.png){kind=icon}
Return
Predictive Analytics in Type-1 Diabetes Using Machine Learning Algorithms
.png){kind=icon}
PRE
AI
DOI：10.1007/978-3-032-05545-3_14.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Type 1 diabetes is a lifelong metabolic disorder that demands prompt and accurate diagnosis to prevent serious health complications. This study examines the use of machine learning and deep learning methods to enhance the precision of Type 1 diabetes detection. The Diabetes Binary Health dataset sourced from Kaggle served as the foundation for this analysis. Feature selection was conducted using a Random Forest classifier, narrowing down the original 22 features to the 15 most impactful ones. Five models-AdaBoost, Random Forest, Neural Networks, Gradient Boosting, and k-Nearest Neighbors-were developed and systematically evaluated. Among them, the AdaBoost model demonstrated superior performance, achieving an AUC of 0.998 and an accuracy rate of 98.4%. These findings highlight the promising role of computational techniques in improving early diagnosis, ultimately supporting better patient management and contributing to the advancement of healthcare strategies for Type 1 diabetes.
Keywords:
Type 1 diabetes
machine learning
deep learning
AdaBoost
Random Forest
feature selection
early detection
medical diagnostics
Journal
P
IF:
0
Papers:
37
Citations:
0
Organization
Cited Papers
A Method to Detect Type 1 Diabetes Based on Physical Activity Measurements Using a Mobile Device
OAAI
OAAIShareSaveThe early detection of type 1 diabetes mellitus and latent autoimmune diabetes in adults (LADA) through rapid test reverse-flow immunochromatography for glutamic acid decarboxylase 65 kDa (GAD65).png)
HELIYONOAAI
HELIYONIF3.6
OAAIShareSave
ShareSaveShareSaveA Novel Proposal for Deep Learning-Based Diabetes Prediction: Converting Clinical Data to Image DataDIAGNOSTICSOAAI
DIAGNOSTICSIF3.3
OAAIShareSave
View more