Related Experiment Video
Updated: Jul 15, 2025
Design and Analysis for Fall Detection System Simplification
Published on: April 6, 2020
Assessment System for Child Head Injury from Falls Based on Neural Network Learning
Ziqian Yang1,2, Baiyu Tsui1,2, Zhihui Wu1,2
1College of Furnishings and Industrial Design, Nanjing Forestry University, Nanjing 210037, China.
Insights
This study developed a real-time AI system to assess toddler head injuries from falls using video analysis. The system achieved 96.67% accuracy, proving the feasibility of AI for monitoring fall-related head trauma.
Area of Science:
- Pediatric Health
- Artificial Intelligence
- Biomechanical Engineering
Background:
- Toddlers are susceptible to severe head injuries from falls at home.
- Existing solutions for fall assessment are limited in effectiveness and precision.
Purpose of the Study:
- To develop a real-time assessment system for head injury in toddlers following falls.
- To utilize AI and video analysis for precise injury evaluation.
Main Methods:
- Phase I: Joint data extraction using Open Pose from surveillance video, followed by Long Short-Term Memory (LSTM) network and 3D transform model integration.
- Phase II: Derivation of head acceleration, calculation of Head Injury Criterion (HIC) values, and development of an injury classification model.
- Data Collection: 200 RGB videos of toddlers (13-30 months) falling, with 500 clips for training/validation and 300 for testing.
Main Results:
- The AI system demonstrated a high classification accuracy of 96.67% in assessing head injuries from falls.
- The framework successfully integrated spatial and temporal information from key body joints.
Conclusions:
- The study validates the feasibility of a real-time AI technique for monitoring and assessing head injuries in toddlers resulting from falls.
- This technology offers a potential solution for early detection and intervention of fall-related head trauma in young children.
Abstract:
Toddlers face serious health hazards if they fall from relatively high places at home during everyday activities and are not swiftly rescued. Still, few effective, precise, and exhaustive solutions exist for such a task. This research aims to create a real-time assessment system for head injury from falls. Two phases are involved in processing the framework: In phase I, the data of joints is obtained by processing surveillance video with Open Pose. The long short-term memory (LSTM) network and 3D transform model are then used to integrate key spots' frame space and time information. In phase II, the head acceleration is derived and inserted into the HIC value calculation, and a classification model is developed to assess the injury. We collected 200 RGB-captured daily films of 13- to 30-month-old toddlers playing near furniture edges, guardrails, and upside-down falls. Five hundred video clips extracted from these are divided in an 8:2 ratio into a training and validation set. We prepared an additional collection of 300 video clips (test set) of toddlers' daily falling at home from their parents to evaluate the framework's performance. The experimental findings revealed a classification accuracy of 96.67%. The feasibility of a real-time AI technique for assessing head injuries in falls through monitoring was proven.
More Related Videos
07:30A Test Bed to Examine Helmet Fit and Retention and Biomechanical Measures of Head and Neck Injury in Simulated Impact
Published on: September 21, 2017
07:02An Investigation of the Effects of Sports-related Concussion in Youth Using Functional Magnetic Resonance Imaging and the Head Impact Telemetry System
Published on: January 12, 2011