Tag Archives: machine learning

Student develops wearable technology aimed to prevent sexual assault

bracelet vs conceal carry

From Daily Mail: A University of Alabama master’s student has developed an unusual tool to combat sexual assault: a bracelet.

The device monitors wearers’ vital signs and can pick up on changes that would signal being under attack. The smart bracelet, still in the prototype stage, alerts a predetermined list of contacts and the police if it senses that the wearer is being harmed.

Additionally, it emits a loud alarm and is equipped with red strobe lights meant to scare away the attacker or get the attention of others in the vicinity.

The team behind the bracelet, Associate Professor Dr. Ragib Hasan and master’s student Jayun Patel, are hopeful that the device can prevent sexual assaults, which are extremely common on college campuses such as theirs.

Dr. Hasan explained in a statement that the ability to call the authorities is usually unavailable to assault victims under attack.

He said: ‘A major challenge to assault prevention is that, during an assault, victims often to not have an easily accessible way to call for help. Whether calling 911 or using an emergency alert app or device, each of these tools requires users to press a button in order to call for help. That is often not possible while a violent act is taking place or if a person is unconscious as a result of the assault.

Dr. Hasan directed Patel while she designed the new product, which relies on machine learning and sensors to detect signs of assault.

The bracelet houses an Adafruit Circuit Playground, pressure sensors, GPS capabilities and microphones, among other tools. It can also determine whether a wearer is standing or lying down.

If the device recognizes that a wearer is in danger it connects to their smartphone via Bluetooth and immediately messages emergency personnel and sends them the wearer’s location. Additionally, it alerts friends the wearer has designated by way of an app.

Patel said: ‘The sensors allow the bracelet to collect user activity and vital signs continuously. A machine learning algorithm detects and differentiates the user’s regular movement and unexpected and sudden movements that can be indicative of an assault.’

The bracelet was designed with the intention of creating a device that can detect assault, but it can be applied to other emergency situations.

For example, the engineers said it could potentially be used to alert the authorities if an elderly person falls or to provide a warning if someone with a disability performs a risky movement.

Patel said she wants to expand the technology and create other ‘smart’ items, such as shoes and earrings, that work to keep people safe.

The wearable devices designed in Dr. Hasan’s lab are low-cost; the prototype for the bracelet costs less than $40.00. The researchers are hopeful the price will drop even more if the product is eventually mass produced.

The technology could benefit a large portion of the population: one-third of women and one-sixth of men in the US alone experience sexual violence.

DCG