1
|
Toseef M, Olayemi Petinrin O, Wang F, Rahaman S, Liu Z, Li X, Wong KC. Deep transfer learning for clinical decision-making based on high-throughput data: comprehensive survey with benchmark results. Brief Bioinform 2023:bbad254. [PMID: 37455245 DOI: 10.1093/bib/bbad254] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/16/2023] [Revised: 06/04/2023] [Accepted: 06/20/2023] [Indexed: 07/18/2023] Open
Abstract
The rapid growth of omics-based data has revolutionized biomedical research and precision medicine, allowing machine learning models to be developed for cutting-edge performance. However, despite the wealth of high-throughput data available, the performance of these models is hindered by the lack of sufficient training data, particularly in clinical research (in vivo experiments). As a result, translating this knowledge into clinical practice, such as predicting drug responses, remains a challenging task. Transfer learning is a promising tool that bridges the gap between data domains by transferring knowledge from the source to the target domain. Researchers have proposed transfer learning to predict clinical outcomes by leveraging pre-clinical data (mouse, zebrafish), highlighting its vast potential. In this work, we present a comprehensive literature review of deep transfer learning methods for health informatics and clinical decision-making, focusing on high-throughput molecular data. Previous reviews mostly covered image-based transfer learning works, while we present a more detailed analysis of transfer learning papers. Furthermore, we evaluated original studies based on different evaluation settings across cross-validations, data splits and model architectures. The result shows that those transfer learning methods have great potential; high-throughput sequencing data and state-of-the-art deep learning models lead to significant insights and conclusions. Additionally, we explored various datasets in transfer learning papers with statistics and visualization.
Collapse
Affiliation(s)
- Muhammad Toseef
- Department of Computer Science, City University of Hong Kong, Hong Kong SAR
| | | | - Fuzhou Wang
- Department of Computer Science, City University of Hong Kong, Hong Kong SAR
| | - Saifur Rahaman
- Department of Computer Science, City University of Hong Kong, Hong Kong SAR
| | - Zhe Liu
- Department of Computer Science, City University of Hong Kong, Hong Kong SAR
| | - Xiangtao Li
- School of Artificial Intelligence, Jilin University, Jilin, China
| | - Ka-Chun Wong
- Department of Computer Science, City University of Hong Kong, Hong Kong SAR
- Hong Kong Institute for Data Science, City University of Hong Kong, Hong Kong SAR
| |
Collapse
|
2
|
He T, Belouali A, Patricoski J, Lehmann H, Ball R, Anagnostou V, Kreimeyer K, Botsis T. Trends and opportunities in computable clinical phenotyping: A scoping review. J Biomed Inform 2023; 140:104335. [PMID: 36933631 DOI: 10.1016/j.jbi.2023.104335] [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: 11/09/2022] [Revised: 03/07/2023] [Accepted: 03/09/2023] [Indexed: 03/18/2023]
Abstract
Identifying patient cohorts meeting the criteria of specific phenotypes is essential in biomedicine and particularly timely in precision medicine. Many research groups deliver pipelines that automatically retrieve and analyze data elements from one or more sources to automate this task and deliver high-performing computable phenotypes. We applied a systematic approach based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to conduct a thorough scoping review on computable clinical phenotyping. Five databases were searched using a query that combined the concepts of automation, clinical context, and phenotyping. Subsequently, four reviewers screened 7960 records (after removing over 4000 duplicates) and selected 139 that satisfied the inclusion criteria. This dataset was analyzed to extract information on target use cases, data-related topics, phenotyping methodologies, evaluation strategies, and portability of developed solutions. Most studies supported patient cohort selection without discussing the application to specific use cases, such as precision medicine. Electronic Health Records were the primary source in 87.1 % (N = 121) of all studies, and International Classification of Diseases codes were heavily used in 55.4 % (N = 77) of all studies, however, only 25.9 % (N = 36) of the records described compliance with a common data model. In terms of the presented methods, traditional Machine Learning (ML) was the dominant method, often combined with natural language processing and other approaches, while external validation and portability of computable phenotypes were pursued in many cases. These findings revealed that defining target use cases precisely, moving away from sole ML strategies, and evaluating the proposed solutions in the real setting are essential opportunities for future work. There is also momentum and an emerging need for computable phenotyping to support clinical and epidemiological research and precision medicine.
Collapse
Affiliation(s)
- Ting He
- Department of Oncology, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Biomedical Informatics and Data Science Section, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
| | - Anas Belouali
- Biomedical Informatics and Data Science Section, Johns Hopkins University School of Medicine, Baltimore, MD, USA
| | - Jessica Patricoski
- Biomedical Informatics and Data Science Section, Johns Hopkins University School of Medicine, Baltimore, MD, USA
| | - Harold Lehmann
- Biomedical Informatics and Data Science Section, Johns Hopkins University School of Medicine, Baltimore, MD, USA
| | - Robert Ball
- Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, US FDA, Silver Spring, MD, USA
| | - Valsamo Anagnostou
- Department of Oncology, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA
| | - Kory Kreimeyer
- Department of Oncology, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Biomedical Informatics and Data Science Section, Johns Hopkins University School of Medicine, Baltimore, MD, USA
| | - Taxiarchis Botsis
- Department of Oncology, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Biomedical Informatics and Data Science Section, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
| |
Collapse
|
3
|
Liu J, Capurro D, Nguyen A, Verspoor K. "Note Bloat" impacts deep learning-based NLP models for clinical prediction tasks. J Biomed Inform 2022; 133:104149. [PMID: 35878821 DOI: 10.1016/j.jbi.2022.104149] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/22/2022] [Revised: 05/28/2022] [Accepted: 07/19/2022] [Indexed: 10/17/2022]
Abstract
One unintended consequence of the Electronic Health Records (EHR) implementation is the overuse of content-importing technology, such as copy-and-paste, that creates "bloated" notes containing large amounts of textual redundancy. Despite the rising interest in applying machine learning models to learn from real-patient data, it is unclear how the phenomenon of note bloat might affect the Natural Language Processing (NLP) models derived from these notes. Therefore, in this work we examine the impact of redundancy on deep learning-based NLP models, considering four clinical prediction tasks using a publicly available EHR database. We applied two deduplication methods to the hospital notes, identifying large quantities of redundancy, and found that removing the redundancy usually has little negative impact on downstream performances, and can in certain circumstances assist models to achieve significantly better results. We also showed it is possible to attack model predictions by simply adding note duplicates, causing changes of correct predictions made by trained models into wrong predictions. In conclusion, we demonstrated that EHR text redundancy substantively affects NLP models for clinical prediction tasks, showing that the awareness of clinical contexts and robust modeling methods are important to create effective and reliable NLP systems in healthcare contexts.
Collapse
Affiliation(s)
- Jinghui Liu
- School of Computing and Information Systems, The University of Melbourne, Victoria, Australia; Australian e-Health Research Centre, CSIRO, Brisbane, Australia.
| | - Daniel Capurro
- School of Computing and Information Systems, The University of Melbourne, Victoria, Australia; Centre for Digital Transformation of Health, Melbourne Medical School, The University of Melbourne, Victoria, Australia.
| | - Anthony Nguyen
- Australian e-Health Research Centre, CSIRO, Brisbane, Australia.
| | - Karin Verspoor
- School of Computing and Information Systems, The University of Melbourne, Victoria, Australia; Centre for Digital Transformation of Health, Melbourne Medical School, The University of Melbourne, Victoria, Australia; School of Computing Technologies, RMIT University, Victoria, Australia.
| |
Collapse
|
4
|
Si Y, Du J, Li Z, Jiang X, Miller T, Wang F, Jim Zheng W, Roberts K. Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review. J Biomed Inform 2021; 115:103671. [PMID: 33387683 PMCID: PMC11290708 DOI: 10.1016/j.jbi.2020.103671] [Citation(s) in RCA: 43] [Impact Index Per Article: 14.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/16/2020] [Revised: 10/23/2020] [Accepted: 12/23/2020] [Indexed: 12/22/2022]
Abstract
OBJECTIVES Patient representation learning refers to learning a dense mathematical representation of a patient that encodes meaningful information from Electronic Health Records (EHRs). This is generally performed using advanced deep learning methods. This study presents a systematic review of this field and provides both qualitative and quantitative analyses from a methodological perspective. METHODS We identified studies developing patient representations from EHRs with deep learning methods from MEDLINE, EMBASE, Scopus, the Association for Computing Machinery (ACM) Digital Library, and the Institute of Electrical and Electronics Engineers (IEEE) Xplore Digital Library. After screening 363 articles, 49 papers were included for a comprehensive data collection. RESULTS Publications developing patient representations almost doubled each year from 2015 until 2019. We noticed a typical workflow starting with feeding raw data, applying deep learning models, and ending with clinical outcome predictions as evaluations of the learned representations. Specifically, learning representations from structured EHR data was dominant (37 out of 49 studies). Recurrent Neural Networks were widely applied as the deep learning architecture (Long short-term memory: 13 studies, Gated recurrent unit: 11 studies). Learning was mainly performed in a supervised manner (30 studies) optimized with cross-entropy loss. Disease prediction was the most common application and evaluation (31 studies). Benchmark datasets were mostly unavailable (28 studies) due to privacy concerns of EHR data, and code availability was assured in 20 studies. DISCUSSION & CONCLUSION The existing predictive models mainly focus on the prediction of single diseases, rather than considering the complex mechanisms of patients from a holistic review. We show the importance and feasibility of learning comprehensive representations of patient EHR data through a systematic review. Advances in patient representation learning techniques will be essential for powering patient-level EHR analyses. Future work will still be devoted to leveraging the richness and potential of available EHR data. Reproducibility and transparency of reported results will hopefully improve. Knowledge distillation and advanced learning techniques will be exploited to assist the capability of learning patient representation further.
Collapse
Affiliation(s)
- Yuqi Si
- School of Biomedical Informatics, The University of Texas Health Science Center at Houston, TX, USA
| | - Jingcheng Du
- School of Biomedical Informatics, The University of Texas Health Science Center at Houston, TX, USA
| | - Zhao Li
- School of Biomedical Informatics, The University of Texas Health Science Center at Houston, TX, USA
| | - Xiaoqian Jiang
- School of Biomedical Informatics, The University of Texas Health Science Center at Houston, TX, USA
| | - Timothy Miller
- Computational Health Informatics Program (CHIP), Boston Children's Hospital and Harvard Medical School, MA, USA
| | - Fei Wang
- Department of Population Health Sciences. Weill Cornell Medicine, Cornell University, NY, USA
| | - W Jim Zheng
- School of Biomedical Informatics, The University of Texas Health Science Center at Houston, TX, USA
| | - Kirk Roberts
- School of Biomedical Informatics, The University of Texas Health Science Center at Houston, TX, USA.
| |
Collapse
|