Skip to content

Hemoglobin Easy Measurement With Optical Artificial Intelligence

A Prospective Feasibility Study to Explore the Utility of Using a Smartphone Camera to Monitor Blood Hemoglobin Levels in Children and Adolescents

Status
Completed
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT04573244
Acronym
HEMO-AI
Enrollment
823
Registered
2020-10-05
Start date
2020-12-02
Completion date
2023-08-31
Last updated
2024-08-23

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Hemoglobin; Deficiency

Brief summary

Blood hemoglobin levels are an extremely important measure for a large swath of medical procedures as they reflect the oxygen-carrying capacity of human blood. The gold standard measure for blood hemoglobin levels involve a venous blood draw followed by a laboratory-based complete blood count (CBC), a process which is both painful and time consuming. To date, various methodologies have been tested to either expediate the process or provide a non-invasive alternative. There remains a need to provide a quick, pain-free/non-invasive and accurate modality to measure blood hemoglobin levels. The objective of this study is to determine whether computer vision technologies can be applied to fingernail images captured via a smartphone camera to quantify blood hemoglobin levels.

Interventions

None listed

Sponsors

MYOR Ltd.
Lead SponsorINDUSTRY

Study design

Observational model
CASE_ONLY
Time perspective
PROSPECTIVE

Eligibility

Sex/Gender
ALL
Age
6 Months to 18 Years
Healthy volunteers
Yes

Inclusion criteria

1. A patient aged 6 months to 18 years. 2. A patient who has undergone a venous blood draw for a CBC since being admitted to the PED no more than 6 hours prior to study enrollment. 3. Parents or legal guardian provide informed written consent.

Exclusion criteria

1. Patient has subungual hematoma, nail bed lacerations or avulsion injuries on both hands. 2. Patient has total leukonychia. 3. Patient has nail polish applied on fingernails. 4. Patient has nailbed darkening or discoloration due to medication. 5. Any other reason that, in the opinion of the investigator, prevents the subject from participating in the study or compromise the patient safety.

Design outcomes

Primary

MeasureTime frameDescription
To determine whether computational learning methods can be applied to fingernail images captured via a smartphone camera to quantify blood hemoglobin levels.6 monthsEvaluated using the hemoglobin portion of a conventional complete blood count (CBC)

Secondary

MeasureTime frameDescription
To determine whether computational learning methods can be applied to fingernail images captured via a smartphone camera to screen for anemia as defined by the WHO6 monthsEvaluated using the hemoglobin portion of a conventional complete blood count (CBC). Anemia cutoff for children aged 6 months to 6 years = 11 g/dL and for children aged 6-14 years = 12 g/dL.
To determine whether computational learning methods can be applied to fingernail images captured via a smartphone camera to quantify other elements of the CBC6 monthsEvaluated using the hemoglobin portion of a conventional complete blood count (CBC)

Countries

Israel

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 5, 2026