Yikuan Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yikuan Li has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
Dexter Canoy
Nuffield Department of Womens Reproductive Health, University of Oxford, Oxford, U.K.
7 shared publicationsGholamreza Salimi-Khorshidi
Nuffield Department of Womens Reproductive Health, University of Oxford, Oxford, U.K.
7 shared publicationsKazem Rahimi
Nuffield Department of Womens Reproductive Health, University of Oxford, Oxford, U.K.
7 shared publicationsAbdelaali Hassaine
Deep Medicine, Oxford Martin School, University of Oxford, Oxford, United Kingdom
6 shared publicationsShishir Rao
Nuffield Department of Womens Reproductive Health, University of Oxford, Oxford, U.K.
6 shared publications- Shishir Rao
Nuffield Department of Women’s & Reproductive Health, University of Oxford, Oxford, UK
5 shared publications
Latest Publications
- Top Cited PapersOpen Access
- 25 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 27 (2) , 1106-1117
Hi-BEHRT: Hierarchical Transformer-Based Model for Accurate Prediction of Clinical Events Using Multimodal Longitudinal Electronic Health Records
Electronic health records (EHR) represent a holistic overview of patients’ trajectories. Their increasing availability has fueled new hopes to leverage them and develop accurate risk prediction models for a wide range of diseases. Given the complex interrelationships of medical records and patient outcomes, deep learning models have shown clear merits in achieving this goal. However, a key limitation of current study remains their capacity in processing long sequences, and long sequence modelling and its application in the context of healthcare and EHR remains unexplored. Capturing the whole history of medical encounters is expected to lead to more accurate predictions, but the inclusion of records collected for decades and from multiple resources can inevitably exceed the receptive field of the most existing deep learning architectures. This can result in missing crucial, long-term dependencies. To address this gap, we present Hi-BEHRT, a hierarchical Transformer-based model that can significantly expand the receptive field of Transformers and extract associations from much longer sequences. Using a multimodal large-scale linked longitudinal EHR, the Hi-BEHRT exceeds the state-of-the-art deep learning models 1% to 5% for area under the receiver operating characteristic (AUROC) curve and 1% to 8% for area under the precision recall (AUPRC) curve on average, and 2% to 8% (AUROC) and 2% to 11% (AUPRC) for patients with long medical history for 5-year heart failure, diabetes, chronic kidney disease, and stroke risk prediction. Additionally, because pretraining for hierarchical Transformer is not well-established, we provide an effective end-to-end contrastive pre-training strategy for Hi-BEHRT using EHR, improving its transferability on predicting clinical events with relatively small training dataset.
- 25 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 27 (2) , 1106-1117
Top Cited PapersOpen AccessHi-BEHRT: Hierarchical Transformer-Based Model for Accurate Prediction of Clinical Events Using Multimodal Longitudinal Electronic Health Records
Electronic health records (EHR) represent a holistic overview of patients’ trajectories. Their increasing availability has fueled new hopes to leverage them and develop accurate risk prediction models for a wide range of diseases. Given the complex interrelationships of medical records and patient outcomes, deep learning models have shown clear merits in achieving this goal. However, a key limitation of current study remains their capacity in processing long sequences, and long sequence modelling and its application in the context of healthcare and EHR remains unexplored. Capturing the whole history of medical encounters is expected to lead to more accurate predictions, but the inclusion of records collected for decades and from multiple resources can inevitably exceed the receptive field of the most existing deep learning architectures. This can result in missing crucial, long-term dependencies. To address this gap, we present Hi-BEHRT, a hierarchical Transformer-based model that can significantly expand the receptive field of Transformers and extract associations from much longer sequences. Using a multimodal large-scale linked longitudinal EHR, the Hi-BEHRT exceeds the state-of-the-art deep learning models 1% to 5% for area under the receiver operating characteristic (AUROC) curve and 1% to 8% for area under the precision recall (AUPRC) curve on average, and 2% to 8% (AUROC) and 2% to 11% (AUPRC) for patients with long medical history for 5-year heart failure, diabetes, chronic kidney disease, and stroke risk prediction. Additionally, because pretraining for hierarchical Transformer is not well-established, we provide an effective end-to-end contrastive pre-training strategy for Hi-BEHRT using EHR, improving its transferability on predicting clinical events with relatively small training dataset.
- Open Access
- 23 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (4) , 5027-5038
Targeted-BEHRT: Deep Learning for Observational Causal Inference on Longitudinal Electronic Health Records
Observational causal inference is useful for decision-making in medicine when randomized clinical trials (RCTs) are infeasible or nongeneralizable. However, traditional approaches do not always deliver unconfounded causal conclusions in practice. The rise of “doubly robust” nonparametric tools coupled with the growth of deep learning for capturing rich representations of multimodal data offers a unique opportunity to develop and test such models for causal inference on comprehensive electronic health records (EHRs). In this article, we investigate causal modeling of an RCT-established causal association: the effect of classes of antihypertensive on incident cancer risk. We develop a transformer-based model, targeted bidirectional EHR transformer (T-BEHRT) coupled with doubly robust estimation to estimate average risk ratio (RR). We compare our model to benchmark statistical and deep learning models for causal inference in multiple experiments on semi-synthetic derivations of our dataset with various types and intensities of confounding. In order to further test the reliability of our approach, we test our model on situations of limited data. We find that our model provides more accurate estimates of relative risk least sum absolute error (SAE) from ground truth compared with benchmark estimations. Finally, our model provides an estimate of class-wise antihypertensive effect on cancer risk that is consistent with results derived from RCTs.
- 23 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (4) , 5027-5038
Open AccessTargeted-BEHRT: Deep Learning for Observational Causal Inference on Longitudinal Electronic Health Records
Observational causal inference is useful for decision-making in medicine when randomized clinical trials (RCTs) are infeasible or nongeneralizable. However, traditional approaches do not always deliver unconfounded causal conclusions in practice. The rise of “doubly robust” nonparametric tools coupled with the growth of deep learning for capturing rich representations of multimodal data offers a unique opportunity to develop and test such models for causal inference on comprehensive electronic health records (EHRs). In this article, we investigate causal modeling of an RCT-established causal association: the effect of classes of antihypertensive on incident cancer risk. We develop a transformer-based model, targeted bidirectional EHR transformer (T-BEHRT) coupled with doubly robust estimation to estimate average risk ratio (RR). We compare our model to benchmark statistical and deep learning models for causal inference in multiple experiments on semi-synthetic derivations of our dataset with various types and intensities of confounding. In order to further test the reliability of our approach, we test our model on situations of limited data. We find that our model provides more accurate estimates of relative risk least sum absolute error (SAE) from ground truth compared with benchmark estimations. Finally, our model provides an estimate of class-wise antihypertensive effect on cancer risk that is consistent with results derived from RCTs.
- Top Cited PapersOpen Access
- 7 May 2022
- Published by Springer Nature in Current Cardiology Reports
- Vol. 24 (7) , 851-860
How Much Lowering of Blood Pressure Is Required to Prevent Cardiovascular Disease in Patients With and Without Previous Cardiovascular Disease?
Purpose of Review: To review the recent large-scale randomised evidence on pharmacologic reduction in blood pressure for the primary and secondary prevention of cardiovascular disease. Recent Findings: Based on findings of the meta-analysis of individual participant-level data from 48 randomised clinical trials and involving 344,716 participants with mean age of 65 years, the relative reduction in the risk of developing major cardiovascular events was proportional to the magnitude of achieved reduction in blood pressure. For each 5-mmHg reduction in systolic blood pressure, the risk of developing cardiovascular events fell by 10% (hazard ratio [HR] (95% confidence interval [CI], 0.90 [0.88 to 0.92]). When participants were stratified by their history of cardiovascular disease, the HRs (95% CI) in those with and without previous cardiovascular disease were 0.89 (0.86 to 0.92) and 0.91 (0.89 to 0.94), respectively, with no significant heterogeneity in these effects (adjusted P for interaction = 1.0). When these patient groups were further stratified by their baseline systolic blood pressure in increments of 10 mmHg from < 120 to ≥ 170 mmHg, there was no significant heterogeneity in the relative risk reduction across these categories in people with or without previous cardiovascular disease (adjusted P for interaction were 1.00 and 0.28, respectively). Summary: Pharmacologic lowering of blood pressure was effective in preventing major cardiovascular disease events both in people with or without previous cardiovascular disease, which was not modified by their baseline blood pressure level. Treatment effects were shown to be proportional to the intensity of blood pressure reduction, but even modest blood pressure reduction, on average, can lead to meaningful gains in the prevention of incident or recurrent cardiovascular disease.
- 7 May 2022
- Published by Springer Nature in Current Cardiology Reports
- Vol. 24 (7) , 851-860
Top Cited PapersOpen AccessHow Much Lowering of Blood Pressure Is Required to Prevent Cardiovascular Disease in Patients With and Without Previous Cardiovascular Disease?
Purpose of Review: To review the recent large-scale randomised evidence on pharmacologic reduction in blood pressure for the primary and secondary prevention of cardiovascular disease. Recent Findings: Based on findings of the meta-analysis of individual participant-level data from 48 randomised clinical trials and involving 344,716 participants with mean age of 65 years, the relative reduction in the risk of developing major cardiovascular events was proportional to the magnitude of achieved reduction in blood pressure. For each 5-mmHg reduction in systolic blood pressure, the risk of developing cardiovascular events fell by 10% (hazard ratio [HR] (95% confidence interval [CI], 0.90 [0.88 to 0.92]). When participants were stratified by their history of cardiovascular disease, the HRs (95% CI) in those with and without previous cardiovascular disease were 0.89 (0.86 to 0.92) and 0.91 (0.89 to 0.94), respectively, with no significant heterogeneity in these effects (adjusted P for interaction = 1.0). When these patient groups were further stratified by their baseline systolic blood pressure in increments of 10 mmHg from < 120 to ≥ 170 mmHg, there was no significant heterogeneity in the relative risk reduction across these categories in people with or without previous cardiovascular disease (adjusted P for interaction were 1.00 and 0.28, respectively). Summary: Pharmacologic lowering of blood pressure was effective in preventing major cardiovascular disease events both in people with or without previous cardiovascular disease, which was not modified by their baseline blood pressure level. Treatment effects were shown to be proportional to the intensity of blood pressure reduction, but even modest blood pressure reduction, on average, can lead to meaningful gains in the prevention of incident or recurrent cardiovascular disease.
- Top Cited PapersOpen Access
- 7 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 26 (7) , 3362-3372
An Explainable Transformer-Based Deep Learning Model for the Prediction of Incident Heart Failure
Predicting the incidence of complex chronic conditions such as heart failure is challenging. Deep learning models applied to rich electronic health records may improve prediction but remain unexplainable hampering their wider use in medical practice. We aimed to develop a deep-learning framework for accurate and yet explainable prediction of 6-month incident heart failure (HF). Using 100,071 patients from longitudinal linked electronic health records across the U.K., we applied a novel Transformer-based risk model using all community and hospital diagnoses and medications contextualized within the age and calendar year for each patient's clinical encounter. Feature importance was investigated with an ablation analysis to compare model performance when alternatively removing features and by comparing the variability of temporal representations. A post-hoc perturbation technique was conducted to propagate the changes in the input to the outcome for feature contribution analyses. Our model achieved 0.93 area under the receiver operator curve and 0.69 area under the precision-recall curve on internal 5-fold cross validation and outperformed existing deep learning models. Ablation analysis indicated medication is important for predicting HF risk, calendar year is more important than chronological age, which was further reinforced by temporal variability analysis. Contribution analyses identified risk factors that are closely related to HF. Many of them were consistent with existing knowledge from clinical and epidemiological research but several new associations were revealed which had not been considered in expert-driven risk prediction models. In conclusion, the results highlight that our deep learning model, in addition high predictive performance, can inform data-driven risk factor identification.
- 7 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 26 (7) , 3362-3372
Top Cited PapersOpen AccessAn Explainable Transformer-Based Deep Learning Model for the Prediction of Incident Heart Failure
Predicting the incidence of complex chronic conditions such as heart failure is challenging. Deep learning models applied to rich electronic health records may improve prediction but remain unexplainable hampering their wider use in medical practice. We aimed to develop a deep-learning framework for accurate and yet explainable prediction of 6-month incident heart failure (HF). Using 100,071 patients from longitudinal linked electronic health records across the U.K., we applied a novel Transformer-based risk model using all community and hospital diagnoses and medications contextualized within the age and calendar year for each patient's clinical encounter. Feature importance was investigated with an ablation analysis to compare model performance when alternatively removing features and by comparing the variability of temporal representations. A post-hoc perturbation technique was conducted to propagate the changes in the input to the outcome for feature contribution analyses. Our model achieved 0.93 area under the receiver operator curve and 0.69 area under the precision-recall curve on internal 5-fold cross validation and outperformed existing deep learning models. Ablation analysis indicated medication is important for predicting HF risk, calendar year is more important than chronological age, which was further reinforced by temporal variability analysis. Contribution analyses identified risk factors that are closely related to HF. Many of them were consistent with existing knowledge from clinical and epidemiological research but several new associations were revealed which had not been considered in expert-driven risk prediction models. In conclusion, the results highlight that our deep learning model, in addition high predictive performance, can inform data-driven risk factor identification.
- Open Access
- 9 November 2021
- Published by Springer Nature in Scientific Reports
- Vol. 11 (1) , 1
Publisher Correction: Deep Bayesian Gaussian processes for uncertainty estimation in electronic health records
- 9 November 2021
- Published by Springer Nature in Scientific Reports
- Vol. 11 (1) , 1
Open AccessPublisher Correction: Deep Bayesian Gaussian processes for uncertainty estimation in electronic health records
- Open Access
- 19 October 2021
- Published by Springer Nature in Scientific Reports
- Vol. 11 (1) , 1-13
Deep Bayesian Gaussian processes for uncertainty estimation in electronic health records
One major impediment to the wider use of deep learning for clinical decision making is the difficulty of assigning a level of confidence to model predictions. Currently, deep Bayesian neural networks and sparse Gaussian processes are the main two scalable uncertainty estimation methods. However, deep Bayesian neural networks suffer from lack of expressiveness, and more expressive models such as deep kernel learning, which is an extension of sparse Gaussian process, captures only the uncertainty from the higher-level latent space. Therefore, the deep learning model under it lacks interpretability and ignores uncertainty from the raw data. In this paper, we merge features of the deep Bayesian learning framework with deep kernel learning to leverage the strengths of both methods for a more comprehensive uncertainty estimation. Through a series of experiments on predicting the first incidence of heart failure, diabetes and depression applied to large-scale electronic medical records, we demonstrate that our method is better at capturing uncertainty than both Gaussian processes and deep Bayesian neural networks in terms of indicating data insufficiency and identifying misclassifications, with a comparable generalization performance. Furthermore, by assessing the accuracy and area under the receiver operating characteristic curve over the predictive probability, we show that our method is less susceptible to making overconfident predictions, especially for the minority class in imbalanced datasets. Finally, we demonstrate how uncertainty information derived by the model can inform risk factor analysis towards model interpretability.
- 19 October 2021
- Published by Springer Nature in Scientific Reports
- Vol. 11 (1) , 1-13
Open AccessDeep Bayesian Gaussian processes for uncertainty estimation in electronic health records
One major impediment to the wider use of deep learning for clinical decision making is the difficulty of assigning a level of confidence to model predictions. Currently, deep Bayesian neural networks and sparse Gaussian processes are the main two scalable uncertainty estimation methods. However, deep Bayesian neural networks suffer from lack of expressiveness, and more expressive models such as deep kernel learning, which is an extension of sparse Gaussian process, captures only the uncertainty from the higher-level latent space. Therefore, the deep learning model under it lacks interpretability and ignores uncertainty from the raw data. In this paper, we merge features of the deep Bayesian learning framework with deep kernel learning to leverage the strengths of both methods for a more comprehensive uncertainty estimation. Through a series of experiments on predicting the first incidence of heart failure, diabetes and depression applied to large-scale electronic medical records, we demonstrate that our method is better at capturing uncertainty than both Gaussian processes and deep Bayesian neural networks in terms of indicating data insufficiency and identifying misclassifications, with a comparable generalization performance. Furthermore, by assessing the accuracy and area under the receiver operating characteristic curve over the predictive probability, we show that our method is less susceptible to making overconfident predictions, especially for the minority class in imbalanced datasets. Finally, we demonstrate how uncertainty information derived by the model can inform risk factor analysis towards model interpretability.
- Open Access
Transfer Learning in Electronic Health Records through Clinical Concept Embedding
Deep learning models have shown tremendous potential in learning representations, which are able to capture some key properties of the data. This makes them great candidates for transfer learning: Exploiting commonalities between different learning tasks to transfer knowledge from one task to another. Electronic health records (EHR) research is one of the domains that has witnessed a growing number of deep learning techniques employed for learning clinically-meaningful representations of medical concepts (such as diseases and medications). Despite this growth, the approaches to benchmark and assess such learned representations (or, embeddings) is under-investigated; this can be a big issue when such embeddings are shared to facilitate transfer learning. In this study, we aim to (1) train some of the most prominent disease embedding techniques on a comprehensive EHR data from 3.1 million patients, (2) employ qualitative and quantitative evaluation techniques to assess these embeddings, and (3) provide pre-trained disease embeddings for transfer learning. This study can be the first comprehensive approach for clinical concept embedding evaluation and can be applied to any embedding techniques and for any EHR concept.
Open AccessTransfer Learning in Electronic Health Records through Clinical Concept Embedding
Deep learning models have shown tremendous potential in learning representations, which are able to capture some key properties of the data. This makes them great candidates for transfer learning: Exploiting commonalities between different learning tasks to transfer knowledge from one task to another. Electronic health records (EHR) research is one of the domains that has witnessed a growing number of deep learning techniques employed for learning clinically-meaningful representations of medical concepts (such as diseases and medications). Despite this growth, the approaches to benchmark and assess such learned representations (or, embeddings) is under-investigated; this can be a big issue when such embeddings are shared to facilitate transfer learning. In this study, we aim to (1) train some of the most prominent disease embedding techniques on a comprehensive EHR data from 3.1 million patients, (2) employ qualitative and quantitative evaluation techniques to assess these embeddings, and (3) provide pre-trained disease embeddings for transfer learning. This study can be the first comprehensive approach for clinical concept embedding evaluation and can be applied to any embedding techniques and for any EHR concept.
- Open Access
Risk factor identification for incident heart failure using neural network distillation and variable selection
Recent evidence shows that deep learning models trained on electronic health records from millions of patients can deliver substantially more accurate predictions of risk compared to their statistical counterparts. While this provides an important opportunity for improving clinical decision-making, the lack of interpretability is a major barrier to the incorporation of these black-box models in routine care, limiting their trustworthiness and preventing further hypothesis-testing investigations. In this study, we propose two methods, namely, model distillation and variable selection, to untangle hidden patterns learned by an established deep learning model (BEHRT) for risk association identification. Due to the clinical importance and diversity of heart failure as a phenotype, it was used to showcase the merits of the proposed methods. A cohort with 788,880 (8.3% incident heart failure) patients was considered for the study. Model distillation identified 598 and 379 diseases that were associated and dissociated with heart failure at the population level, respectively. While the associations were broadly consistent with prior knowledge, our method also highlighted several less appreciated links that are worth further investigation. In addition to these important population-level insights, we developed an approach to individual-level interpretation to take account of varying manifestation of heart failure in clinical practice. This was achieved through variable selection by detecting a minimal set of encounters that can maximally preserve the accuracy of prediction for individuals. Our proposed work provides a discovery-enabling tool to identify risk factors in both population and individual levels from a data-driven perspective. This helps to generate new hypotheses and guides further investigations on causal links.
Open AccessRisk factor identification for incident heart failure using neural network distillation and variable selection
Recent evidence shows that deep learning models trained on electronic health records from millions of patients can deliver substantially more accurate predictions of risk compared to their statistical counterparts. While this provides an important opportunity for improving clinical decision-making, the lack of interpretability is a major barrier to the incorporation of these black-box models in routine care, limiting their trustworthiness and preventing further hypothesis-testing investigations. In this study, we propose two methods, namely, model distillation and variable selection, to untangle hidden patterns learned by an established deep learning model (BEHRT) for risk association identification. Due to the clinical importance and diversity of heart failure as a phenotype, it was used to showcase the merits of the proposed methods. A cohort with 788,880 (8.3% incident heart failure) patients was considered for the study. Model distillation identified 598 and 379 diseases that were associated and dissociated with heart failure at the population level, respectively. While the associations were broadly consistent with prior knowledge, our method also highlighted several less appreciated links that are worth further investigation. In addition to these important population-level insights, we developed an approach to individual-level interpretation to take account of varying manifestation of heart failure in clinical practice. This was achieved through variable selection by detecting a minimal set of encounters that can maximally preserve the accuracy of prediction for individuals. Our proposed work provides a discovery-enabling tool to identify risk factors in both population and individual levels from a data-driven perspective. This helps to generate new hypotheses and guides further investigations on causal links.
- 1 December 2020
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 112, 103606
Learning multimorbidity patterns from electronic health records using Non-negative Matrix Factorisation
- 1 December 2020
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 112, 103606
Learning multimorbidity patterns from electronic health records using Non-negative Matrix Factorisation
- Top Cited PapersOpen Access
- 28 April 2020
- Published by Springer Nature in Scientific Reports
- Vol. 10 (1) , 1-12
BEHRT: Transformer for Electronic Health Records
Today, despite decades of developments in medicine and the growing interest in precision healthcare, vast majority of diagnoses happen once patients begin to show noticeable signs of illness. Early indication and detection of diseases, however, can provide patients and carers with the chance of early intervention, better disease management, and efficient allocation of healthcare resources. The latest developments in machine learning (including deep learning) provides a great opportunity to address this unmet need. In this study, we introduce BEHRT: A deep neural sequence transduction model for electronic health records (EHR), capable of simultaneously predicting the likelihood of 301 conditions in one’s future visits. When trained and evaluated on the data from nearly 1.6 million individuals, BEHRT shows a striking improvement of 8.0–13.2% (in terms of average precision scores for different tasks), over the existing state-of-the-art deep EHR models. In addition to its scalability and superior accuracy, BEHRT enables personalised interpretation of its predictions; its flexible architecture enables it to incorporate multiple heterogeneous concepts (e.g., diagnosis, medication, measurements, and more) to further improve the accuracy of its predictions; its (pre-)training results in disease and patient representations can be useful for future studies (i.e., transfer learning).
- 28 April 2020
- Published by Springer Nature in Scientific Reports
- Vol. 10 (1) , 1-12
Top Cited PapersOpen AccessBEHRT: Transformer for Electronic Health Records
Today, despite decades of developments in medicine and the growing interest in precision healthcare, vast majority of diagnoses happen once patients begin to show noticeable signs of illness. Early indication and detection of diseases, however, can provide patients and carers with the chance of early intervention, better disease management, and efficient allocation of healthcare resources. The latest developments in machine learning (including deep learning) provides a great opportunity to address this unmet need. In this study, we introduce BEHRT: A deep neural sequence transduction model for electronic health records (EHR), capable of simultaneously predicting the likelihood of 301 conditions in one’s future visits. When trained and evaluated on the data from nearly 1.6 million individuals, BEHRT shows a striking improvement of 8.0–13.2% (in terms of average precision scores for different tasks), over the existing state-of-the-art deep EHR models. In addition to its scalability and superior accuracy, BEHRT enables personalised interpretation of its predictions; its flexible architecture enables it to incorporate multiple heterogeneous concepts (e.g., diagnosis, medication, measurements, and more) to further improve the accuracy of its predictions; its (pre-)training results in disease and patient representations can be useful for future studies (i.e., transfer learning).