Imperial College London

DrBernardHernandez Perez

Faculty of MedicineDepartment of Infectious Disease

Research Fellow



b.hernandez-perez Website CV




B420Electrical EngineeringSouth Kensington Campus





Publication Type

14 results found

Hernandez Perez B, Stiff O, Ming D, Ho Quang C, Vuong Nguyen L, Tuan Nguyen M, Chau Nguyen VV, Nguyet Nguyen M, Huy Nguyen Q, Lam Phung K, Tam Dong Thi H, Trung Dinh T, Trieu Huynh T, Wills B, Cameron Paul S, Holmes A, Yacoub S, Georgiou Pet al., 2023, Learning meaningful latent space representations for patient risk stratification: model development and validation for dengue and other acute febrile illness, Frontiers in Digital Health, Vol: 5, Pages: 1-16, ISSN: 2673-253X

Background: Increased data availability has prompted the creation of clinical decision support systems. These systems utilise clinical information to enhance health care provision, both to predict the likelihood of specific clinical outcomes or evaluate the risk of further complications. However, their adoption remains low due to concerns regarding the quality of recommendations, and a lack of clarity on how results are best obtained and presented.Methods: We used autoencoders capable of reducing the dimensionality of complex datasets in order to produce a 2D representation denoted as latent space to support understanding of complex clinical data. In this output, meaningful representations of individual patient profiles are spatially mapped in an unsupervised manner according to their input clinical parameters. This technique was then applied to a large real-world clinical dataset of over 12,000 patients with an illness compatible with dengue infection in Ho Chi Minh City, Vietnam between 1999 and 2021. Dengue is a systemic viral disease which exerts significant health and economic burden worldwide, and up to 5% of hospitalised patients develop life-threatening complications.Results: The latent space produced by the selected autoencoder aligns with established clinical characteristics exhibited by patients with dengue infection, as well as features of disease progression. Similar clinical phenotypes are represented close to each other in the latent space and clustered according to outcomes broadly described by the World Health Organisation dengue guidelines. Balancing distance metrics and density metrics produced results covering most of the latent space, and improved visualisation whilst preserving utility, with similar patients grouped closer together. In this case, this balance is achieved by using the sigmoid activation function and one hidden layer with three neurons, in addition to the latent dimension layer, which produces the output (Pearson, 0.840; Spearman

Journal article

Ming D, Nguyen QH, An LP, Chanh HQ, Tam DTH, Truong NT, Huy VX, Hernandez B, Van Nuil JI, Paton C, Georgiou P, Nguyen NM, Holmes A, Tho PV, Yacoub S, Ming Det al., 2023, Mapping patient pathways and understanding clinical decision-making in dengue management to inform the development of digital health tools, BMC Medical Informatics and Decision Making, Vol: 23, Pages: 1-9, ISSN: 1472-6947

BackgroundDengue is a common viral illness and severe disease results in life-threatening complications. Healthcare services in low- and middle-income countries treat the majority of dengue cases worldwide. However, the clinical decision-making processes which result in effective treatment are poorly characterised within this setting. In order to improve clinical care through interventions relating to digital clinical decision-support systems (CDSS), we set out to establish a framework for clinical decision-making in dengue management to inform implementation.MethodsWe utilised process mapping and task analysis methods to characterise existing dengue management at the Hospital for Tropical Diseases, Ho Chi Minh City, Vietnam. This is a tertiary referral hospital which manages approximately 30,000 patients with dengue each year, accepting referrals from Ho Chi Minh city and the surrounding catchment area. Initial findings were expanded through semi-structured interviews with clinicians in order to understand clinical reasoning and cognitive factors in detail. A grounded theory was used for coding and emergent themes were developed through iterative discussions with clinician-researchers.ResultsKey clinical decision-making points were identified: (i) at the initial patient evaluation for dengue diagnosis to decide on hospital admission and the provision of fluid/blood product therapy, (ii) in those patients who develop severe disease or other complications, (iii) at the point of recurrent shock in balancing the need for fluid therapy with complications of volume overload. From interviews the following themes were identified: prioritising clinical diagnosis and evaluation over existing diagnostics, the role of dengue guidelines published by the Ministry of Health, the impact of seasonality and caseload on decision-making strategies, and the potential role of digital decision-support and disease scoring tools.ConclusionsThe study highlights the contemporary priorities i

Journal article

Bolton W, Rawson T, Hernandez B, Wilson R, Antcliffe D, Georgiou P, Holmes Aet al., 2022, Machine learning and synthetic outcome estimation for individualised antimicrobial cessation, Frontiers in Digital Health, Vol: 4, Pages: 1-12, ISSN: 2673-253X

The decision on when it is appropriate to stop antimicrobial treatment in an individual patient is complex and under-researched. Ceasing too early can drive treatment failure, while excessive treatment risks adverse events. Under- and over-treatment can promote the development of antimicrobial resistance (AMR). We extracted routinely collected electronic health record data from the MIMIC-IV database for 18,988 patients (22,845 unique stays) who received intravenous antibiotic treatment during an intensive care unit (ICU) admission. A model was developed that utilises a recurrent neural network autoencoder and a synthetic control-based approach to estimate patients’ ICU length of stay (LOS) and mortality outcomes for any given day, under the alternative scenarios of if they were to stop vs. continue antibiotic treatment. Control days where our model should reproduce labels demonstrated minimal difference for both stopping and continuing scenarios indicating estimations are reliable (LOS results of 0.24 and 0.42 days mean delta, 1.93 and 3.76 root mean squared error, respectively). Meanwhile, impact days where we assess the potential effect of the unobserved scenario showed that stopping antibiotic therapy earlier had a statistically significant shorter LOS (mean reduction 2.71 days, p-value <0.01). No impact on mortality was observed. In summary, we have developed a model to reliably estimate patient outcomes under the contrasting scenarios of stopping or continuing antibiotic treatment. Retrospective results are in line with previous clinical studies that demonstrate shorter antibiotic treatment durations are often non-inferior. With additional development into a clinical decision support system, this could be used to support individualised antimicrobial cessation decision-making, reduce the excessive use of antibiotics, and address the problem of AMR.

Journal article

Le V-KD, Hai BH, Karolcik S, Hernandez B, Greeff H, Van HN, Nguyen QKP, Thanh PL, Thwaites L, Georgiou P, Clifton Det al., 2022, vital_sqi: A Python package for physiological signal quality control, FRONTIERS IN PHYSIOLOGY, Vol: 13

Journal article

Ming DK, Hernandez B, Sangkaew S, Vuong NL, Lam PK, Nguyet NM, Tam DTH, Trung DT, Tien NTH, Tuan NM, Chau NVV, Tam CT, Chanh HQ, Trieu HT, Simmons CP, Wills B, Georgiou P, Holmes AH, Yacoub Set al., 2022, Applied machine learning for the risk-stratification and clinical decision support of hospitalised patients with dengue in Vietnam, PLOS Digital Health, Vol: 1, Pages: e0000005-e0000005

BackgroundIdentifying patients at risk of dengue shock syndrome (DSS) is vital for effective healthcare delivery. This can be challenging in endemic settings because of high caseloads and limited resources. Machine learning models trained using clinical data could support decision-making in this context.MethodsWe developed supervised machine learning prediction models using pooled data from adult and paediatric patients hospitalised with dengue. Individuals from 5 prospective clinical studies in Ho Chi Minh City, Vietnam conducted between 12th April 2001 and 30th January 2018 were included. The outcome was onset of dengue shock syndrome during hospitalisation. Data underwent random stratified splitting at 80:20 ratio with the former used only for model development. Ten-fold cross-validation was used for hyperparameter optimisation and confidence intervals derived from percentile bootstrapping. Optimised models were evaluated against the hold-out set.FindingsThe final dataset included 4,131 patients (477 adults and 3,654 children). DSS was experienced by 222 (5.4%) of individuals. Predictors were age, sex, weight, day of illness at hospitalisation, indices of haematocrit and platelets over first 48 hours of admission and before the onset of DSS. An artificial neural network model (ANN) model had best performance with an area under receiver operator curve (AUROC) of 0.83 (95% confidence interval [CI], 0.76–0.85) in predicting DSS. When evaluated against the independent hold-out set this calibrated model exhibited an AUROC of 0.82, specificity of 0.84, sensitivity of 0.66, positive predictive value of 0.18 and negative predictive value of 0.98.InterpretationThe study demonstrates additional insights can be obtained from basic healthcare data, when applied through a machine learning framework. The high negative predictive value could support interventions such as early discharge or ambulatory patient management in this population. Work is underway to incorporate t

Journal article

Hernandez B, Herrero-Viñas P, Rawson TM, Moore LSP, Holmes A, Georgiou P, Hernandez Perez B, Herrero Viñas P, Rawson TM, Moore LSP, Holmes A, Georgiou Pet al., 2021, Resistance trend estimation using regression analysis to enhance antimicrobial surveillance: a multi-centre study in London 2009-2016, Antibiotics, Vol: 10, Pages: 1-16, ISSN: 2079-6382

In the last years, there has been an increase of antimicrobial resistance rates around the world with the misuse and overuse of antimicrobials as one of the main leading drivers. In response to this threat, a variety of initiatives have arisen to promote the efficient use of antimicrobials. These initiatives rely on antimicrobial surveillance systems to promote appropriate prescription practices and are provided by national or global health care institutions with limited consideration of the variations within hospitals. As a consequence, physicians’ adherence to these generic guidelines is still limited. To fill this gap, this work presents an automated approach to performing local antimicrobial surveillance from microbiology data. Moreover, in addition to the commonly reported resistance rates, this work estimates secular resistance trends through regression analysis to provide a single value that effectively communicates the resistance trend to a wider audience. The methods considered for trend estimation were ordinary least squares regression, weighted least squares regression with weights inversely proportional to the number of microbiology records available and autoregressive integrated moving average. Among these, weighted least squares regression was found to be the most robust against changes in the granularity of the time series and presented the best performance. To validate the results, three case studies have been thoroughly compared with the existing literature: (i) Escherichia coli in urine cultures; (ii) Escherichia coli in blood cultures; and (iii) Staphylococcus aureus in wound cultures. The benefits of providing local rather than general antimicrobial surveillance data of a higher quality is two fold. Firstly, it has the potential to stimulate engagement among physicians to strengthen their knowledge and awareness on antimicrobial resistance which might encourage prescribers to change their prescription habits more willingly. Moreover, it pro

Journal article

Ming DKY, Myall A, Hernandez B, Weisse A, Peach R, Barahona M, Rawson T, Holmes Aet al., 2021, Informing antimicrobial management in the context of COVID-19: understanding the longitudinal dynamics of C-reactive protein and procalcitonin, BMC Infectious Diseases, Vol: 21, ISSN: 1471-2334

Background:To characterise the longitudinal dynamics of C-reactive protein (CRP) and Procalcitonin (PCT) in a cohort of hospitalised patients with COVID-19 and support antimicrobial decision-making.Methods:Longitudinal CRP and PCT concentrations and trajectories of 237 hospitalised patients with COVID-19 were modelled. The dataset comprised of 2,021 data points for CRP and 284 points for PCT. Pairwise comparisons were performed between: (i) those with or without significant bacterial growth from cultures, and (ii) those who survived or died in hospital.Results:CRP concentrations were higher over time in COVID-19 patients with positive microbiology (day 9: 236 vs 123 mg/L, p < 0.0001) and in those who died (day 8: 226 vs 152 mg/L, p < 0.0001) but only after day 7 of COVID-related symptom onset. Failure for CRP to reduce in the first week of hospital admission was associated with significantly higher odds of death. PCT concentrations were higher in patients with COVID-19 and positive microbiology or in those who died, although these differences were not statistically significant.Conclusions:Both the absolute CRP concentration and the trajectory during the first week of hospital admission are important factors predicting microbiology culture positivity and outcome in patients hospitalised with COVID-19. Further work is needed to describe the role of PCT for co-infection. Understanding relationships of these biomarkers can support development of risk models and inform optimal antimicrobial strategies.

Journal article

Rawson T, Moore L, Castro Sanchez E, Charani E, Hernandez Perez B, Alividza V, Husson F, Toumazou C, Ahmad R, Georgiou P, Holmes Aet al., 2018, Development of a patient-centred intervention to improve knowledge and understanding of antibiotic therapy in secondary care, Antimicrobial Resistance and Infection Control, Vol: 7, ISSN: 2047-2994

Introduction: We developed a personalised antimicrobial information module co-designed with patients. This study aimed to evaluate the potential impact of this patient-centred intervention on short-term knowledge and understanding of antimicrobial therapy in secondary care. Methods:Thirty previous patients who had received antibiotics in hospital within 12 months were recruited to co-design an intervention to promote patient engagement with infection management. Two workshops, containing five focus-groups were held. These were audio-recorded. Data were analysed using a thematic framework developed deductively based on previous work. Line-by-line coding was performed with new themes added to the framework by two researchers. This was used to inform the development of a patient information module, embedded within an electronic decision support tool (CDSS). The intervention was piloted over a four-week period at Imperial College Healthcare NHS Trust on 30 in-patients. Pre- and post-intervention questionnaires were developed and implemented to assess short term changes in patient knowledge and understanding and provide feedback on the intervention. Data were analysed using SPSS and NVIVO software. Results: Within the workshops, there was consistency in identified themes. The participants agreed upon and co-designed a personalised PDF document that could be integrated into an electronic CDSS to be used by healthcare professionals at the point-of-care. Their aim for the tool was to provide individualised practical information, signpost to reputable information sources, and enhance communication between patients and healthcare professionals.Eighteen out of thirty in-patients consented to participant in the pilot evaluation with 15/18(83%) completing the study. Median (range) age was 66(22-85) years. The majority were male (10/15;66%). Pre-intervention, patients reported desiring further information regarding their infections and antibiotic therapy, including side effects

Journal article

Hernandez Perez B, Herrero Viñas P, Miles Rawson T, SP Moore L, Evans B, Toumazou C, H Holmes A, Georgiou Pet al., 2017, Supervised Learning for Infection Risk Inference Using Pathology Data, BMC Medical Informatics and Decision Making, Vol: 17, ISSN: 1472-6947

Background: Antimicrobial Resistance is threatening our ability to treat common infectious diseases and overuse of antimicrobials to treat human infections in hospitals is accelerating this process. Clinical Decision Support Systems (CDSSs) have been proven to enhance quality of care by promoting change in prescription practices through antimicrobial selection advice. However, bypassing an initial assessment to determine the existence of an underlying disease that justifies the need of antimicrobial therapy might lead to indiscriminate and often unnecessary prescriptions.Methods: From pathology laboratory tests, six biochemical markers were selected and combined with microbiology outcomes from susceptibility tests to create a unique dataset with over one and a half million daily profiles to perform infection risk inference. Outliers were discarded using the inter-quartile range rule and several sampling techniques were studied to tackle the class imbalance problem. The first phase selects the most effective and robust model during training using four-fold stratified cross-validation. The second phase evaluates the final model after isotonic calibration in scenarios with missing inputs and imbalanced class distributions. Results: More than 50\% of infected profiles have daily requested laboratory tests for the six biochemical markers with very promising infection inference results: area under the receiver operating characteristic curve (0.80-0.83), sensitivity (0.64-0.75) and specificity (0.92-0.97). Standardization consistently outperforms normalization and sensitivity is enhanced by using the SMOTE sampling technique. Furthermore, models operated without noticeable loss in performance if at least four biomarkers were available.Conclusion: The selected biomarkers comprise enough information to perform infection risk inference with a high degree of confidence even in the presence of incomplete and imbalanced data. Since they are commonly available in hospitals, Clini

Journal article

Rawson T, moore L, Hernandez B, Charani E, Castro Sanchez E, Herrero P, Hayhoe B, Hope W, Georgiou P, Holmes Aet al., 2017, A systematic review of clinical decision support systems for antimicrobial management: Are we failing to investigate these interventions appropriately?, Clinical Microbiology and Infection, Vol: 23, Pages: 524-532, ISSN: 1469-0691

ObjectivesClinical decision support systems (CDSS) for antimicrobial management can support clinicians to optimise antimicrobial therapy. We reviewed all original literature (qualitative and quantitative) to understand the current scope of CDSS for antimicrobial management and analyse existing methods used to evaluate and report such systems. MethodPRISMA guidelines were followed. Medline, EMBASE, HMIC Health and Management, and Global Health databases were searched from 1st January 1980 to 31st October 2015. All primary research studies describing CDSS for antimicrobial management in adults in primary or secondary care were included. For qualitative studies, thematic synthesis was performed. Quality was assessed using Integrated quality Criteria for the Review Of Multiple Study designs (ICROMS) criteria. CDSS reporting was assessed against a reporting framework for behaviour change intervention implementation.ResultsFifty-eight original articles were included describing 38 independent CDSS. The majority of systems target antimicrobial prescribing (29/38;76%), are platforms integrated with electronic medical records (28/38;74%), and have rules based infrastructure providing decision support (29/38;76%). On evaluation against the intervention reporting framework, CDSS studies fail to report consideration of the non-expert, end-user workflow. They have narrow focus, such as antimicrobial selection, and use proxy outcome measures. Engagement with CDSS by clinicians was poor.ConclusionGreater consideration of the factors that drive non-expert decision making must be considered when designing CDSS interventions. Future work must aim to expand CDSS beyond simply selecting appropriate antimicrobials with clear and systematic reporting frameworks for CDSS interventions developed to address current gaps identified in the reporting of evidence.

Journal article

Hernandez B, Herrero P, Rawson TM, Moore LSP, Charani E, Holmes AH, Georgiou Pet al., 2017, Data-drivenWeb-based Intelligent Decision Support System for Infection Management at Point-Of-Care: Case-Based Reasoning Benefits and Limitations, 10th International Joint Conference on Biomedical Engineering Systems and Technologies, Publisher: SCITEPRESS, Pages: 119-127

Conference paper

Rawson T, Charani E, Moore L, Hernandez B, Castro Sanchez E, Herrero Vinas P, Georgiou P, Holmes Aet al., 2016, Mapping the decision pathways of acute infection management in secondary care among UK medical physicians: a qualitative study, BMC Medicine, Vol: 14, ISSN: 1741-7015

BackgroundThe inappropriate use of antimicrobials drives antimicrobial resistance. We conducted a study to map physician decision making processes for acute infection management in secondary care to identify potential targets for quality improvement interventions.MethodsNewly qualified to Consultant level physicians participated in semi-structured interviews. Interviews were audio recorded and transcribed verbatim for analysis using NVIVO11.0 software. Grounded theory methodology was applied. Analytical categories were created using constant comparison approach to the data and participants were recruited to the study until thematic saturation was reached. ResultsTwenty physicians were interviewed. The decision pathway for the management of acute infections follows a Bayesian-like step-wise approach, with information processed and systematically added to prior assumptions to guide management. The main emerging themes identified as determinants of the decision making of individual physicians were; (i) perceptions of providing “optimal” care for the patient with infection by providing rapid and often intravenous therapy; (ii) perceptions that stopping/de-escalating therapy was a senior doctor decision with junior trainees not expected to contribute; (iii) expectation of interactions with local guidelines and microbiology service advice. Feedback on review of junior doctor prescribing decisions was often lacking, causing frustration and confusion on appropriate practice within this cohort. ConclusionInterventions to improve infection management must incorporate mechanisms to promote distribution of responsibility for decisions made. The disparity between expectations of prescribers to start but not review/stop therapy requires addressing urgently with mechanisms to improve communication and feedback to junior prescribers to facilitate their continued development as prudent antimicrobial prescribers.

Journal article

Rawson T, Moore L, Hernandez B, Castro Sanchez E, Charani E, Georgiou P, Ahmad R, Holmes Aet al., 2016, Patient engagement with infection management in secondary care: a qualitative investigation of current experiences, BMJ Open, Vol: 6, ISSN: 2044-6055

Objective To understand patient engagement with decision-making for infection management in secondary care and the consequences associated with current practices.Design A qualitative investigation using in-depth focus groups.Participants Fourteen members of the public who had received antimicrobials from secondary care in the preceding 12 months in the UK were identified for recruitment. Ten agreed to participate. All participants had experience of infection management in secondary care pathways across a variety of South-East England healthcare institutes. Study findings were subsequently tested through follow-up focus groups with 20 newly recruited citizens.Results Participants reported feelings of disempowerment during episodes of infection in secondary care. Information is communicated in a unilateral manner with individuals ‘told’ that they have an infection and will receive an antimicrobial (often unnamed), leading to loss of ownership, frustration, anxiety and ultimately distancing them from engaging with decision-making. This poor communication drives individuals to seek information from alternative sources, including online, which is associated with concerns over reliability and individualisation. Failures in communication and information provision by clinicians in secondary care influence individuals’ future ideas about infections and their management. This alters their future actions towards antimicrobials and can drive prescription non-adherence and loss to follow-up.Conclusions Current infection management and antimicrobial prescribing practices in secondary care fail to engage patients with the decision-making process. Secondary care physicians must not view infection management episodes as discrete events, but as cumulative experiences which have the potential to shape future patient behaviour and understanding of antimicrobial use.

Journal article

Herrero P, El-Sharkawy M, Pesl P, Hernandez B, Choi L, Awara OM, Lee Y, Lim J, Yusof MM, Sheah A, Yu L, Georgiou Pet al., 2016, Live Demonstrator: Challenging the Bio-inspired Artificial Pancreas with a Mixed-Meal Model Library, IEEE International Symposium on Circuits and Systems (ISCAS), Publisher: IEEE, Pages: 1444-1444, ISSN: 0271-4302

Conference paper

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: Request URI: /respub/WEB-INF/jsp/search-html.jsp Query String: respub-action=search.html&id=00824551&limit=30&person=true