Individualising intensive systolic blood pressure reduction in hypertension using computational trial phenomaps and machine learning: a post-hoc analysis of randomised clinical trials

Oikonomou, Evangelos; Spatz, Erica S.; Suchard, Marc A.; Khera, Rohan

doi:10.1016/s2589-7500(22)00170-4

Cited by 38 publications

(30 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach extends to a large variety of human activities and in hypertension has produced promising results for an improvement of the ability to predict the risk of incident hypertension and future organ damage [1733,1734]. Promising results have also been obtained on the possibility to personalize antihypertensive treatment [1735,1736]. Improved prediction of future hypertension and HMOD development by these approaches would be of particular importance, because of the possibility to focus intensive preventive treatment on people at greater risk (Table 27).…”

Section: Gaps In Evidence and Future Opportunitiesmentioning

confidence: 99%

2023 ESH Guidelines for the management of arterial hypertension The Task Force for the management of arterial hypertension of the European Society of Hypertension

Chairperson

Co-Chair

Brunström

et al. 2023

Journal of Hypertension

1,096

280

View full text Add to dashboard Cite

Document Reviewers: Luis Alcocer (Mexico), Christina Antza (Greece), Mustafa Arici (Turkey), Eduardo Barbosa (Brazil), Adel Berbari (Lebanon), Luís Bronze (Portugal), John Chalmers (Australia), Tine De Backer (Belgium), Alejandro de la Sierra (Spain), Kyriakos Dimitriadis (Greece), Dorota Drozdz (Poland), Béatrice Duly-Bouhanick (France), Brent M. Egan (USA), Serap Erdine (Turkey), Claudio Ferri (Italy), Slavomira Filipova (Slovak Republic), Anthony Heagerty (UK), Michael Hecht Olsen (Denmark), Dagmara Hering (Poland), Sang Hyun Ihm (South Korea), Uday Jadhav (India), Manolis Kallistratos (Greece), Kazuomi Kario (Japan), Vasilios Kotsis (Greece), Adi Leiba (Israel), Patricio López-Jaramillo (Colombia), Hans-Peter Marti (Norway), Terry McCormack (UK), Paolo Mulatero (Italy), Dike B. Ojji (Nigeria), Sungha Park (South Korea), Priit Pauklin (Estonia), Sabine Perl (Austria), Arman Postadzhian (Bulgaria), Aleksander Prejbisz (Poland), Venkata Ram (India), Ramiro Sanchez (Argentina), Markus Schlaich (Australia), Alta Schutte (Australia), Cristina Sierra (Spain), Sekib Sokolovic (Bosnia and Herzegovina), Jonas Spaak (Sweden), Dimitrios Terentes-Printzios (Greece), Bruno Trimarco (Italy), Thomas Unger (The Netherlands), Bert-Jan van den Born (The Netherlands), Anna Vachulova (Slovak Republic), Agostino Virdis (Italy), Jiguang Wang (China), Ulrich Wenzel (Germany), Paul Whelton (USA), Jiri Widimsky (Czech Republic), Jacek Wolf (Poland), Grégoire Wuerzner (Switzerland), Eugene Yang (USA), Yuqing Zhang (China).

show abstract

Section: Gaps In Evidence and Future Opportunitiesmentioning

confidence: 99%

2023 ESH Guidelines for the management of arterial hypertension The Task Force for the management of arterial hypertension of the European Society of Hypertension

Chairperson

Co-Chair

Brunström

et al. 2023

Journal of Hypertension

1,096

280

View full text Add to dashboard Cite

show abstract

“…In past work, we highlighted that the sample size at which the training loss curve of an autoencoder converges corresponds to the sample size at which a training loss curve for a neural network classifier trained on the same data converges. We showed that this holds in a variety of computer vision datasets and on latent spaces belonging to a variety of structures, which included varying the number of classes, number of informative features, number of total features, number of class-specific subgroups, and the relationships between different variables within the dataset 68,69 . In this work, we build on that approach to the problem of bias mitigation in healthcare.…”

Section: Discussionmentioning

confidence: 87%

“…For example, the use of diffusion algorithms in conjunction with data-centric strategies may yield compounding effects 5 . Third, the labels for the MIMIC-CXR dataset were obtained using the CheXpert system of natural language processing of radiology reports, which is known to have a substantial and varying amount of mislabeling for different diagnoses 69 . Results from Zhang et al suggest there is no statistically detectable label bias by race in MIMIC-CXR for the ‘no finding’ label 72 .…”

Section: Discussionmentioning

confidence: 99%

Derivation and Validation of a Machine Learning Approach to Detect and Mitigate Biases in Healthcare Data

Gulamali,

Sawant,

Liharska

et al. 2023

Preprint

View full text Add to dashboard Cite

The adoption of diagnosis and prognostic algorithms in healthcare has led to concerns about the perpetuation of bias against disadvantaged groups of individuals. Deep learning methods to detect and mitigate bias have revolved around modifying models, optimization strategies, and threshold calibration with varying levels of success. Here, we generate a data-centric, model-agnostic, task-agnostic approach to evaluate dataset bias by investigating the relationship between how easily different groups are learned at small sample sizes (AEquity). We then apply a systematic analysis of AEq values across subpopulations to identify and mitigate manifestations of racial bias in two known cases in healthcare – Chest X-rays diagnosis with deep convolutional neural networks and healthcare utilization prediction with multivariate logistic regression. AEq is a novel and broadly applicable metric that can be applied to advance equity by diagnosing and remediating bias in healthcare datasets.

show abstract

“…It has been changing, though. Currently, new research results based on analysis of the SPRINT and ACCORD trial data incorporating ML methods demonstrated potential for the individualization of hypertension treatment [37].…”

Section: Discussionmentioning

confidence: 99%

Answering Clinical Questions Using Machine Learning: Should We Look at Diastolic Blood Pressure When Tailoring Blood Pressure Control?

Siński

Berka

Lewandowski

et al. 2022

JCM

View full text Add to dashboard Cite

Background: The guidelines recommend intensive blood pressure control. Randomized trials have focused on the relevance of the systolic blood pressure (SBP) lowering, leaving the safety of the diastolic blood pressure (DBP) reduction unresolved. There are data available which show that low DBP should not stop clinicians from achieving SBP targets; however, registries and analyses of randomized trials present conflicting results. The purpose of the study was to apply machine learning (ML) algorithms to determine, whether DBP is an important risk factor to predict stroke, heart failure (HF), myocardial infarction (MI), and primary outcome in the SPRINT trial database. Methods: ML experiments were performed using decision tree, random forest, k-nearest neighbor, naive Bayesian, multi-layer perceptron, and logistic regression algorithms, including and excluding DBP as the risk factor in an unselected and selected (DBP < 70 mmHg) study population. Results: Including DBP as the risk factor did not change the performance of the machine learning models evaluated using accuracy, AUC, mean, and weighted F-measure, and was not required to make proper predictions of stroke, MI, HF, and primary outcome. Conclusions: Analyses of the SPRINT trial data using ML algorithms imply that DBP should not be treated as an independent risk factor when intensifying blood pressure control.

show abstract

Individualising intensive systolic blood pressure reduction in hypertension using computational trial phenomaps and machine learning: a post-hoc analysis of randomised clinical trials

Cited by 38 publications

References 35 publications

2023 ESH Guidelines for the management of arterial hypertension The Task Force for the management of arterial hypertension of the European Society of Hypertension

2023 ESH Guidelines for the management of arterial hypertension The Task Force for the management of arterial hypertension of the European Society of Hypertension

Derivation and Validation of a Machine Learning Approach to Detect and Mitigate Biases in Healthcare Data

Answering Clinical Questions Using Machine Learning: Should We Look at Diastolic Blood Pressure When Tailoring Blood Pressure Control?

Contact Info

Product

Resources

About