Use of smartphones to estimate carbohydrates in foods for diabetes management Jurong HUANG, Hang DING, Simon MCBRIDE, David IRELAND, Mohan KARUNANITHI Presenter: Hang Ding | hang.ding@csiro.au | HIC 2015 3 August 2015 HEALTH AND BIOSECURITY
Prevalence of diabetes Adults with 380 million diabetes worldwide Deaths directly 1.5 million caused by diabetes 2 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Traditional estimation of carbohydrate 3 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Smartphone Approach Food Classifier OpenCV Volume Camera Estimator Nutrition OS Database Carbohydrate Calculation 4 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Food Classifier Three Features Texture Shape Colour (scale Invariant Feature Transform) (Local Binary Pattern) (RGB elements) Support Vector Machine 5 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Volume Estimator Object with calibrated size Food Photo Objects extracted Estimated Volume 6 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Evaluation of Classification 10 types of fruits, 60 photos each (orange, apple, pear, tomato, strawberry, banana, mango, avocado, pineapple, and kiwi fruit) Randomized Training Data Test Data 10 types, 50 photos each 10 types, 10 photos each ACC = (TP + TN) Optimized Classification Accuracy of TP + TN + FP + FN Parameters Classification 7 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Accuracy of Classification Classification Accuracy Types of Tested Fruits 8 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Accuracy of Carbo estimation Table 1. Summary of the volume and carbohydrate estimations, compared with the actual values measured from the water displacement and weight scale. Test Item Model Actual Error Rate Estimated Actual Carbs Error Rate Volume Volume (%) Carbs (g) (g) (%) (ml) (ml) Peach 158 151 4.43 16.3 15.9 2.45 Apple1 165 173 4.85 18.5 21.3 15.1 Apple2 172 190 13.9 19.3 22.4 16.1 Apple3 201 198 1.49 22.5 23.7 3.56 Tomato1 21 22 4.76 0.74 0.78 5.41 Tomato2 17 19 11.7 0.62 0.66 6.45 Average Error 6.86 8.18 9 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Future work • Improvement of the approach • Combination with other techniques • Evaluation through clinical studies 10 | Use of smartphones to estimate carbohydrates in foods for diabetes management | Hang Ding, 3 August 2015
Prof. Len Gray Mr. Jurong Huang Dr. David Hansen Prof. Anthony Russell Dr. Mohan Karunanithi Ms. Denise Bennetts Dr. Simon McBride Ms. Dominique Bird Dr. David Ireland Dr. Farhad Fatehi Thank you Contact Us Phone: 1300 363 400 or +61 3 9545 2176 Email: enquiries@csiro.au Web: www.csiro.au
Recommend
More recommend