# Sensor Fusion System For Early And Accurate Fall Detection and Injury Protection

> **NIH NIH R43** · GRACEFALL, INC. · 2021 · $298,873

## Abstract

Project Summary/Abstract
Fall is the leading cause of injury among elderly. One in three adults over 65 falls each year. Of those who fall,
20% to 30% suffer moderate to severe injuries and increase their risk of early death. In 2015, the total medical
costs related to fall injuries for people 65 and older was over $50 billion. Currently, there are devices that use
motion sensors (accelerometers) to detect imminent falls and inflate micro-airbags located in garments worn by
the users to protect from injury. The literature has shown that wearable solutions based on motion sensing
have low detection accuracy and suffer from false-positive events (the airbag may erroneously deploy during
daily activities after interpreting abrupt movements as falls). GraceFall, Inc. (GFI) will develop a patent
protected fall detection device based on a sensor-fusion algorithm that combines brain (EEG) and body motion
signals to allow reliable fall prediction and injury protection. Our initial findings, along with supporting literature,
show that a reliable EEG signal preceding an unexpected loss of balance could be the key to developing a
complete, reliable, ergonomic solution for fall detection and injury prevention, and would have a major impact
on maintaining mobility and quality of life in our aging population. Accelerometers reflect body movement and it
is difficult to distinguish between loss of stability and other non-fall related activities. The key difference
between intended actions and unintended loss of balance is the appearance of a “startle” response that can be
captured on most EEG channels. Using EEG sensors will allow us to identify the difference between a fall and
other acceleration scenarios. Our goal is to create a device that primarily uses these reliable brain responses,
coupled with motion sensors, to accurately detect loss of balance and stability, thereby preventing injuries due
to falling. The goal of the proposed Phase I project is to provide a proof of concept for a future product. Using
existing fall protection products that rely on motion sensing to detect an imminent fall, we will identify scenarios
in which these products have either false-positive (the airbag erroneously deploying in daily activities after
interpreting an abrupt movement as a fall) or false-negative (the airbag not deploying in a real fall scenario)
events. We will simulate these same scenarios on human subjects (Aim 1) and we will characterize the
physiological parameters of the startle response in an elderly population (Aim 2) to refine the sensor fusion
algorithm. The purpose of this proposal is proof of concept that adding a sensor fusion algorithm that combines
the EEG information with the acceleration data, improves the performance and reliability of the protection
system.

## Key facts

- **NIH application ID:** 10256574
- **Project number:** 1R43AG067843-01A1
- **Recipient organization:** GRACEFALL, INC.
- **Principal Investigator:** Emily A Keshner
- **Activity code:** R43 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $298,873
- **Award type:** 1
- **Project period:** 2021-09-30 → 2023-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10256574

## Citation

> US National Institutes of Health, RePORTER application 10256574, Sensor Fusion System For Early And Accurate Fall Detection and Injury Protection (1R43AG067843-01A1). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10256574. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*