SAM Photo Diagnosis App Project
ISRCTN | ISRCTN71157571 |
---|---|
DOI | https://doi.org/10.1186/ISRCTN71157571 |
Secondary identifying numbers | ES/SAMphotoD/01 |
- Submission date
- 24/09/2018
- Registration date
- 29/10/2018
- Last edited
- 09/09/2020
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Nutritional, Metabolic, Endocrine
Plain English Summary
Background and study aims
Action against Hunger (ACH) is an international humanitarian organization that fights against the causes and effects of hunger. We save lives of malnourished children and guarantee families access to safe water, food, training and basic health care. The SAM Photo Diagnosis App® project has developed an innovative easy-to-use smartphone tool which can quick and accurately diagnose nutritional status in children aged 6-59 months by means of a photo (body shape analysis). Our innovation responds to the need to improve screening, diagnosis and management of malnutrition, providing a unique opportunity to address it at scale at community level.
Despite significant progress over recent decades, maternal and child undernutrition remains a significant problem throughout the developing world, with 45% of deaths of children younger than five years old attributed to malnutrition. Acute malnutrition (wasting) in children is the life-threatening result of hunger and/or disease. Children suffering from acute malnutrition face an increased risk of morbidity and mortality, and therefore require urgent treatment to survive. In 2016, over 52 million children under five years of age suffered from acute malnutrition, with almost 17 million of these children experiencing the most severe form. Only 3.2 million children were successfully diagnosed and treated in 2016, leaving over 80% of children in need but unable to access treatment. It is important that health centres are equipped with the tools and expertise they need to be able to diagnose and manage cases of acute malnutrition. By the same token, it’s important that caretakers are also empowered to continue the treatment and regularly monitor the condition until the child is fully recovered.
Historically, two methods of anthropological measurement have been used for the diagnosis of moderate and severe acute malnutrition – weight-for-height (WHZ) and mid-upper arm circumference (MUAC). While these techniques provide an accurate diagnosis of acute malnutrition, there are still significant operational challenges for their use at scale. We have demonstrated that image-based techniques can be applied as an alternative diagnosis approach.
This project reinforces decentralization and sustainability of community management of malnutrition by empowering caretakers to diagnose and monitor the nutrition status of their children. It also presents a great opportunity for governments and local health systems to revolutionize nutrition assessments and improve the impact and coverage of nutrition programs.
Who can participate?
Children aged 6-59 months with severe acute malnutrition
What does the study involve?
Phase 1 is the pilot phase. This phase is complete and involved developing an innovative smartphone tool which could quick and accurately diagnose nutritional status, by analysing the child’s body shape registered in photos taken by the device. The development of the algorithm required taking photographs of a sample of well-nourished children in Spain and well-nourished and malnourished children in Senegal. The sample was composed of 107 Spanish children, 154 Senegalese well-nourished children and 142 Senegalese children with severe acute malnutrition. Overall, the Phase 1 project collected over 2,060 images across the sample of children in the study. These images captured the full body photos in different views and allowed the study of both the whole body and split body regions. These images provided the information needed to generate the templates used for diagnosis.
Phase 2 involves adapting the technology to the users. Here the aim is to reinforce both dimensions of the App as a research and diagnostic tool. At conception, it was recognised that a second phase of the project would be needed to further develop the tool and expand its use to different contexts and explore further use of the tool at scale. Thus, as both the programming of the App and the first preliminary results were being generated, new research questions and concerns were arising and opening-up potential future App functionality and usability paths to be developed.
Phase 3 will be a scaling up phase and is currently under development.
What are the possible benefits and risks of participating?
Both Spanish and Senegalese children receive a personalised report (written for Spanish children and both written and/or oral for Senegalese children) regarding their nutritional status, as well as important information related to health and nutrition care practices. All malnutrition diagnosis (both under and overnutrition) are reported directly to the families, informing them of next steps. In Senegal, when a MAM or SAM child is identified, they are immediately referred to the health system. All participants families are informed of the associated technology that was going to be created and the important role they were carrying out for that population in risk or suffering malnutrition by allowing their children to participate. They are also informed that such technology would be available for free in the near future for everyone who wishes to use it.
However, the possible risk is that children do not always wish to participate, even if their parents do, especially babies. Children may get tired during the sampling process.
Where is the study run from?
Action Against Hunger (Spain)
When is the study starting and how long is it expected to run for?
November 2015 to December 2021
Who is funding the study?
1. Children's Investment Fund Foundation (UK)
2. World Food Programme (Italy)
3. Grand Challenges Canada (GCC, Canada)
4. Agencia Española de Cooperación (AECID, Spain)
Who is the main contact?
Laura Medialdea Marcos, Principal Researcher
lmedialdea@accioncontraelhambre.org
Contact information
Scientific
C/ Duque de Sevilla, 3
Madrid
28003
Spain
Phone | +34913915300, ext. 1345 |
---|---|
lmedialdea@achesp.org |
Study information
Study design | Observational epidemiological study with pilot phase and validation phase |
---|---|
Primary study design | Observational |
Secondary study design | Epidemiological study |
Study setting(s) | Other |
Study type | Diagnostic |
Participant information sheet | ISRCTN71157571_PIS_18Oct2018_FRENCH.pdf |
Scientific title | Severe Acute Malnutrition (SAM) Photo Diagnosis Application Project for nutritional assesment |
Study acronym | SAMphotoD |
Study hypothesis | The starting hypothesis this project was that it is possible to develop an innovative smartphone tool which can quick and accurately diagnose nutritional status, by analysing morphometrically the child’s body shape, registered in photos taken by the device. Once this hypothesis was verified, a second phase of the project was designed in order to answer the following hypotheses: 1. Shape variations between this two-extreme nutritional status are also measurable if considering Moderate Acute Malnutrition (MAM) and underweight 2. Children’s body shape with SAM present homogeneous growth and body shape pattern among populations with different contexts and culture 3. Chronic malnutrition approach is possible using SAM Photo Diagnosis® methodology 4. A user centred design is possible for addressing community management of malnutrition |
Ethics approval(s) | 1. Clinical Research Ethical Committee, San Carlos Clinical Hospital, Madrid, Spain, 15/01/2016, reference number: C.P. CMAM 2.0 - C.I. 16/013-E 2. National Ethics Committee for Health Research, Dakar, Senegal, 23/11/2016, reference number: SEN16/35 |
Condition | Acute and chronic malnutrition in children aged 6-59 months |
Intervention | Current intervention as of 07/09/2020: Phase I of this study was the pilot phase, with the following objectives: 1. Validate geometric morphometric (GM) techniques for use in children under 5 years of age 2. Validate the use of a morphometric model to diagnose severe acute malnutrition (SAM) in a high-prevalence context 3. Document the lessons and knowledge on the application and its use as a diagnostic tool to inform future operational and research use The development of the algorithm required taking photographs of a sample of well-nourished children in Spain and well-nourished and malnourished children in Senegal. Overall, 2,060 images were collected across the sample of 450 children in the study. These images captured the full body photos in different views and allowed the study of both the whole body and split body regions. These images provided the information needed to generate the templates used for diagnosis. According to bioethical considerations as well as personal data protection policies (Senegalese 2008/12 and European GDPR 16/679), all health data, including images, were treated under a high-level data protection framework. Furthermore, required ethical clearances and consent forms were obtained prior to data collection. Regarding the app development, since the images of body regions are used to diagnose malnutrition with SAM Photo Diagnosis App®, children will not be identifiable from the photos taken and all images will be deleted from mobiledevices once the diagnosis has been provided. Phase II is the validation phase, with the following objectives: 1. Technical validation of SAM Photo Diagnosis App in two selected pilot countries, Senegal and Guatemala 2. Integration of machine learning technology to allow automated diagnosis of malnutrition 3. Validation of a user-centered design approach in pilot countries Pictures of body regions are being taken with SAM Photo Diagnosis App®, applying the same ethical considerations used in Phase I. Surveys are being conducted to legal guardians of participants to assess nutritional, familiar and context information related to undernutrition and provide useful information for final users. Focal groups are being constructed to carry out workshops and get useful feedbacks to develop the app's interface. Phase III is the scaling up phase is still under ideation. One of the main objectives in this stage is to create a platform where SAM Photo Diagnosis App® is integrated with national health systems to provide a holistic community-based management of malnutrition. _____ Previous intervention: Phase I of this study was the pilot phase, with the following objectives: 1. Validate geometric morphometric (GM) techniques for use in children under 5 years of age 2. Validate the use of a morphometric model to diagnose severe acute malnutrition (SAM) is a high-prevalence context 3. Document the lessons and knowledge on the application and its use as a diagnostic tool to inform future operational and research use The development of the algorithm required taking photographs of a sample of well-nourished children in Spain and well-nourished and malnourished children in Senegal. Overall, 2,060 images were collected across the sample of children in the study. These images captured the full body photos in different views and allowed the study of both the whole body and split body regions. These images provided the information needed to generate the templates used for diagnosis. According to bioethical considerations as well as personal data protection policies (Senegalese 2008/12 and European GDPR 16/679), all health data, including images, were treated under a high-level data protection framework. Furthermore, required ethical clearances and consent forms were obtained prior to data collection. Regarding the app development, since the images of body regions are used to diagnose malnutrition with SAM Photo Diagnosis App®, children will not be identifiable from the photos taken and all images will be deleted from mobile devices once the diagnosis has been provided. Phase II is the validation phase, with the following objectives: 1. Technical validation of SAM Photo Diagnosis App in two selected pilot countries, Senegal and the Democratic Republic of the Congo 2. Creation of a platform where SAM Photo Diagnosis App(R) is integrated integration for holistic community-based management of malnutrition 3. Integration of machine learning technology to allow automated diagnosis of malnutrition 4. Validation of the approach in pilot countries Pictures of body regions were taken with SAM Photo Diagnosis App(R), and the same ethical considerations applied in Phase I were applied to Phase II. Surveys were conducted to assess nutritional, familiar and context information related to undernutrition and provide useful information for final users. Phase III is the scaling up phase and has not yet been designed. |
Intervention type | Device |
Pharmaceutical study type(s) | |
Phase | |
Drug / device / biological / vaccine name(s) | |
Primary outcome measure | Current primary outcome measures as of 07/09/2020: 1. Anthropometric measurements: 1.1. Weight (kg), assessed using a portable electronic scale (phases I & II) 1.2. Height (cm), assessed using (phases I & II): 1.2.1. Portable infantometer for children under 87 cm 1.2.2. Anthropometer GPM for children over 87 cm 1.3. Mid-upper arm circumference (MUAC, cm), assessed using a self-rectracting flat metal tape with a blank lead-in strip (phases I & II) 1.4. Sitting height (cm), assessed using (phase I): 1.4.1. Portable infantometer for children under 87 cm 1.4.2. Anthropometer GPM for children over 87 cm 1.5. Cephalic perimeter (cm), assessed using a self-rectracting flat metal tape with a blank lead-in strip (phase I) 2. Determination of which photographic angles would provide better image information regarding body shape, assessed through anatomically identifying and marking 68 points on the child's body and analysing the images (phase I) 3. Information regarding the health, nutrition and origin of the child participating in the study, assessed using a survey completed by the parents (phases I & II) 4. Feedback obtained in user-centered design activities carried out with focal groups (phase II) _____ Previous primary outcome measures: 1. Anthropometric measurements, taken during phase I of the study: 1.1. Weight (kg), assessed using a portable electronic scale 1.2. Height (cm), assessed using: 1.2.1. Portable infantometer for children under 87 cm 1.2.2. Anthropometer GPM for children over 87 cm 1.3. Mid-upper arm circumference (cm), assessed using a self-rectracting flat metal tape with a blank lead-in strip 1.4. Sitting height (cm), assessed using: 1.4.1. Portable infantometer for children under 87 cm 1.4.2. Anthropometer GPM for children over 87 cm 1.5. Cephalic perimeter (cm), assessed using a self-rectracting flat metal tape with a blank lead-in strip 2. Determination of which photographic angles would provide better image information regarding body shape, assessed through anatomically identifying and marking 68 points on the child's body and analysing the images. This was done during phase I of the study 3. Information regarding the health, nutrition and origin of the child participating in the study, assessed using a survey completed by the parents at the point of photographs being taken (during phase I) |
Secondary outcome measures | N/A |
Overall study start date | 16/11/2015 |
Overall study end date | 31/12/2021 |
Eligibility
Participant type(s) | Mixed |
---|---|
Age group | Child |
Lower age limit | 6 Months |
Upper age limit | 59 Months |
Sex | Both |
Target number of participants | Phase I: n= 450. 2 countries (150 Spain, 300 Senegal), nutritional status (normalweight: 150 Spanish, 150 Senegalese; SAM: 150 Senegalese). Phase II: n=1800 (1200 Senegal, 600 Guatemala),Senegal= 4 nutritional conditions decribed by WHO (25% each), Guatemala= wasting (presence/absence) + stunting (presence/absence) equally distributed. Phases I & II: 2 sexes (50% each group), 2 age groups (50% each group) |
Participant inclusion criteria | Current participant inclusion criteria as of 07/09/2020: Phase I 1. Aged 6-59 months 2. Normal weight: p30 - p70 weight-for-height Z-score (WHZ) or mid-upper arm circumference Z-score (MUACZ) 3. Severe acute malnourishment (WHZ and MUACZ <-3 SD Phase II 1. Aged 6-59 months 2. Any nutritional condition estimated by WHZ and MUACZ following WHO criteria _____ Previous participant inclusion criteria: 1. Aged 6-59 months 2. Normal weight: p30 - p70 weight-for-height Z-score (WHZ) or mid-upper arm circumference Z-score (MUACZ) 3. Severe acute malnourishment (WHZ and MUACZ <-3 SD |
Participant exclusion criteria | Current participant exclusion criteria as of 07/09/2020: 1. No signed consent 2. Billateral oedema. 3. Unwilling to participate 4. Health complications other than malnutrition 5. Morphological abnormalities _____ Previous participant exclusion criteria: 1. No signed consent 2. Unwilling to participate 3. Health complications 4. Morphological abnormalities |
Recruitment start date | 01/02/2016 |
Recruitment end date | 31/12/2020 |
Locations
Countries of recruitment
- Guatemala
- Senegal
- Spain
Study participating centres
Madrid
28003
Spain
Madrid
28043
Spain
Madrid
28044
Spain
Madrid
28034
Spain
Madrid
28005
Spain
Madrid
28027
Spain
Madrid
28025
Spain
Madrid
28045
Spain
Madrid
28047
Spain
Madrid
28031
Spain
Madrid
28002
Spain
Madrid
28040
Spain
Matam
Diela
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Matam
N/A
Senegal
Madrid
28019
Spain
Madrid
28045
Spain
Sponsor information
Charity
7 Clifford Street
London
W1S 2FT
United Kingdom
Website | https://ciff.org/ |
---|---|
https://ror.org/00jfgrn87 |
Funders
Funder type
Charity
Private sector organisation / Trusts, charities, foundations (both public and private)
- Alternative name(s)
- CIFF
- Location
- United Kingdom
No information available
Government organisation / National government
- Alternative name(s)
- Grands Défis Canada, GCC
- Location
- Canada
No information available
Results and Publications
Intention to publish date | 31/12/2020 |
---|---|
Individual participant data (IPD) Intention to share | No |
IPD sharing plan summary | Data sharing statement to be made available at a later date |
Publication and dissemination plan | Current publication and dissemination plan as of 07/09/2020: We have produced a promotional video of the project (https://www.youtube.com/watch?v=y0VlLxYKpZY&feature=youtu.be) and a general diffusion article (https://www.ennonline.net//fex/57/samphotodiagnosisapp). A manuscript is under review with the American Journal of Physical Anthropology. We aim to prepare another two manuscripts to be sent to peer-reviewed journals in December 2020 and March 2021 Previous publication and dissemination plan: We have produced a promotional video of the project (https://www.youtube.com/watch?v=y0VlLxYKpZY&feature=youtu.be) and a general diffusion article (https://www.ennonline.net//fex/57/samphotodiagnosisapp). A manuscript is under review with the American Journal of Physical Anthropology. We aim to prepare another two manuscripts to be sent to peer-reviewed journals in December 2018 and March 2019 |
IPD sharing plan | The data sharing plans for the current study are unknown and will be made available at a later date |
Study outputs
Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
---|---|---|---|---|---|
Participant information sheet | 18/10/2018 | 02/04/2019 | No | Yes | |
Participant information sheet | 18/10/2018 | 02/04/2019 | No | Yes | |
Results article | results | 01/04/2019 | 09/07/2019 | Yes | No |
Additional files
- ISRCTN71157571_PIS_18Oct2018_FRENCH.pdf
- Uploaded 02/04/2019
- ISRCTN71157571_PIS_18Oct2018_SPANISH.pdf
- Uploaded 02/04/2019
Editorial Notes
09/09/2020: Internal review.
07/09/2020: The following changes have been made:
1. The overall trial end date has been changed from 30/06/2020 to 31/12/2021 and the plain English summary updated accordingly.
2. The intervention has been changed.
3. The primary outcome measures have been changed.
4. The condition has been changed from "Acute and chronic malnutrition in children" to "Acute and chronic malnutrition in children aged 6-59 months".
5. The participant inclusion criteria have been changed.
6. The target number of participants has been changed from "Minimum 450. 2 countries (150 Spain, 300 Senegal), nutritional status (normalweight: 150 Spanish, 150 Senegalese; SAM: 150 Senegalese), 2 sexes (50% each group), 2 age groups (50% each group)" to "Phase I: n= 450. 2 countries (150 Spain, 300 Senegal), nutritional status (normalweight: 150 Spanish, 150 Senegalese; SAM: 150 Senegalese). Phase II: n=1800 (1200 Senegal, 600 Guatemala), Senegal= 4 nutritional conditions decribed by WHO (25% each), Guatemala= wasting (presence/absence) + stunting (presence/absence) equally distributed. Phases I & II: 2 sexes (50% each group), 2 age groups (50% each group)".
7. The participant exclusion criteria have been changed.
8. The recruitment end date has been changed from 20/05/2017 to 31/12/2020.
9. The total target enrolment has been changed from 450 to 2250.
10. The total final enrolment number of 258 has been deleted, as this onlyreferred to phase I of the study.
11. Guatemala has been added to the countries of recruitment.
12. Grand Challenges Canada and Agencia Española de Cooperación (AECID) have been added as funders and the plain English summary updated accordingly.
13. The publication and dissemination plan has been updated.
14. The intention to publish date has been changed from 01/01/2019 to 31/12/2020.
09/07/2019: The following changes were made to the trial record:
1. Publication reference added.
2. The total final enrolment was added.
02/04/2019: The participant information sheet has been uploaded.