GlucoseDAO
Innovative decentralized science project focused on glucose prediction and health optimization, combining technology with community-driven healthcare solutions.
The project aims to develop an open-source glucose prediction system while building a collaborative community focused on health and longevity. Built on existing language models, working towards integrating RAG (Retrieval-Augmented Generation), hybrid search, and AI agent workflows. Developed using open-source libraries from Longevity Genie and Glucobench Also transformer model used is from Gluformer repo
- Glucose based repositories: All content is oriented to this particular approach, be it prediction model, sorting and processing data or presenting
- Hybrid Search Capabilities: Combines traditional search methods with advanced semantic search techniques.
- AI Agent Workflows: Utilizes orchestration logic to streamline interactions between LLMs and external tools.
- just-chat: Fork which on the branch glucosedao has the code for the chat deployment used in this project.
- sugar-sugar: The base for the sugar prediciton game.
- glucosedao: The base for the huggingface UI deployment.
- dia-linkeness: OAKvar module for finding diabetic related genes.
- glucobench: Fork for testing various prediction models.
We actually have a chat for people to ask questions directly about the organisation and project on this adress. The repo to do a similar chat for your project/organisation can be found on the Longevity Genie page, just-chat
Popular repositories Loading
-
-
cgm_format cgm_format Public
Converter from different popular CGM data export formats and datasets to a unified format suitable for ML training and inference
-
-
-
glucose_data_processing glucose_data_processing Public
Script to process glucose data for machine learning.
-
glucose-neuralforecast glucose-neuralforecast Public
repository to experiment with neuralforecast library to predict glucose
Python 2
Repositories
- sugar-data-processing Public
Statistical analysis pipeline for Sugar Sugar study results (H1-H5 per study design)
- glucose-forecasting Public
Multimodal glucose forecasting pipeline with GluMind, NHITS/TFT/NBEATSx baselines, and cross-cohort evaluation on AI-READI-style datasets.
- cgm_format Public
Converter from different popular CGM data export formats and datasets to a unified format suitable for ML training and inference
Top languages
Loading…
Most used topics
Loading…