Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Navigating Weight Prediction with Diet Diary

Yinxuan Gui, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yu-Gang Jiang

TL;DR

A novel task of weight prediction with a dietary diary that aims to leverage historical food intake and weight to predict future weights is proposed, and a model-agnostic time series forecasting framework is proposed to tackle this task.

Abstract

Current research in food analysis primarily concentrates on tasks such as food recognition, recipe retrieval and nutrition estimation from a single image. Nevertheless, there is a significant gap in exploring the impact of food intake on physiological indicators (e.g., weight) over time. This paper addresses this gap by introducing the DietDiary dataset, which encompasses daily dietary diaries and corresponding weight measurements of real users. Furthermore, we propose a novel task of weight prediction with a dietary diary that aims to leverage historical food intake and weight to predict future weights. To tackle this task, we propose a model-agnostic time series forecasting framework. Specifically, we introduce a Unified Meal Representation Learning (UMRL) module to extract representations for each meal. Additionally, we design a diet-aware loss function to associate food intake with weight variations. By conducting experiments on the DietDiary dataset with two state-of-the-art time series forecasting models, NLinear and iTransformer, we demonstrate that our proposed framework achieves superior performance compared to the original models. We make our dataset, code, and models publicly available at: https://yxg1005.github.io/weight-prediction/.

Navigating Weight Prediction with Diet Diary

TL;DR

A novel task of weight prediction with a dietary diary that aims to leverage historical food intake and weight to predict future weights is proposed, and a model-agnostic time series forecasting framework is proposed to tackle this task.

Abstract

Current research in food analysis primarily concentrates on tasks such as food recognition, recipe retrieval and nutrition estimation from a single image. Nevertheless, there is a significant gap in exploring the impact of food intake on physiological indicators (e.g., weight) over time. This paper addresses this gap by introducing the DietDiary dataset, which encompasses daily dietary diaries and corresponding weight measurements of real users. Furthermore, we propose a novel task of weight prediction with a dietary diary that aims to leverage historical food intake and weight to predict future weights. To tackle this task, we propose a model-agnostic time series forecasting framework. Specifically, we introduce a Unified Meal Representation Learning (UMRL) module to extract representations for each meal. Additionally, we design a diet-aware loss function to associate food intake with weight variations. By conducting experiments on the DietDiary dataset with two state-of-the-art time series forecasting models, NLinear and iTransformer, we demonstrate that our proposed framework achieves superior performance compared to the original models. We make our dataset, code, and models publicly available at: https://yxg1005.github.io/weight-prediction/.
Paper Structure (18 sections, 6 equations, 11 figures, 6 tables)

This paper contains 18 sections, 6 equations, 11 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Time Series Forecasting
  4. Food Analysis
  5. Dataset Construction
  6. DietDiary Dataset
  7. Food Images and Ingredient Annotation
  8. Method
  9. Task Definition
  10. Proposed Framework
  11. Unified Meal Representation Learning
  12. Experiment
  13. Experiment Settings
  14. Performance Comparison
  15. Multi-modal Fusion of Food Intake
  16. ...and 3 more sections

Figures (11)

  • Figure 1: The overview of the proposed weight prediction with diet diary task. Given a historical food intake and corresponding weight measurement in the past $L$ days, the task aims to predict the weights in the following $T$ days.
  • Figure 2: Distribution of the number of recording days for the participants in DietDiary. The y-axis is in the log scale.
  • Figure 3: Examples of food images from three meals in DietDiary.
  • Figure 4: Distribution of Daily Weight Change of all records in DietDiary.
  • Figure 5: Top 10 ingredients by occurrence frequency in DietDiary.
  • ...and 6 more figures