introduction
play

Introduction People with T1DM are usually in basal-bolus therapy - PowerPoint PPT Presentation

CBR BASED BOLUS RECOMMENDER SYSTEM Ferran Torrent-Fontbona Introduction People with T1DM are usually in basal-bolus therapy Timely and accurate insulin dosage avoids hyperglycaemia and its consequent complications and reduces the risk of


  1. CBR BASED BOLUS RECOMMENDER SYSTEM Ferran Torrent-Fontbona

  2. Introduction  People with T1DM are usually in basal-bolus therapy  Timely and accurate insulin dosage avoids hyperglycaemia and its consequent complications and reduces the risk of hypoglycaemia  Bolus calculators: – Available in market products: pumps, glucose meters, apps… – They have been proved useful at improving glycaemic self-control – Drawbacks: difficulty setting parameters, need to regularly adjust them… – Far from achieving optimal results June 24, 2017 2/16

  3. Objectives  Provide a method capable of: – Estimating the personalised bolus calculator parameters – Learning from past experiences to adapt to new situations – Providing personalised adaptive bolus recommendations CASE BASED REASONING June 24, 2017 3/16

  4. Case based reasoning  Lazy learning method  Propose new solutions using past experiences  Good results with small amounts of data Query case Retrieve Maintenance Reuse Case base Revise June 24, 2017 4/16

  5. Bolus recommender system  The CBR estimates the Insulin to Carbs Ratio (ICR) and Insulin Sensitivity Factor (ISF)  Then, it calculates the bolus dose New case: • Carbohydrates • Blood Glucose • Time of day Retrieve • Exercise • Stress • … Maintenance Reuse Case base Revise June 24, 2017 5/16

  6. Retrieve Objective: select similar past experiences  ISF and ICR depend on several factors: stress, time of day, menstruations, illnesses…  Not all factors have the same impact  Proposed retrieve consists of two steps: – Context reasoning (select the case base corresponding to the context) – Similarity measure and case retrieval June 24, 2017 6/16

  7. Reuse Objective: Adapt past solutions to the new case  Reuse ICR from retrieved cases – Weighted average according to the similarity  Calculate ISF using the ICR 𝐽𝑇𝐺 = 341.94 ∙ 𝐽𝐷𝑆 Walsh et al. (2011). Journal of Diabetes Science and Technology 𝑋  Calculate bolus dose 𝐽𝐷𝑆 + 𝐻 𝑑 − 𝐻 𝑡𝑞 𝐶 = 𝐷𝐼𝑃 − 𝐽𝑃𝐶 𝐽𝑇𝐺 𝐷𝐼𝑃 : carbs 𝐽𝑃𝐶 : insulin on board 𝐻 𝑑 : blood glucose level 𝑋 : body weight 𝐻 𝑡𝑞 : blood glucose target June 24, 2017 7/16

  8. Revise Objective: revise and repair the proposed solution  Revise: check minimum postprandial blood glucose and correct the recommended bolus (and ICR and ISF) to bring the value to the target one ෢ 𝐽𝐷𝑆 = 1 − 𝛽 𝐽𝐷𝑆 𝑠𝑓𝑣𝑡𝑓 + 𝛽𝐽𝐷𝑆 𝑑 𝐻 𝑑 𝐻 𝑛𝑗𝑜 Postprandial phase Meal time June 24, 2017 8/16

  9. Maintenance Objective: manage the case base to keep it updated and efficient  Concept drift problem  Proposed maintenance – Save the revise query case – If there are similar enough cases to the query case in the case base, then remove them June 24, 2017 9/16

  10. Experimentation  11 virtual adults using UVA/PADOVA simulator  Intra-day and physical activity variability have been added  50 simulations of 90-days  Comparison with a run-to-run algorithm June 24, 2017 10/16

  11. Results (without exercise) June 24, 2017 11/16

  12. Results (without exercise) June 24, 2017 12/16

  13. Results (with exercise) June 24, 2017 13/16

  14. Results (with exercise) June 24, 2017 14/16

  15. Conclusions  The proposed system: – Automatically estimates the personalised ICR and ISF – Is capable of adapting the parameters to new situations  Results are promising June 24, 2017 15/16

  16. T HANK YOU FOR YOUR ATTENTION ferran.torrent@udg.edu

  17. Results (summary) CBR R2R ( avg ± std) ( avg ± std) Without exercise In target (%) 86.62 ± 1.73 78.07 ± 6.01 Below target (%) 2.74 ± 0.85 7.05 ± 4.16 Above target (%) 10.6 3 ± 1.40 14.88 ± 2.68 With exercise In target (%) 82.51 ± 1.43 75.00 ± 4.93 Below target (%) 4.51 ± 1.29 8.41 ± 3.43 Above target (%) 12.98 ± 0.73 16.59 ± 2.26 June 24, 2017 17/16

  18. Future work  Automatically learn similarity measure weights  Similarity measure capable to deal with missing values  Adaptable learning rate June 24, 2017 18/16

  19. Results (I)  Without physical activity June 24, 2017 19/16

  20. Results (II)  With physical activity June 24, 2017 20/16

Recommend


More recommend