AI-StuntDetect: Development of Early Stunting Detection System for Young Children Based on CNN and MediaPipe Pose

Asmaul Husna  RS; Purnamawati Purnamawati; Hendra Jaya; Anas Arfandi

doi:10.35882/jeeemi.v8i4.2041

Asmaul Husna RS Doctoral Study Program in Engineering Vocational Education, Makassar State University, South Sulawesi, Indonesia https://orcid.org/0009-0002-0320-217X
Purnamawati Purnamawati Doctoral Study Program in Engineering Vocational Education, Makassar State University, South Sulawesi, Indonesia https://orcid.org/0000-0001-8516-4167
Hendra Jaya Doctoral Study Program in Engineering Vocational Education, Makassar State University, South Sulawesi, Indonesia https://orcid.org/0000-0003-3312-5603
Anas Arfandi Doctoral Study Program in Engineering Vocational Education, Makassar State University, South Sulawesi, Indonesia https://orcid.org/0000-0002-2513-6786

DOI: https://doi.org/10.35882/jeeemi.v8i4.2041

Keywords: Anthropometry, CNN, MediaPipe Pose, Posyandu, Stunting

Abstract

Stunting affects a substantial proportion of Indonesian children under five and is monitored through routine anthropometric measurement at posyandu, community health posts staffed largely by volunteer cadres. This measurement depends on calibrated instruments and consistent technique, which are not uniformly available across sites, so a low-cost screening aid could extend coverage where conventional practice is weakest. This study aims to develop AI-StuntDetect, a web-based system that estimates children's height and weight from a single mobile-phone photograph, classifies nutritional status, and evaluates whether its image-derived estimates add value beyond simple demographic predictors. A Convolutional Neural Network estimates weight directly from the image, while MediaPipe Pose landmarks are converted into body-proportion features and regressed onto height with a Random Forest. The models were developed on 117 children (1,593 images) using child-level partitioning and validated externally on 48 children from three independent posyandu. Performance was compared against a mean-value baseline and against age-only and age-plus-sex regressions fitted to the training data, and predicted heights were classified against the WHO and Kemenkes reference standards. Height estimation achieved an MAE of 4.092 cm and an R² of 0.863, reducing error by 31.9% relative to the strongest training-derived baseline. Weight estimation achieved an MAE of 1.771 kg and an R² of 0.590 but did not improve on the age-only regression (1.675 kg). Nutritional status was classified with 70.8% accuracy and a macro-averaged F1 of 0.546, with weak sensitivity for the intermediate stunted category (0.167). A single photograph therefore supports useful height estimation, whereas image-based weight estimation adds nothing beyond what age predicts. AI-StuntDetect is best characterized as an experimental screening prototype suited to triage under manual confirmation.

Downloads

Download data is not yet available.

References

S. Rahardiantoro et al., “Spatio-temporal modeling to identify factors associated with stunting in Indonesia using a modified generalized lasso,” Spat. Spatiotemporal Epidemiol., vol. 51, Art. no. 100694, 2024, doi: 10.1016/j.sste.2024.100694.

P. K. Arya, K. Sur, T. Kundu, S. Dhote, and S. K. Singh, “Unveiling predictive factors for household-level stunting in India: A machine learning approach using NFHS-5 and satellite-driven data,” Nutrition, vol. 132, Art. no. 112674, 2025, doi: 10.1016/j.nut.2024.112674.

A. Ndagijimana, G. Nduwayezu, T. Lind, and A. Mansourian, “Machine learning techniques to model child low height-for-age in the Northern Province of Rwanda: The role of climatological and environmental factors and their interactions,” Clin. Epidemiol. Glob. Health, vol. 37, Art. no. 102284, 2026, doi: 10.1016/j.cegh.2025.102284.

V. G. Biradar, H. H, P. K. Pareek, S. T, K. Aishwarya, and K. Aneeth, “Performance analysis of deep learning models for detection of malnutrition in children,” in Proc. 2024 Int. Conf. Data Sci. Netw. Secur. (ICDSNS), Jul. 2024, pp. 1–8, doi: 10.1109/ICDSNS62112.2024.10691135.

S. Jain, T. Khanam, A. J. Abedi, and A. A. Khan, “Efficient machine learning for malnutrition prediction among under-five children in India,” in Proc. 2022 IEEE Delhi Sect. Conf. (DELCON), 2022, pp. 1–10, doi: 10.1109/DELCON54057.2022.9753080.

N. Akseer et al., “Economic costs of childhood stunting to the private sector in low- and middle-income countries,” EClinicalMedicine, vol. 45, Art. no. 101320, 2022, doi: 10.1016/j.eclinm.2022.101320.

I. Mostafa et al., “A microbiota-directed complementary food intervention in 12–18-month-old Bangladeshi children improves linear growth,” eBioMedicine, vol. 104, Art. no. 105166, 2024, doi: 10.1016/j.ebiom.2024.105166.

Md. M. Islam et al., “Application of machine learning based algorithm for prediction of malnutrition among women in Bangladesh,” Int. J. Cogn. Comput. Eng., vol. 3, pp. 46–57, 2022, doi: 10.1016/j.ijcce.2022.02.002.

R. E. Kofinti, I. Koomson, J. A. Paintsil, and E. K. Ameyaw, “Reducing children’s malnutrition by increasing mothers’ health insurance coverage: A focus on stunting and underweight across 32 sub-Saharan African countries,” Econ. Model., vol. 117, Art. no. 106049, 2022, doi: 10.1016/j.econmod.2022.106049.

E. Gayawan and O. A. Egbon, “Spatio-temporal mapping of stunting and wasting in Nigerian children: A bivariate mixture modeling,” Spat. Stat., vol. 58, Art. no. 100785, 2023, doi: 10.1016/j.spasta.2023.100785.

S. V. Subramanian, O. Karlsson, and R. Kim, “Revisiting the stunting metric for monitoring and evaluating nutrition policies,” Lancet Glob. Health, vol. 10, no. 2, pp. e179–e180, 2022, doi: 10.1016/S2214-109X(21)00504-0.

L. Bevis, K. Kim, and D. Guerena, “Soil zinc deficiency and child stunting: Evidence from Nepal,” J. Health Econ., vol. 87, Art. no. 102691, 2023, doi: 10.1016/j.jhealeco.2022.102691.

B. Haile and D. Headey, “Growth in milk consumption and reductions in child stunting: Historical evidence from cross-country panel data,” Food Policy, vol. 118, Art. no. 102485, 2023, doi: 10.1016/j.foodpol.2023.102485.

UNICEF, WHO, and World Bank, “Levels and trends in child malnutrition: UNICEF/WHO/World Bank Group joint child malnutrition estimates: Key findings of the 2023 edition,” World Health Organization, Geneva, Switzerland, 2023. ISBN: 978-92-4-007379-1.

R. D. Logarajan, N. M. Nor, S. Ibrahim, and R. Said, “Social determinants of stunting in Malay children aged <5 years in Malaysia,” Nutrition, vol. 111, Art. no. 112030, 2023, doi: 10.1016/j.nut.2023.112030.

M. B. Arndt et al., “Global, regional, and national progress towards the 2030 global nutrition targets and forecasts to 2050: A systematic analysis for the Global Burden of Disease Study 2021,” Lancet, vol. 404, no. 10471, pp. 2543–2583, 2024, doi: 10.1016/S0140-6736(24)01821-X.

H. A. Santoso, N. S. Dewi, S.-C. Haw, A. D. Pambudi, and S. A. Wulandari, “Enhancing nutritional status prediction through attention-based deep learning and explainable AI,” Intell.-Based Med., vol. 11, Art. no. 100255, 2025, doi: 10.1016/j.ibmed.2025.100255.

N. Khalid, R. Nawaz, Z. Zhou, and S. Gong, “Child malnutrition in Pakistan: Pathways linking maternal intimate partner violence, gender inequality, healthcare utilization, and age at marriage,” Child. Youth Serv. Rev., vol. 186, Art. no. 109039, 2026, doi: 10.1016/j.childyouth.2026.109039.

K. Simamora, B. W. Otok, and S. Setiawan, “Multilevel confirmatory factor analysis for modelling the determinants of malnutrition among under-five children in Indonesia,” in Proc. 2025 Int. Conf. Artif. Intell., Blockchain, Cloud Comput., Data Anal. (ICoABCD), 2025, pp. 1–5, doi: 10.1109/ICoABCD67551.2025.11470742.

R. Mukherjee et al., “Accuracy of a smartphone-based 3D imaging tool for height measurement and stunting detection among children 24–59 months of age in Nepal: A validation study,” Lancet Reg. Health Southeast Asia, vol. 48, Art. no. 100774, 2026, doi: 10.1016/j.lansea.2026.100774.

H. Anastasia et al., “Determinants of stunting in children under five years old in South Sulawesi and West Sulawesi Province: 2013 and 2018 Indonesian Basic Health Survey,” PLoS One, vol. 18, no. 5, Art. no. e0281962, 2023, doi: 10.1371/journal.pone.0281962.

Health Development Policy Agency, Ministry of Health of the Republic of Indonesia, Indonesian Health Survey 2023 in Numbers. Jakarta, Indonesia: Ministry of Health of the Republic of Indonesia, 2024.

S. Ichikawa, M. Hamada, and H. Sugimori, “A deep-learning method using computed tomography scout images for estimating patient body weight,” Sci. Rep., vol. 11, no. 1, Art. no. 15627, 2021, doi: 10.1038/s41598-021-95170-9.

A. Demircioğlu, A. S. Quinsten, L. Umutlu, M. Forsting, K. Nassenstein, and D. Bos, “Determining body height and weight from thoracic and abdominal CT localizers in pediatric and young adult patients using deep learning,” Sci. Rep., vol. 13, no. 1, Art. no. 19010, 2023, doi: 10.1038/s41598-023-46080-5.

L. D. B. Lima et al., “A scale-equivariant CNN-based method for estimating human weight and height from multi-view clinic silhouette images,” Expert Syst. Appl., vol. 256, Art. no. 124879, 2024, doi: 10.1016/j.eswa.2024.124879.

V. G. Sonar et al., “3D camera-based body weight estimation using artificial intelligence in emergency care settings,” Emerg. Care Med., vol. 2, no. 4, Art. no. 55, 2025, doi: 10.3390/ecm2040055.

H. Shu, L. Ren, L. Pan, D. Huang, H. Lu, and W. Wang, “Single image based infant body height and weight estimation,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Workshops (CVPRW), 2023, pp. 6052–6059, doi: 10.1109/CVPRW59228.2023.00644.

M. Sakina, I. Muhammad, and S. S. Abdullahi, “A multi-factor approach for height estimation of an individual using 2D image,” Procedia Comput. Sci., vol. 231, pp. 765–770, 2024, doi: 10.1016/j.procs.2023.12.140.

Risfendra, Aripriharta, Suherman, G. F. Ananda, D. S. Putra, and Fahmi, “Contactless infant height measurement for enhanced early detection of stunting using computer vision techniques,” IEEE Access, vol. 13, pp. 42364–42376, 2025, doi: 10.1109/ACCESS.2025.3548159.

L. Qiao et al., “Contactless height measurement for infants and toddlers based on computer vision,” in Proc. 2024 4th Int. Conf. Consum. Electron. Comput. Eng. (ICCECE), 2024, pp. 306–310, doi: 10.1109/ICCECE61317.2024.10504244.

P. Nguyen Trung et al., “A non-contacted height measurement method in two-dimensional space,” Sensors, vol. 24, no. 21, Art. no. 6796, 2024, doi: 10.3390/s24216796.

R. E. Caraka et al., “Bayesian network analysis of spatial disparities in child nutrition status and health determinants across Indonesia,” Int. J. Appl. Earth Obs. Geoinf., vol. 144, Art. no. 104936, 2025, doi: 10.1016/j.jag.2025.104936.

S. Bakri, D. A. Puspaningrum, K. N. Berawi, and R. Zuraida, “Determinants of stunting severity among under-fives: Comparison of agricultural and nonagricultural households in South Lampung Regency, Indonesia,” Adv. Public Health, vol. 2025, no. 1, Art. no. 7443296, 2025, doi: 10.1155/adph/7443296.

W. R. Mgomezulu, P. Thangata, B. Mkandawire, and N. Amoah, “Advancing predictive analytics in child malnutrition: Machine, ensemble and deep learning models with balanced class distribution for early detection of stunting and wasting,” Hum. Nutr. Metab., vol. 42, Art. no. 200340, 2025, doi: 10.1016/j.hnm.2025.200340.

A. P. Joshi and B. V. Patel, “Data preprocessing: The techniques for preparing clean and quality data for data analytics process,” Orient. J. Comput. Sci. Technol., vol. 13, no. 2–3, pp. 78–81, 2021, doi: 10.13005/ojcst13.0203.03.

C. Fan, M. Chen, X. Wang, J. Wang, and B. Huang, “A review on data preprocessing techniques toward efficient and reliable knowledge discovery from building operational data,” Front. Energy Res., vol. 9, Art. no. 652801, 2021, doi: 10.3389/fenrg.2021.652801.

L. Wang and L. Liu, “Athlete body type recognition based on spectral co-clustering and CNN,” Int. J. Cloth. Sci. Technol., vol. 38, no. 3, pp. 389–414, 2026, doi: 10.1108/IJCST-04-2023-0055.

Z. Li, F. Liu, W. Yang, S. Peng, and J. Zhou, “A survey of convolutional neural networks: Analysis, applications, and prospects,” IEEE Trans. Neural Netw. Learn. Syst., vol. 33, no. 12, pp. 6999–7019, 2022, doi: 10.1109/TNNLS.2021.3084827.

C. Chen, N. A. Mat Isa, and X. Liu, “A review of convolutional neural network based methods for medical image classification,” Comput. Biol. Med., vol. 185, Art. no. 109507, 2025, doi: 10.1016/j.compbiomed.2024.109507.

S. R. Dubey, S. K. Singh, and B. B. Chaudhuri, “Activation functions in deep learning: A comprehensive survey and benchmark,” Neurocomputing, vol. 503, pp. 92–108, 2022, doi: 10.1016/j.neucom.2022.06.111.

Y. Li et al., “A survey on dropout methods and experimental verification in recommendation,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 7, pp. 6595–6615, 2023, doi: 10.1109/TKDE.2022.3187013.

H. Khoshvaght, R. R. Permala, A. Razmjou, and M. Khiadani, “A critical review on selecting performance evaluation metrics for supervised machine learning models in wastewater quality prediction,” J. Environ. Chem. Eng., vol. 13, no. 6, Art. no. 119675, 2025, doi: 10.1016/j.jece.2025.119675.

F. Roggio, B. Trovato, M. Sortino, and G. Musumeci, “A comprehensive analysis of the machine learning pose estimation models used in human movement and posture analyses: A narrative review,” Heliyon, vol. 10, no. 21, Art. no. e39977, 2024, doi: 10.1016/j.heliyon.2024.e39977.

A. Latreche, R. Kelaiaia, A. Chemori, and A. Kerboua, “Reliability and validity analysis of MediaPipe-based measurement system for some human rehabilitation motions,” Measurement, vol. 214, Art. no. 112826, 2023, doi: 10.1016/j.measurement.2023.112826.

C. I. Allen Akselrud, “Random forest regression models in ecology: Accounting for messy biological data and producing predictions with uncertainty,” Fish. Res., vol. 280, Art. no. 107161, 2024, doi: 10.1016/j.fishres.2024.107161.

R. Dunne et al., “Thresholding Gini variable importance with a single-trained random forest: An empirical Bayes approach,” Comput. Struct. Biotechnol. J., vol. 21, pp. 4354–4360, 2023, doi: 10.1016/j.csbj.2023.08.033.

T. Thaventhiran et al., “A digital health solution for child growth monitoring at home: Testing the accuracy of a novel ‘GrowthMonitor’ smartphone application to detect abnormal height and body mass indices,” Mayo Clin. Proc. Digit. Health, vol. 1, no. 4, pp. 498–509, 2023, doi: 10.1016/j.mcpdig.2023.08.001.

D. A. Rusakov, “A misadventure of the correlation coefficient,” Trends Neurosci., vol. 46, no. 2, pp. 94–96, 2023, doi: 10.1016/j.tins.2022.09.009.