NiADA (Non-invasive Anemia Detection App), a Smartphone-Based Application With Artificial Intelligence to Measure Blood Hemoglobin in Real-Time: A Clinical Validation
Bibliographic record
Abstract
Background Anemia is a severe public health problem in India affecting more than half of the population. To reduce its burden on the population, the Government of India under the Anemia Mukt Bharat program has adopted a monitoring strategy for the diagnosis and treatment of anemia. Point-of-care testing (POCT) devices play a pivotal role in testing hemoglobin at a community level where sophisticated laboratory instruments are not available. The majority of the currently available POCT devices are invasive in nature which have their own limitations. A non-invasive method of hemoglobin estimation will address many of the limitations of an invasive POCT instrument, which will further improve people’s acceptability for hemoglobin testing. The Non-Invasive Anemia Detection App (NiADA) (Monere AI Private Limited, Kolkata, West Bengal, India), a non-invasive POCT application, uses artificial intelligence (AI) to predict the hemoglobin level from lower eyelid images. This real-time, point-of-care, low-cost solution uses a custom computer vision deep-learning algorithm to determine blood hemoglobin value. Method The study validates an AI-based smartphone application NiADA against laboratory hemoglobin estimation and a widely used point-of-care hemoglobin estimation instrument (HemoCue Hb 301; HemoCue AB, Angelholm, Skane County, Sweden). The study was conducted in a tertiary care hospital in Eastern India and recruited a total of 556 participants. These included 58 pediatric patients, 51 pregnant women, 214 adult females, and 224 adult males. Statistical analysis was performed using Python (Python Software Foundation, Wilmington, Delaware, United States). A p-value of < 0.05 was taken to be significant. Result The mean difference observed between NiADA and laboratory-estimated hemoglobin values came out to be -0.29 g/dL and -0.89 g/dL for adult females and males respectively, and 0.61 g/dL for pregnant women and -0.69g/dL for the pediatric population. The limits of agreement for NiADA were narrow at 2.77 to -2.18 g/dL for adult females, 3.76 to -1.96 g/dL for adult males, 1.89 to -3.29 g/dL for pregnant women, and 3.28 to -2.08 g/dL for the pediatric population. The sensitivity and specificity of the NiADA application against the laboratory estimation method were 75.8% and 53.8% for adult females, 70.0% and 48.3% for adult males, 23.8% and 90% for pregnant females, and 75% and 57% for the pediatric population. Conclusion As a non-invasive application, NiADA’s performance is satisfactory and comparable with minimally invasive tools like HemoCue Hb 301 and other POCT devices.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".