Publications by authors named "M Sharabiani"
Article Synopsis
- - The study focuses on improving the detection of familial hypercholesterolemia (FH), a common genetic disorder that is often underdiagnosed, which could help prevent heart-related issues linked to it.
- - Researchers used machine learning algorithms on data from the UK Biobank to identify individuals with FH-causing genetic variants, comparing the performance of these algorithms against existing clinical diagnostic criteria.
- - The findings showed that their machine learning model outperformed traditional diagnostic methods in identifying FH, with better sensitivity and accuracy, and reduced the number of individuals needed to be screened to find a case of FH.
Background: Electronic health records provide the opportunity to identify undiagnosed individuals likely to have a given disease using machine learning techniques, and who could then benefit from more medical screening and case finding, reducing the number needed to screen with convenience and healthcare cost savings. Ensemble machine learning models combining multiple prediction estimates into one are often said to provide better predictive performances than non-ensemble models. Yet, to our knowledge, no literature review summarises the use and performances of different types of ensemble machine learning models in the context of medical pre-screening.
View Article and Find Full Text PDFTo assess 20-year retrospective trajectories of cardio-metabolic factors preceding dementia diagnosis among people with type 2 diabetes (T2D). We identified 227,145 people with T2D aged > 42 years between 1999 and 2018. Annual mean levels of eight routinely measured cardio-metabolic factors were extracted from the Clinical Practice Research Datalink.
View Article and Find Full Text PDFBackground: Continuous vital sign monitoring may identify changes sooner than current standard monitoring.
Objective: To investigate if the use of real-time digital alerts sent to healthcare staff can improve the time taken to identify unwell patients and those with sepsis.
Design: A prospective cohort study design.
Objective: Pulmonary arterial hypertension (PAH) can lead to left main coronary artery compression (LMCo), but data on the impact, screening and treatment are limited. A meta-analysis of LMCo cases could fill the knowledge gaps in this topic.
Methods: Electronic databases were searched for all LMCo/PAH studies, abstracts and case reports including pulmonary artery (PA) size.