Using Automated Infant Posture Recognition to Reduce SIDS Risk

Isha Narang

Overview

According to the CDC, approximately 3500 infants die annually in the United States from sleep-related infant deaths, including Sudden Infant Death Syndrome (SIDS). My project aims to reduce the extrinsic risk factors of SIDS by initiating an automatic alert when an infant’s posture is high-risk (as considered by the American Academy of Pediatrics). I downloaded pictures of infants in various lying positions and produced about 50,000 video frames. On each frame, I ran PoseNet: a model that generates coordinates and a certainty score for different body parts. I used it to recognize each infant’s nose, eyes, ears, shoulders, elbows, and wrists. I progressively added more features, such as angles and distances between certain body parts. Using this dataset, I built a geometric algorithm and a machine learning (ML) model. I changed the geometric algorithm to include threshold values for all measurements. Then, I imported the CSV data file into WEKA (an ML software) and experimented with different algorithms. The ML models with the lowest and highest accuracy were generated by Decision Tables (64.78%) and Random Forests (91.07%) respectively. After refining the Random Forests (RF) model by optimizing hyperparameters (objective functions and number of trees), the accuracy improved to 96.67% with a root mean square error of 0.072 when using 10-fold cross-validation. I incorporated this RF model into my “SIDS Pose Recognition” application. With my user-friendly app, caregivers would immediately receive an alert when an infant’s position is unsafe or high-risk for SIDS.

My Story

While my research ended with 'SIDS Pose Recognition' -- a user-friendly webcam app which initiates audio/visual alerts when an infant's lying posture is high-risk for SIDS -- it started with my simple desire to improve prevention of SIDS.

Personally, my interest to find a way to reduce risk of SIDS came from an experience that my aunt had before I was born. Her first child passed away from SIDS, and ever since, all of my relatives on my mom's side had been very paranoid about how they positioned their sleeping infants. If an infant was lying down, they felt the need to always be in the room and supervise. This problem affects many other people, as according to the CDC, approximately 3,500 infants die annually in the United States from sleep-related infant deaths, including Sudden Infant Death Syndrome (SIDS).

I worked on my research for about seven months. The hardest part was deriving my procedure, as I started from scratch and often had to backtrack when unexpected obstacles popped up. When I was finally done conducting my project, I synthesized all of my material into a research paper, virtual poster, powerpoint, etc.

After qualifying for and competing in the North Carolina Student Academy of Science's State competition, I placed 1st in the Computer Science category. Thus, I was inducted as an AJAS fellow to represent the state of North Carolina.