Machine learning for classification of hypertension subtypes using multi-omics: A multi-centre, retrospective, data-driven study

Reel, Parminder Singh; Reel, Smarti; Kralingen, Josie C van; Langton, Katharina; Lang, Katharina; Erlic, Zoran; Larsen, Casper K.; Amar, Laurence; Pamporaki, Christina; Mulatero, Paolo; Blanchard, Anne; Kabat, Marek; Robertson, Stephen M.; MacKenzie, Scott; Peitzsch, Mirko; Ceccato, Filippo; Scaroni, Carla; Reincke, Martín; Kroiß, Matthias; Dennedy, Michael Conall; Pecori, Alessio; Monticone, Silvia; Deinum, Jaap; Rossi, Gian Paolo; Lenzini, Livia; McClure, John; Nind, Thomas; Riddell, Alexandra; Stell, Anthony; Callebaut, Christian; Sudano, Isabella; Prehn, Cornelia; Adamski, Jerzy; Gimenez‐Roqueplo, Anne‐Paule; Assié, Guillaume; Arlt, Wiebke; Beuschlein, Felix; Eisenhofer, Graeme; Davies, Eleanor; Zennaro, Maria‐Christina; Jefferson, Emily

doi:10.1016/j.ebiom.2022.104276

Cited by 36 publications

(17 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Continuing medical education programs have a role by targeting all healthcare professionals involved in the care of pregnancy; and decision support tools are gaining traction for highlighting the possibility of a rare disease in otherwise common clinical scenarios (e.g. hypertension in pregnancy) [ 42 ]. Patient advocacy is also crucial, particularly in the setting of a women with prior history of PPGL or hereditary PPGL syndrome.…”

Section: Discussionmentioning

confidence: 99%

The diagnosis and management of pheochromocytoma and paraganglioma during pregnancy

Clifton‐Bligh

2023

Rev Endocr Metab Disord

View full text Add to dashboard Cite

Diagnosis of pheochromocytoma or paraganglioma (PPGL) in pregnancy has been associated historically with high rates of materno-fetal morbidity and mortality. Recent evidence suggests outcomes are improved by recognition of PPGL before or during pregnancy and appropriate medical management with alpha-blockade. Whether antepartum surgery (before the third trimester) is required remains controversial and open to case-based merits. Women with PPGL in pregnancy are more commonly delivered by Caesarean section, although vaginal delivery appears to be safe in selected cases. At least some PPGLs express the luteinizing hormone/chorionic gonadotropin receptor (LHCGR) which may explain their dramatic manifestation in pregnancy. PPGLs in pregnancy are often associated with heritable syndromes, and genetic counselling and testing should be offered routinely in this setting. Since optimal outcomes are only achieved by early recognition of PPGL in (or ideally before) pregnancy, it is incumbent for clinicians to be aware of this diagnosis in a pregnant woman with hypertension occurring before 20 weeks’ gestation, and acute and/or refractory hypertension particularly if paroxysmal and accompanied by sweating, palpitations and/or headaches. All women with a past history of PPGL and/or heritable PPGL syndrome should be carefully assessed for the presence of residual or recurrent disease before considering pregnancy.

show abstract

Section: Discussionmentioning

confidence: 99%

The diagnosis and management of pheochromocytoma and paraganglioma during pregnancy

Clifton‐Bligh

2023

Rev Endocr Metab Disord

View full text Add to dashboard Cite

show abstract

“…Random Forest, as the leading class of machine-learning algorithms, has shown high accuracy and low overfitting risk in diverse biological analyses, especially for the high-dimensionality datasets, such as multi-omics data [ 37 , 38 , 39 ]. In RF analysis, a VIM value is generated for each variable accessed by RF, the higher the ‘VIM’ value is, the more important the variable is for generation a prediction in the decision trees [ 23 ].…”

Section: Discussionmentioning

confidence: 99%

Integration of the Microbiome, Metabolome and Transcriptome Reveals Escherichia coli F17 Susceptibility of Sheep

Chen

Cao

et al. 2023

Animals

View full text Add to dashboard Cite

Escherichia coli (E. coli) F17 is one of the most common pathogens causing diarrhea in farm livestock. In the previous study, we accessed the transcriptomic and microbiomic profile of E. coli F17-antagonism (AN) and -sensitive (SE) lambs; however, the biological mechanism underlying E. coli F17 infection has not been fully elucidated. Therefore, the present study first analyzed the metabolite data obtained with UHPLC-MS/MS. A total of 1957 metabolites were profiled in the present study, and 11 differential metabolites were identified between E. coli F17 AN and SE lambs (i.e., FAHFAs and propionylcarnitine). Functional enrichment analyses showed that most of the identified metabolites were related to the lipid metabolism. Then, we presented a machine-learning approach (Random Forest) to integrate the microbiome, metabolome and transcriptome data, which identified subsets of potential biomarkers for E. coli F17 infection (i.e., GlcADG 18:0-18:2, ethylmalonic acid and FBLIM1); furthermore, the PCCs were calculated and the interaction network was constructed to gain insight into the crosstalk between the genes, metabolites and bacteria in E. coli F17 AN/SE lambs. By combing classic statistical approaches and a machine-learning approach, our results revealed subsets of metabolites, genes and bacteria that could be potentially developed as candidate biomarkers for E. coli F17 infection in lambs.

show abstract

“…Accordingly, AI technology should be implemented cautiously; to be a partner of clinicians, there is still a long way to go, but it can serve as a virtual assistant and enable clinicians to promote quality and increase efficiency. Based on EMRs from Fuwai Hospital, five ML prediction models with good performance and applicability to the etiology detection of secondary hypertension were developed by Campo [ 95 ], which demonstrated that ML approaches were feasible and effective in diagnosing secondary hypertension. Reel and colleagues [ 95 ] showed that the MOmics approach provided better discriminatory power compared to single-omics (monoomics) data analysis and appropriately classified different forms of endocrine hypertension with high sensitivity and specificity, providing potential diagnostic biomarker combinations for diagnosing secondary hypertension subtypes.…”

Section: Ai In Secondary Arterial Hypertensionmentioning

confidence: 99%

“…Based on EMRs from Fuwai Hospital, five ML prediction models with good performance and applicability to the etiology detection of secondary hypertension were developed by Campo [ 95 ], which demonstrated that ML approaches were feasible and effective in diagnosing secondary hypertension. Reel and colleagues [ 95 ] showed that the MOmics approach provided better discriminatory power compared to single-omics (monoomics) data analysis and appropriately classified different forms of endocrine hypertension with high sensitivity and specificity, providing potential diagnostic biomarker combinations for diagnosing secondary hypertension subtypes. However, there still needs to be more data in the literature on the application of AI in the field of secondary hypertension; consequently, these innovative and clinically relevant prediction models still require further validation and more clinical tests before being implemented into clinical practice.…”

Section: Ai In Secondary Arterial Hypertensionmentioning

confidence: 99%

Artificial Intelligence in Hypertension Management: An Ace up Your Sleeve

Visco

Izzo

Mancusi

et al. 2023

JCDD

View full text Add to dashboard Cite

Arterial hypertension (AH) is a progressive issue that grows in importance with the increased average age of the world population. The potential role of artificial intelligence (AI) in its prevention and treatment is firmly recognized. Indeed, AI application allows personalized medicine and tailored treatment for each patient. Specifically, this article reviews the benefits of AI in AH management, pointing out diagnostic and therapeutic improvements without ignoring the limitations of this innovative scientific approach. Consequently, we conducted a detailed search on AI applications in AH: the articles (quantitative and qualitative) reviewed in this paper were obtained by searching journal databases such as PubMed and subject-specific professional websites, including Google Scholar. The search terms included artificial intelligence, artificial neural network, deep learning, machine learning, big data, arterial hypertension, blood pressure, blood pressure measurement, cardiovascular disease, and personalized medicine. Specifically, AI-based systems could help continuously monitor BP using wearable technologies; in particular, BP can be estimated from a photoplethysmograph (PPG) signal obtained from a smartphone or a smartwatch using DL. Furthermore, thanks to ML algorithms, it is possible to identify new hypertension genes for the early diagnosis of AH and the prevention of complications. Moreover, integrating AI with omics-based technologies will lead to the definition of the trajectory of the hypertensive patient and the use of the most appropriate drug. However, AI is not free from technical issues and biases, such as over/underfitting, the “black-box” nature of many ML algorithms, and patient data privacy. In conclusion, AI-based systems will change clinical practice for AH by identifying patient trajectories for new, personalized care plans and predicting patients’ risks and necessary therapy adjustments due to changes in disease progression and/or therapy response.

show abstract

Machine learning for classification of hypertension subtypes using multi-omics: A multi-centre, retrospective, data-driven study

Cited by 36 publications

References 42 publications

The diagnosis and management of pheochromocytoma and paraganglioma during pregnancy

The diagnosis and management of pheochromocytoma and paraganglioma during pregnancy

Integration of the Microbiome, Metabolome and Transcriptome Reveals Escherichia coli F17 Susceptibility of Sheep

Artificial Intelligence in Hypertension Management: An Ace up Your Sleeve

Contact Info

Product

Resources

About