Photo showing man wearing robotic soft suit aided by two people

Harvard’s robotic exoskeleton can improve walking, decrease falls in people with Parkinson’s

Photo showing man wearing robotic soft suit aided by two people

Image Credits: Harvard Biodesign Lab

If you follow the world of robotic exoskeletons with any frequency, you’re no doubt aware of the two primary categories. The first is aimed firmly at workers with strenuous or repetitive jobs such as lifting and transporting packages. The second is assistive, largely focused at individuals with mobility impairments.

The latter category is often the domain of soft robotic exoskeletons — those with fabric parts designed to be more of a day-to-day assistive accessory. These systems are also increasingly focused on specific issues that impact people’s ability to get around. After all, not all mobility impairments are created equal.

In the case of people with Parkinson’s disease, “freezing” is a frequent issue that impacts the ability to walk, while increasing the likelihood of falls. The symptom causes people to lose their footing while walking, shortening steps before stopping altogether.

There have been a number of attempts to tackle the issue, from medication to physical therapy to surgery, but existing solutions are limited in their efficacy.

New research from a joint team from Harvard and Boston University, published in Nature Medicine, demonstrates how soft robotic exoskeletons can address the issue.

The wearable technology features sensors that detect movement and utilize algorithms to estimate the walker’s gait. Cable-driven actuators kick in, assisting walking midstride.

“We found that just a small amount of mechanical assistance from our soft robotic apparel delivered instantaneous effects and consistently improved walking across a range of conditions for the individual in our study,” says Harvard professor Conor Walsh.

Researchers worked with a 73-year-old man with Parkinson’s who suffers from freezing more than 10 times a day. In their research, which was carried out over the course of six months, Walsh’s team was able to eliminate freezing in patients while walking indoors. In turn, they were also able to walk farther and faster.

Harvard notes, “The effect was instantaneous. Without any special training, the patient was able to walk without any freezing indoors and with only occasional episodes outdoors. He was also able to walk and talk without freezing, a rarity without the device.”

The Harvard Biodesign Lab is the same group that designed an exosuit for stroke patients that was subsequently licensed by the startup ReWalk Robotics. If the promising early results are any indication, this new technology could someday follow in its footsteps to commercialization.

Photo showing man wearing robotic soft suit aided by two people

Harvard’s robotic exoskeleton can improve walking, decrease falls in people with Parkinson’s

Photo showing man wearing robotic soft suit aided by two people

Image Credits: Harvard Biodesign Lab

If you follow the world of robotic exoskeletons with any frequency, you’re no doubt aware of the two primary categories. The first is aimed firmly at workers with strenuous or repetitive jobs such as lifting and transporting packages. The second is assistive, largely focused at individuals with mobility impairments.

The latter category is often the domain of soft robotic exoskeletons — those with fabric parts designed to be more of a day-to-day assistive accessory. These systems are also increasingly focused on specific issues that impact people’s ability to get around. After all, not all mobility impairments are created equal.

In the case of people with Parkinson’s disease, “freezing” is a frequent issue that impacts the ability to walk, while increasing the likelihood of falls. The symptom causes people to lose their footing while walking, shortening steps before stopping altogether.

There have been a number of attempts to tackle the issue, from medication to physical therapy to surgery, but existing solutions are limited in their efficacy.

New research from a joint team from Harvard and Boston University, published in Nature Medicine, demonstrates how soft robotic exoskeletons can address the issue.

The wearable technology features sensors that detect movement and utilize algorithms to estimate the walker’s gait. Cable-driven actuators kick in, assisting walking midstride.

“We found that just a small amount of mechanical assistance from our soft robotic apparel delivered instantaneous effects and consistently improved walking across a range of conditions for the individual in our study,” says Harvard professor Conor Walsh.

Researchers worked with a 73-year-old man with Parkinson’s who suffers from freezing more than 10 times a day. In their research, which was carried out over the course of six months, Walsh’s team was able to eliminate freezing in patients while walking indoors. In turn, they were also able to walk farther and faster.

Harvard notes, “The effect was instantaneous. Without any special training, the patient was able to walk without any freezing indoors and with only occasional episodes outdoors. He was also able to walk and talk without freezing, a rarity without the device.”

The Harvard Biodesign Lab is the same group that designed an exosuit for stroke patients that was subsequently licensed by the startup ReWalk Robotics. If the promising early results are any indication, this new technology could someday follow in its footsteps to commercialization.

An image of the word Reputation with Karma3 Labs' logo

Karma3 Labs raises $4.5M to improve trust in web3 with ratings and recommendations

An image of the word Reputation with Karma3 Labs' logo

Image Credits: Karma3 Labs (opens in a new window)

It’s becoming harder to discern which platforms and people have good intentions online, especially as automated content and anonymous profiles are on the rise. Some Web 2.0 companies like Uber, Amazon and Airbnb offer rating systems for businesses and individuals, but those types of systems are few and far between in the web3 world.

But Karma3 Labs is hoping to change that with $4.5 million in fresh capital backing its decentralized reputation protocol OpenRank. This is the protocol’s first capital raise, Sahil Dewan, founder and CEO of Karma3 Labs, told TechCrunch exclusively.

The round was led by Galaxy and IDEO CoLab Ventures. Additional investors include Spartan, SevenX, HashKey, Flybridge, Delta Fund, Draper Dragon and Compa Capital. The capital will be used toward growing OpenRank’s adoption and helping launch the protocol’s initial version to developers.

“We are really obsessed with solving trust and safety issues for crypto,” Dewan said. “After the last [crypto] bull market, the DeFi and NFT mania happened, and a lot of people came into crypto, but a lot of people were getting scammed.”

There’s no reputation system in the decentralized world of web3, so it’s hard to figure out which entities and individuals to trust and depend on, Dewan said.

Since the internet’s inception, there have been peer-to-peer environments that allow businesses and individuals to publish and buy things. But these businesses have an advantage: “They can take the value from users, define the rules of what’s right or not, and sit on the data,” Dewan said. “It’s not a public good, but transactional relationships between centralized parties and users.”

Decentralization of ratings and reputation systems is important because it prevents a single entity from owning the reputation scores and being able to manipulate or alter them, Dewan said. OpenRank aims to help developers and web3 protocols launch consumer apps, communities and marketplaces with open rankings and recommendations, without the need for a centralized entity running it. “We wanted to create a protocol and generalized system, not as a source of trust, but for anyone to come and build reputation systems,” Dewan said.

This could create a foundation for peer-to-peer interactions and community ownership of ratings online.

The OpenRank protocol allows any developer to use its “Reputation Graphs” for ratings, ranking or recommendations for applications or communities. This means developers, consumer applications and marketplaces can integrate specific rankings and recommendations, while also leveraging rankings and reputations from other ecosystems and communities to build a foundation for their own.

To start, OpenRank is working with Metamask Snaps; providing ranking and recommendation APIs for Lens and Farcaster; and helping with on-chain discovery feeds for consumer apps, crypto wallets and reputation-based voting and governance, Dewan said.

“They can post it internally or use it behind the scenes to power search and recommendation, it’s up to developers,” Dewan said. “We aren’t going to tell them what number to attach. We want to create a ranking system that can be used for whatever utility they want to give to end users.”

The protocol also plans to put “resistant mechanisms” in place to prevent bad actors or scammers who try to cheat the system by wash trading or sharing malicious links.

Ratings also help reduce the cost of searching and discovering that’s on-chain or in the crypto ecosystem, Dewan noted. “If you don’t have anything rated, you won’t know what to buy or trust. User engagement won’t happen the same way we see in web2 if there isn’t a ranking.”

These rankings can be relative and specific to different people. What may pop up on one person’s recommendations might not reach others, based on their past interests and interactions. “Today you can’t challenge what Google or Amazon shows you,” Dewan said. “But third-party developers have a market to create new ranking systems, and that’s a guiding force to help us choose and show the most value for users.”

In the near term, the startup plans to continue working with its launch partners and open up OpenRank to help people find, buy and vote for what they trust on-chain. Its next goal is to open the protocol to any third-party developer who wants to implement a ranking and reputation system.

“Over time we want a self-serve model for OpenRank, so any developer can create their own rankings without permission and without having to do hard work on data and computing,” Dewan said.