Related Experiment Video
Updated: Jul 19, 2026
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
Prediction of malnutrition in kids by integrating ResNet-50-based deep learning technique using facial images
S Aanjankumar1, Malathy Sathyamoorthy2, Rajesh Kumar Dhanaraj3
1School of Computing Science and Engineering, VIT Bhopal University, Bhopal-Indore Highway, Kothrikalan, 466114, Sehore, Madhya Pradesh, India.
Insights
This study introduces an AI image segmentation technique to predict severe acute malnutrition (SAM) in children, simplifying diagnosis and reducing the need for manual tests. The ResNet-50 model achieved 98.49% accuracy, outperforming other deep learning methods.
Area of Science:
- Artificial Intelligence
- Medical Imaging
- Pediatric Health
Background:
- Severe acute malnutrition (SAM) affects millions of children globally, with India facing significant challenges.
- Traditional malnutrition diagnosis relies on time-consuming methods like BMI monitoring and manual tests, often inaccessible in rural areas.
- Parental lack of awareness and missed medical appointments exacerbate malnutrition issues.
Purpose of the Study:
- To develop an artificial intelligence-based image segmentation technique for early malnutrition prediction in children.
- To simplify the diagnostic process by eliminating the need for multiple manual tests and expert consultations.
- To leverage deep learning for efficient and accurate malnutrition detection.
Main Methods:
- Utilized a deep learning model, specifically ResNet-50, incorporating shortcut connections to address the vanishing gradient problem.
- Employed image segmentation techniques for direct malnutrition prediction from visual data.
- Compared the proposed model's performance against other deep learning models like XG Boost, VGG 16, Xception, and MobileNet.
Main Results:
- The ResNet-50 model achieved a high accuracy of 98.49% in identifying malnourished children.
- The proposed AI system demonstrated superior performance compared to XG Boost (75.29%), VGG 16 (94%), Xception (95.41%), and MobileNet (92.42%).
- The AI-driven approach effectively detects malnutrition without requiring predictive analysis functions or medical expert advice.
Conclusions:
- The developed AI image segmentation technique offers an effective and efficient method for early malnutrition detection in children.
- This approach significantly simplifies diagnosis, particularly benefiting children in remote areas with limited access to healthcare.
- The ResNet-50 model shows promise in improving pediatric malnutrition screening and management.
Abstract:
In recent times, severe acute malnutrition (SAM) in India is considered a serious issue as per UNICEF 2022 records. In that record, 35.5% of children under age 5 are stunted, 19.3% are wasted, and 32% are underweight. Malnutrition, defined as these three conditions, affects 5.7 million children globally. This research utilizes an artificial intelligence-based image segmentation technique to predict malnutrition in children. The primary goal of this research is to use a deep learning model to eliminate the need for multiple manual diagnostic tests and simplify the prediction of malnutrition in kids. The traditional model uses text-based data and takes more time with continuous monitoring of kids by analysing body mass index (BMI) over different periods. Children in rural areas often miss medical expert appointments, and a lack of knowledge among parents can lead to severe malnutrition. The aim of the proposed system is to eliminate the need for manual blood tests and regular visits to medical experts. This study uses the ResNet-50 deep learning model's built-in shortcut connection to solve the image-based vanishing gradient problem. This makes training more efficient for image segmentation tasks in predicting malnutrition. The model is 98.49% accurate in predicting the kids who are malnourished among the kids who are healthy. It is evident from the results that the proposed system serves better than other deep learning models, such as XG Boost (75.29% accuracy), VGG 16 (94% accuracy), Xception (95.41% accuracy), and MobileNet (92.42% accuracy). Hence, the proposed technique is effective in detecting malnutrition and diagnose it earlier, without using predictive analysis function or advice from the medical experts.
