A Significance Assessment of Diabetes Diagnostic Biomarkers Using Machine Learning

2021 ◽  
Author(s):  
Ran Cui ◽  
Elena Daskalaki ◽  
Md Zakir Hossain ◽  
Artem Lenskiy ◽  
Christopher J Nolan ◽  
...  

Diabetes can be diagnosed by either Fasting Plasma Glucose or Hemoglobin A1c. The aim of our study was to explore the differences between the two criteria through the development of a machine learning based diabetes diagnostic algorithm and analysing the predictive contribution of each input biomarker. Our study concludes that fasting insulin is predictive of diabetes defined by FPG, but not by HbA1c. Besides, 28 other fasting blood biomarkers were not significant predictors of diabetes.

2014 ◽  
Vol 52 (3) ◽  
pp. 631-632 ◽  
Author(s):  
Jonas Blum ◽  
Stefanie Aeschbacher ◽  
Tobias Schoen ◽  
Matthias Bossard ◽  
Katrin Pumpol ◽  
...  

2012 ◽  
Vol 6 (3) ◽  
pp. 721-722
Author(s):  
B. Shivananda Nayak ◽  
Kameel Mungrue ◽  
Shawn Seepersad ◽  
Justin Sookram ◽  
Shalina Singh ◽  
...  

1990 ◽  
Vol 29 (2) ◽  
pp. 132-137 ◽  
Author(s):  
Hiroshi SOBAJIMA ◽  
Masahiko SHIKANO ◽  
Makoto TANIKAWA ◽  
Toyoo Niwa ◽  
Tetsuo HAYAKAWA

2012 ◽  
Vol 56 (7) ◽  
pp. 449-455 ◽  
Author(s):  
Joel C. Exebio ◽  
Gustavo G. Zarini ◽  
Joan A. Vaccaro ◽  
Cristobal Exebio ◽  
Fatma G. Huffman

OBJECTIVE: To evaluate the validity of hemoglobin A1C (A1C) as a diagnostic tool for type 2 diabetes and to determine the most appropriate A1C cutoff point for diagnosis in a sample of Haitian-Americans. SUBJECTS AND METHODS: Subjects (n = 128) were recruited from Miami-Dade and Broward counties, FL. Receiver operating characteristics (ROC) analysis was run in order to measure sensitivity and specificity of A1C for detecting diabetes at different cutoff points. RESULTS: The area under the ROC curve was 0.86 using fasting plasma glucose ≥ 7.0 mmol/L as the gold standard. An A1C cutoff point of 6.26% had sensitivity of 80% and specificity of 74%, whereas an A1C cutoff point of 6.50% (recommended by the American Diabetes Association - ADA) had sensitivity of 73% and specificity of 89%. CONCLUSIONS: A1C is a reliable alternative to fasting plasma glucose in detecting diabetes in this sample of Haitian-Americans. A cutoff point of 6.26% was the optimum value to detect type 2 diabetes.


2020 ◽  
Vol 67 (5) ◽  
pp. 509-513 ◽  
Author(s):  
Keiichi Torimoto ◽  
Yosuke Okada ◽  
Yoshiya Tanaka ◽  
Atsuko Matsuoka ◽  
Yushi Hirota ◽  
...  

Diabetes Care ◽  
2012 ◽  
Vol 36 (2) ◽  
pp. 429-435 ◽  
Author(s):  
J. B. Buse ◽  
F. R. Kaufman ◽  
B. Linder ◽  
K. Hirst ◽  
L. El ghormli ◽  
...  

2013 ◽  
Vol 34 (suppl 1) ◽  
pp. P4290-P4290
Author(s):  
J. Blum ◽  
S. Aeschbacher ◽  
T. Schoen ◽  
K. Pumpol ◽  
M. Risch ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document