The BGLP Challenge: Results, Papers, and Source Code


Results: Official Rankings

Of the 16 systems with papers that were accepted for publication, 8 systems had results that conformed to The BGLP Challenge Rules. These 8 systems were all evaluated using the exact same test points for each of the 6 data contributors in the OhioT1DM dataset. They are ranked based on their RMSE and MAE scores for the 30 minute and 60 minute prediction horizons. The 4 scores are added together to compute an overall score, and the 8 systems are ranked in increasing order of this total score. The overall ranking is shown in the first table below. The following tables show the rankings computed based on each of the 4 scores separately.

Official ranking (April 26, 2020).

 
 
RMSE at 30 minutes ranking.
 
MAE at 30 minutes ranking.
 
RMSE at 60 minutes ranking.
 
MAE at 60 minutes ranking.
 

Additional Unofficial Rankings

For the camera ready deadline (July 15, 2020), the authors of accepted papers had the opportunity to submit an updated set of results. Overall, 13 out of the 16 systems were evaluated using the exact same test points for each of the 6 data contributors in the OhioT1DM dataset, as specified in The BGLP Challenge Rules. The table below shows the overall ranking based on results reported in the camera ready version. Note that this ranking is for informational purpose only. The official rankings are the ones based on the test results submitted by the April 26, 2020, deadline.  

Ranking of camera ready results (July 15, 2020).

Papers and Source Code

The proceedings have been submitted for publication to CEUR. Paper titles will be linked to the papers once the proceedings are published. In the meantime, you can get a PDF file containing the full workshop proceedings here.

  1. Experiments in Non-Personalized Future Blood Glucose Level Prediction
    by Robert Bevan and Frans Coenen
    Source code
  2. Blood Glucose Level Prediction as Time-Series Modeling using Sequence-to-Sequence Neural Networks
    by Ananth Reddy Bhimireddy, Priyanshu Sinha, Bolu Oluwalade, Judy Wawira Gichoya and Saptarshi Purkayastha
    Source code
  3. A Personalized and Interpretable Deep Learning Based Approach to Predict Blood Glucose Concentration in Type 1 Diabetes
    by Giacomo Cappon, Lorenzo Meneghetti, Francesco Prendin, Jacopo Pavan, Giovanni Sparacino, Simone Del Favero and Andrea Facchinetti
    Source code
  4. Personalised Glucose Prediction via Deep Multitask Networks
    by John Daniels, Pau Herrero and Pantelis Georgiou
    Source code
  5. A Deep Learning Approach for Blood Glucose Prediction and Monitoring of Type 1 Diabetes Patients
    by Jonas Freiburghaus, Aïcha Rizzotti-Kaddouri and Fabrizio Albertetti
    Source code
  6. Investigating Potentials and Pitfalls of Knowledge Distillation across Datasets for Blood Glucose Forecasting
    by Hadia Hameed and Samantha Kleinberg
    Source code
  7. Analysis of the Performance of Genetic Programming on the Blood Glucose Level Prediction Challenge 2020
    by David Joedicke, Oscar Garnica, Gabriel Kronberger, José Manuel Colmenar, Stephan Winkler, Jose Manuel Velasco, Sergio Contador and Ignacio Hidalgo
    Source code
  8. Multi-lag Stacking Approach for Blood Glucose Level Prediction
    by Heydar Khadem, Hoda Nemat, Jackie Elliott and Mohammed Benaissa
    Source code
  9. Online Blood Glucose Prediction Using Autoregressive Moving Average Model with Residual Compensation Network
    by Ning Ma, Yuhang Zhao, Shuang Wen, Tao Yang, Ruikun Wu, Rui Tao, Xia Yu and Hongru Li
    Source code
  10. Neural Multi-Class Classification Approach to Blood Glucose Level Forecasting with Prediction Uncertainty Visualisation
    by Michael Mayo and Tomas Koutny
    Source code
  11. Data Fusion of Activity and CGM for Predicting Blood Glucose Level
    by Hoda Nemat, Heydar Khadem, Jackie Elliott and Mohammed Benaissa
    Source code
  12. Personalized Machine Learning Algorithm based on Shallow Network and Error Imputation Module for an Improved Blood Glucose Prediction
    by Jacopo Pavan, Francesco Prendin, Lorenzo Meneghetti, Giacomo Cappon, Giovanni Sparacino, Andrea Facchinetti and Simone Del Favero
    Source code
  13. Deep Residual Time-Series Forecasting: Application to Blood Glucose Prediction
    by Harry Rubin-Falcone, Ian Fox and Jenna Wiens
    Source code
  14. Prediction of Blood Glucose Levels for People with Type 1 Diabetes using Latent-Variable-based Model
    by Xiaoyu Sun, Mudassir Rashid, Mert Sevil, Nicole Hobbs, Rachel Brandt, Mohammad Reza Askari, Andrew Shahidehpour and Ali Cinar
    Source code
  15. Multi-Scale Long Short-Term Memory Network with Multi-Lag Structure for Blood Glucose Prediction
    by Tao Yang, Ruikun Wu, Rui Tao, Shuang Wen, Ning Ma, Yuhang Zhao, Xia Yu and Hongru Li
    Source code
  16. Blood Glucose Prediction for Type 1 Diabetes Using Generative Adversarial Networks
    by Taiyu Zhu, Xi Yao, Kezhi Li, Pau Herrero and Pantelis Georgiou
    Source code

Last updated: November 27, 2024.