An Analysis of Predicting Diabetes using Machine Learning

Mr. Ujjwal Anand, Dr. Amit Sehgal, Mr. Shashank Tripathi, Mr. Gagandeep Singh, Mr. Rinku Sharma, Ms. Manisha

Abstract


Diabetes comes under chronic disease, in which cells are not able to use blood sugar (glucose) efficient enough for energy. This condition arrives when the cells become non-responsive to insulin and the blood sugar increases gradually. The known types of diabetes are, Type1, Type 2 and Type 3. Type 1 and type 2 diabetes are in hyperglycemia category (caused by increase in blood sugar), while Type 3 diabetes (Alzheimer's disease) which is caused by resistance to insulin in the brain. Prediction of preliminary stage diabetes is very important as it becomes worse in next stages. This Prediction can be done using machine learning classification models, which are more widely being used for other medical purposes. To predict, if a person is diabetic we need data about insulin, blood pressure, skin thickness and glucose. This data will be fitted in classification models of machine learning with a target vector of conclusion. This will prepare a model that can predict if a certain patient is a diabetic or not. We have implemented many classification models on the data. We have also used neural networks to serve the same purpose.


Full Text:

PDF

References


HLA Nomenclature in WMDA file format–details of current HLA alleles and where known their unambiguous, possible or assumed serologically equivalent antigens. http://hla.alleles.org/wmda/rel_dna_ser.txt. Accessed April 13, 20164

International Diabetes Federation. Diabetes Atlas. 5th ed. Brussels, Belgium: IDF Publications. (2011) The Global Burden of Diabetes; pp. 7–13. Available from http://www.idf.org/diabetesatlas/news/fifth-edition-release. Accessed 25 May2015

Georgiev, D., Houdová, L., Fetter, M., Jindra, P.: A Scalable Method for Efficient Stem Cell Donor HLA Genotype Match Determination. In: Energy, Environment, Biology and Biomedicine, Proceedings of the 2014 International Conference on Biology and Biomedicine II, pp. 28-32. INASE, Prague (2014)3

Hayuhardhika W, Putra N, Sugiyanto, Sarno R, Sidiq M (2013) Weighted Ontology and Weighted Tree Similarity Algorithm for Diagnosing Diabetes Mellitus. IEEE International Conference on Computer, Control, Informatics and Its Applications pp.267-272.

Chen R, Huang Y, Bau C, Chen S (2012) A recommendation system based on domain ontology and SWRL for anti-diabetic drugs selection. Expert Systems with Applications 39: 3995–4006

NMDP allele codes. https://bioinformatics.bethematchclinical.org/hla-resources/ allele- codes/. Accessed April 13,2016

A. S. Abdelmoneim, D. T. Eurich, J.-M. Gamble, and S. H. Simpson, “Use patterns of antidiabetic regimens by patients with type 2 diabetes,” Canadian Journal of Diabetes, vol. 37, no. 6, pp. 394–400, 2013.

World Marrow Donor Association International Standards for Unrelated Hematopoietic Stem Cell Donor registries (2014), https://www.wmda.info/images/ pdf/20140101-STDC- WMDA_Standards_New_Housestyle.pdf. Accessed April 7,2016

Y. Handelsman, Z. T. Bloomgarden, G. Grunberger et al., “American Association of Clinical Endocrinologists and American College of Endocrinology - clinical practice guidelines for developing a diabetes mellitus comprehensive care plan - 2015,” Endocrine Practice, vol. 21, Supplement 1, pp. 1–87,2015


Refbacks

  • There are currently no refbacks.