Predicting misdiagnosed adult-onset type 1 diabetes using machine learning

It is now understood that almost half of newly diagnosed cases of type 1 diabetes are adult-onset. However, type 1 and type 2 diabetes are difficult to initially distinguish clinically in adults, potentially leading to ineffective care. In this study a machine learning model was developed to identif...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Diabetes research and clinical practice Ročník 191; s. 110029
Hlavní autoři:	Cheheltani, Rabee, King, Nicholas, Lee, Suyin, North, Benjamin, Kovarik, Danny, Evans-Molina, Carmella, Leavitt, Nadejda, Dutta, Sanjoy
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Ireland Elsevier B.V 01.09.2022
Témata:	Adult Adult-onset type 1 diabetes Blood Glucose Clinical decision support tool Diabetes Mellitus, Type 1 - diagnosis Diabetes Mellitus, Type 1 - epidemiology Diabetes Mellitus, Type 2 - diagnosis Diabetes Mellitus, Type 2 - drug therapy Diabetes Mellitus, Type 2 - epidemiology Diagnostic Errors Glycated Hemoglobin Humans Machine Learning Misdiagnosis Predictive algorithm Retrospective Studies Misdiagnosis AI Clinical decision support tool Adult-onset type 1 diabetes Predictive algorithm Machine learning
ISSN:	0168-8227, 1872-8227, 1872-8227
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	It is now understood that almost half of newly diagnosed cases of type 1 diabetes are adult-onset. However, type 1 and type 2 diabetes are difficult to initially distinguish clinically in adults, potentially leading to ineffective care. In this study a machine learning model was developed to identify type 1 diabetes patients misdiagnosed as type 2 diabetes. In this retrospective study, a machine learning model was developed to identify misdiagnosed type 1 diabetes patients from a population of patients with a prior type 2 diabetes diagnosis. Using Ambulatory Electronic Medical Records (AEMR), features capturing relevant information on age, demographics, risk factors, symptoms, treatments, procedures, vitals, or lab results were extracted from patients' medical history. The model identified age, BMI/weight, therapy history, and HbA1c/blood glucose values among top predictors of misdiagnosis. Model precision at low levels of recall (10 %) was 17 %, compared to <1 % incidence rate of misdiagnosis at the time of the first type 2 diabetes encounter in AEMR. This algorithm shows potential for being translated into screening guidelines or a clinical decision support tool embedded directly in an EMR system to reduce misdiagnosis of adult-onset type 1 diabetes and implement effective care at the outset.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	0168-8227 1872-8227 1872-8227
DOI:	10.1016/j.diabres.2022.110029