Short-term hypoglycemia classification using AI-models with DiaData

Cinar, Beyza; Maleshkova, Maria

doi:10.13140/RG.2.2.10172.60801

Short-term hypoglycemia classification using AI-models with DiaData

Publication date

2026-06

Document type

Wissenschaftliche Vortragsfolien, Präsentation

Author

Cinar, Beyza

Maleshkova, Maria

Organisational unit

Data Engineering

DOI

10.13140/RG.2.2.10172.60801

URI

https://openhsu.ub.hsu-hh.de/handle/10.24405/24208

Conference

Internal workshop

Publisher

ResearchGate

Part of the university bibliography

✅

Language

English

Description

This work presents the creation and preprocessing of DiaData, a large integrated dataset comprising over 149 million CGM values, and benchmarks deep learning (ResNet, FCN) and machine learning (XGBoost) models for classifying hypoglycemia up to 120 minutes before onset. Also, personalization, age-wise performance, and generalizability are investigated.

Version

Published version

Access right on openHSU

Metadata only access