140 Exploring dimensionality reduction to facilitate visualisation and analysis of GOSH EHR data
Bibliographic record
Abstract
Background Electronic Health Record (EHR) systems offer valuable insights into patient care and outcomes. Yet, exploitation of EHR data has been limited. The complexity and high dimensionality of this data complicates visualisation, interpretation and efficient analysis. Dimensionality reduction (DR) methods transform the data into a low-dimensional set of features and can facilitate visualisation and subsequent analysis.Methods This study explored the application of DR methods to EHR data at Great Ormond Street Hospital (GOSH) and the challenges around using EHR data for clinical definitions e.g. infection. Our predefined aims were to 1) investigate blood biomarkers in bacterial infection and visually distinguish between causative pathogens and 2) explore the relationship between clinical, physical, and demographic variables in patients diagnosed with cardiomyopathy. Defining an infectious case and observation period was complex and involved various approaches, including ICD-10 diagnoses and lab results. The DR methods employed were PCA, MCA, t-SNE and UMAP, as appropriate. A four-year cardiomyopathy and a 1-year general dataset were extracted using tools developed by the GOSH Digital Research Environment. Datasets shared a common data cleaning protocol followed by pre-processing steps specific to their analysis.Results The results of each clinical use-case were as follows. 1) PCA-derived loadings from blood biomarkers revealed patterns, although bootstrap confidence intervals did not confirm their stability. t-SNE and UMAP did not exhibit a clear advantage over PCA in enhancing visualisation power. 2) We identified clinical dimensions in the MCA results, including ‘cardiopulmonary function’ and ‘age-related physiological response’.Conclusion We explored the application of DR methods to EHR data, evaluating its benefits and challenges. These methods helped identify patterns, such as groups of infections, and understand key factors in cardiomyopathy. Our work is intended to inspire further, more comprehensive investigations on larger datasets and different clinical use cases.Acknowledgements for Funding or Support This work is supported by the NIHR GOSH BRC. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. This work is supported by the Great Ormond Street Hospital Children’s Charity.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.002 |
| Open science | 0.000 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.010 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".