AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review
AbstractObjective Human error estimating food intake is a major source of bias in nutrition research. Artificial intelligence (AI) methods may reduce bias, but the overall accuracy of AI estimates is unknown. This study was a systematic review of peer-reviewed journal articles comparing fully automa...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2023-12-01
|
| Series: | Annals of Medicine |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/07853890.2023.2273497 |
| _version_ | 1797302140756033536 |
|---|---|
| author | Eleanor Shonkoff Kelly Copeland Cara Xuechen (Anna) Pei Mei Chung Shreyas Kamath Karen Panetta Erin Hennessy |
| author_facet | Eleanor Shonkoff Kelly Copeland Cara Xuechen (Anna) Pei Mei Chung Shreyas Kamath Karen Panetta Erin Hennessy |
| author_sort | Eleanor Shonkoff |
| collection | DOAJ |
| description | AbstractObjective Human error estimating food intake is a major source of bias in nutrition research. Artificial intelligence (AI) methods may reduce bias, but the overall accuracy of AI estimates is unknown. This study was a systematic review of peer-reviewed journal articles comparing fully automated AI-based (e.g. deep learning) methods of dietary assessment from digital images to human assessors and ground truth (e.g. doubly labelled water).Materials and Methods Literature was searched through May 2023 in four electronic databases plus reference mining. Eligible articles reported AI estimated volume, energy, or nutrients. Independent investigators screened articles and extracted data. Potential sources of bias were documented in absence of an applicable risk of bias assessment tool.Results Database and hand searches identified 14,059 unique publications; fifty-two papers (studies) published from 2010 to 2023 were retained. For food detection and classification, 79% of papers used a convolutional neural network. Common ground truth sources were calculation using nutrient tables (51%) and weighed food (27%). Included papers varied widely in food image databases and results reported, so meta-analytic synthesis could not be conducted. Relative errors were extracted or calculated from 69% of papers. Average overall relative errors (AI vs. ground truth) ranged from 0.10% to 38.3% for calories and 0.09% to 33% for volume, suggesting similar performance. Ranges of relative error were lower when images had single/simple foods.Conclusions Relative errors for volume and calorie estimations suggest that AI methods align with – and have the potential to exceed – accuracy of human estimations. However, variability in food image databases and results reported prevented meta-analytic synthesis. The field can advance by testing AI architectures on a limited number of large-scale food image and nutrition databases that the field determines to be adequate for training and testing and by reporting accuracy of at least absolute and relative error for volume or calorie estimations. |
| first_indexed | 2024-03-07T23:32:31Z |
| format | Article |
| id | doaj.art-54b210240da54f8abde5d81cdb800df3 |
| institution | Directory Open Access Journal |
| issn | 0785-3890 1365-2060 |
| language | English |
| last_indexed | 2024-03-07T23:32:31Z |
| publishDate | 2023-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Annals of Medicine |
| spelling | doaj.art-54b210240da54f8abde5d81cdb800df32024-02-20T11:58:23ZengTaylor & Francis GroupAnnals of Medicine0785-38901365-20602023-12-0155210.1080/07853890.2023.2273497AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic reviewEleanor Shonkoff0Kelly Copeland Cara1Xuechen (Anna) Pei2Mei Chung3Shreyas Kamath4Karen Panetta5Erin Hennessy6School of Health Sciences, Merrimack College, North Andover, MA, USAFriedman School of Nutrition Science and Policy, Tufts University, Boston, MA, USAFriedman School of Nutrition Science and Policy, Tufts University, Boston, MA, USAFriedman School of Nutrition Science and Policy, Tufts University, Boston, MA, USASchool of Engineering, Tufts University, Medford, MA, USASchool of Engineering, Tufts University, Medford, MA, USAFriedman School of Nutrition Science and Policy, Tufts University, Boston, MA, USAAbstractObjective Human error estimating food intake is a major source of bias in nutrition research. Artificial intelligence (AI) methods may reduce bias, but the overall accuracy of AI estimates is unknown. This study was a systematic review of peer-reviewed journal articles comparing fully automated AI-based (e.g. deep learning) methods of dietary assessment from digital images to human assessors and ground truth (e.g. doubly labelled water).Materials and Methods Literature was searched through May 2023 in four electronic databases plus reference mining. Eligible articles reported AI estimated volume, energy, or nutrients. Independent investigators screened articles and extracted data. Potential sources of bias were documented in absence of an applicable risk of bias assessment tool.Results Database and hand searches identified 14,059 unique publications; fifty-two papers (studies) published from 2010 to 2023 were retained. For food detection and classification, 79% of papers used a convolutional neural network. Common ground truth sources were calculation using nutrient tables (51%) and weighed food (27%). Included papers varied widely in food image databases and results reported, so meta-analytic synthesis could not be conducted. Relative errors were extracted or calculated from 69% of papers. Average overall relative errors (AI vs. ground truth) ranged from 0.10% to 38.3% for calories and 0.09% to 33% for volume, suggesting similar performance. Ranges of relative error were lower when images had single/simple foods.Conclusions Relative errors for volume and calorie estimations suggest that AI methods align with – and have the potential to exceed – accuracy of human estimations. However, variability in food image databases and results reported prevented meta-analytic synthesis. The field can advance by testing AI architectures on a limited number of large-scale food image and nutrition databases that the field determines to be adequate for training and testing and by reporting accuracy of at least absolute and relative error for volume or calorie estimations.https://www.tandfonline.com/doi/10.1080/07853890.2023.2273497Artificial intelligencemachine learningnutrition assessmentnutrition surveysfood images |
| spellingShingle | Eleanor Shonkoff Kelly Copeland Cara Xuechen (Anna) Pei Mei Chung Shreyas Kamath Karen Panetta Erin Hennessy AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review Annals of Medicine Artificial intelligence machine learning nutrition assessment nutrition surveys food images |
| title | AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review |
| title_full | AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review |
| title_fullStr | AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review |
| title_full_unstemmed | AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review |
| title_short | AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review |
| title_sort | ai based digital image dietary assessment methods compared to humans and ground truth a systematic review |
| topic | Artificial intelligence machine learning nutrition assessment nutrition surveys food images |
| url | https://www.tandfonline.com/doi/10.1080/07853890.2023.2273497 |
| work_keys_str_mv | AT eleanorshonkoff aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview AT kellycopelandcara aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview AT xuechenannapei aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview AT meichung aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview AT shreyaskamath aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview AT karenpanetta aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview AT erinhennessy aibaseddigitalimagedietaryassessmentmethodscomparedtohumansandgroundtruthasystematicreview |