
- Sunyang Fu
- Mayo Clinic, Rochester, MN, United States; University of Texas Health Science Center, Houston, TX, United States.
| | - Heling Jia
- Mayo Clinic, Rochester, MN, United States.
| | | | | | - Yifang Dang
- University of Texas Health Science Center, Houston, TX, United States.
| | - Yujia Zhou
- University of Texas Health Science Center, Houston, TX, United States.
| | | | | | | | | | - Liwei Wang
- Mayo Clinic, Rochester, MN, United States.
| | - Andrew Wen
- University of Texas Health Science Center, Houston, TX, United States.
| | - Fang Li
- University of Texas Health Science Center, Houston, TX, United States.
| | - Hua Xu
- Yale University, New Haven, CT, United States.
| | - Cui Tao
- University of Texas Health Science Center, Houston, TX, United States.
| | | | - Hongfang Liu
- Mayo Clinic, Rochester, MN, United States; University of Texas Health Science Center, Houston, TX, United States.
| | | |
Collapse
|
3
|
Almalki J, Alshahrani SM, Khan NA. A comprehensive secure system enabling healthcare 5.0 using federated learning, intrusion detection and blockchain. PeerJ Comput Sci 2024; 10:e1778. [PMID: 38259900 PMCID: PMC10803090 DOI: 10.7717/peerj-cs.1778] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/27/2023] [Accepted: 12/05/2023] [Indexed: 01/24/2024]
Abstract
Recently, the use of the Internet of Medical Things (IoMT) has gained popularity across various sections of the health sector. The historical security risks of IoMT devices themselves and the data flowing from them are major concerns. Deploying many devices, sensors, services, and networks that connect the IoMT systems is gaining popularity. This study focuses on identifying the use of blockchain in innovative healthcare units empowered by federated learning. A collective use of blockchain with intrusion detection management (IDM) is beneficial to detect and prevent malicious activity across the storage nodes. Data accumulated at a centralized storage node is analyzed with the help of machine learning algorithms to diagnose disease and allow appropriate medication to be prescribed by a medical healthcare professional. The model proposed in this study focuses on the effective use of such models for healthcare monitoring. The amalgamation of federated learning and the proposed model makes it possible to reach 93.89 percent accuracy for disease analysis and addiction. Further, intrusion detection ensures a success rate of 97.13 percent in this study.
Collapse
Affiliation(s)
- Jameel Almalki
- Department of Computer Science, College of Computer in Al-Lith, Umm Al-Qura University, Makkah, Makkah, Saudi Arabia
| | - Saeed M. Alshahrani
- Department of Computer Science, College of Computing and Information Technology, Shaqra University, Shaqra, Riyadh, Saudi Arabia
| | - Nayyar Ahmed Khan
- Department of Computer Science, College of Computing and Information Technology, Shaqra University, Shaqra, Riyadh, Saudi Arabia
| |
Collapse
|
4
|
Huang CT, Wang TJ, Kuo LK, Tsai MJ, Cia CT, Chiang DH, Chang PJ, Chong IW, Tsai YS, Chu YC, Liu CJ, Chen CH, Pai KC, Wu CL. Federated machine learning for predicting acute kidney injury in critically ill patients: a multicenter study in Taiwan. Health Inf Sci Syst 2023; 11:48. [PMID: 37822805 PMCID: PMC10562351 DOI: 10.1007/s13755-023-00248-5] [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: 07/05/2023] [Accepted: 09/20/2023] [Indexed: 10/13/2023] Open
Abstract
Purpose To address the contentious data sharing across hospitals, this study adopted a novel approach, federated learning (FL), to establish an aggregate model for acute kidney injury (AKI) prediction in critically ill patients in Taiwan. Methods This study used data from the Critical Care Database of Taichung Veterans General Hospital (TCVGH) from 2015 to 2020 and electrical medical records of the intensive care units (ICUs) between 2018 and 2020 of four referral centers in different areas across Taiwan. AKI prediction models were trained and validated thereupon. An FL-based prediction model across hospitals was then established. Results The study included 16,732 ICU admissions from the TCVGH and 38,424 ICU admissions from the other four hospitals. The complete model with 60 features and the parsimonious model with 21 features demonstrated comparable accuracies using extreme gradient boosting, neural network (NN), and random forest, with an area under the receiver-operating characteristic (AUROC) curve of approximately 0.90. The Shapley Additive Explanations plot demonstrated that the selected features were the key clinical components of AKI for critically ill patients. The AUROC curve of the established parsimonious model for external validation at the four hospitals ranged from 0.760 to 0.865. NN-based FL slightly improved the model performance at the four centers. Conclusion A reliable prediction model for AKI in ICU patients was developed with a lead time of 24 h, and it performed better when the novel FL platform across hospitals was implemented. Supplementary Information The online version contains supplementary material available at 10.1007/s13755-023-00248-5.
Collapse
Affiliation(s)
- Chun-Te Huang
- Institute of Emergency and Critical Care Medicine, National Yang-Ming Chiao Tung University, Taipei, Taiwan
- Nephrology and Critical Care Medicine, Department of Internal Medicine and Critical Care Medicine, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Tsai-Jung Wang
- Nephrology and Critical Care Medicine, Department of Internal Medicine and Critical Care Medicine, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Li-Kuo Kuo
- Department of Critical Care Medicine, MacKay Memorial Hospital, Taipei, Taiwan
| | - Ming-Ju Tsai
- Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, School of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
| | - Cong-Tat Cia
- Division of Critical Care Medicine, Department of Internal Medicine, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan
| | - Dung-Hung Chiang
- Department of Critical Care Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
| | - Po-Jen Chang
- Department of Information Technology, MacKay Memorial Hospital, Taipei, Taiwan
| | - Inn-Wen Chong
- Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, School of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
| | - Yi-Shan Tsai
- Department of Diagnostic Radiology, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan
| | - Yuan-Chia Chu
- Department of Information Technology, Taipei Veterans General Hospital, Taipei, Taiwan
| | - Chia-Jen Liu
- Institute of Emergency and Critical Care Medicine, National Yang-Ming Chiao Tung University, Taipei, Taiwan
| | - Cheng-Hsu Chen
- Division of Nephrology, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Kai-Chih Pai
- College of Engineering, Tunghai University, Taichung, Taiwan
| | - Chieh-Liang Wu
- College of Medicine, National Chung Hshin University, Taichung, Taiwan
| |
Collapse
|
5
|
Wolff J, Matschinske J, Baumgart D, Pytlik A, Keck A, Natarajan A, von Schacky CE, Pauling JK, Baumbach J. Federated machine learning for a facilitated implementation of Artificial Intelligence in healthcare - a proof of concept study for the prediction of coronary artery calcification scores. J Integr Bioinform 2022; 19:jib-2022-0032. [PMID: 36054833 PMCID: PMC9800042 DOI: 10.1515/jib-2022-0032] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/14/2022] [Revised: 08/03/2022] [Accepted: 08/11/2022] [Indexed: 01/09/2023] Open
Abstract
The implementation of Artificial Intelligence (AI) still faces significant hurdles and one key factor is the access to data. One approach that could support that is federated machine learning (FL) since it allows for privacy preserving data access. For this proof of concept, a prediction model for coronary artery calcification scores (CACS) has been applied. The FL was trained based on the data in the different institutions, while the centralized machine learning model was trained on one allocation of data. Both algorithms predict patients with risk scores ≥5 based on age, biological sex, waist circumference, dyslipidemia and HbA1c. The centralized model yields a sensitivity of c. 66% and a specificity of c. 70%. The FL slightly outperforms that with a sensitivity of 67% while slightly underperforming it with a specificity of 69%. It could be demonstrated that CACS prediction is feasible via both, a centralized and an FL approach, and that both show very comparable accuracy. In order to increase accuracy, additional and a higher volume of patient data is required and for that FL is utterly necessary. The developed "CACulator" serves as proof of concept, is available as research tool and shall support future research to facilitate AI implementation.
Collapse
Affiliation(s)
- Justus Wolff
- Chair of Experimental Bioinformatics, TUM School of Life Sciences Weihenstephan, Technical University of Munich, Maximus-von-Imhof-Forum 3, 85354Freising, Germany
- Syte – Strategy Institute for Digital Health, Hohe Bleichen 8, 20354Hamburg, Germany
| | - Julian Matschinske
- Chair of Computational Systems Biology, University of Hamburg, Notkestreet 9-11, 22607Hamburg, Germany
| | - Dietrich Baumgart
- Preventicum Essen, Theodor-Althoff-Str. 47 45133Essen, Germany
- Preventicum Duesseldorf, Koenigsallee 11, 40212Duesseldorf, Germany
| | - Anne Pytlik
- Preventicum Essen, Theodor-Althoff-Str. 47 45133Essen, Germany
- Preventicum Duesseldorf, Koenigsallee 11, 40212Duesseldorf, Germany
| | - Andreas Keck
- Syte – Strategy Institute for Digital Health, Hohe Bleichen 8, 20354Hamburg, Germany
| | - Arunakiry Natarajan
- Independent Researcher, Digital Health, Informatics and Data Science, Lower Saxony, Germany
| | - Claudio E. von Schacky
- Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Ismaningerstr. 22, 81675Munich, Germany
| | - Josch K. Pauling
- Chair of Experimental Bioinformatics, TUM School of Life Sciences Weihenstephan, Technical University of Munich, Maximus-von-Imhof-Forum 3, 85354Freising, Germany
- LipiTUM, Chair of Experimental Bioinformatics, TUM School of Life Sciences, Technical University of Munich, Maximus-von-Imhof-Forum 3, 85354Freising, Germany
| | - Jan Baumbach
- Chair of Computational Systems Biology, University of Hamburg, Notkestreet 9-11, 22607Hamburg, Germany
- Computational BioMedicine Lab, Institute of Mathematics and Computer Science, University of Southern Denmark, Campusvej 55, 5230Odense, Denmark
| |
Collapse
|