Future of robotics: Nanoflex gears up to launch robot for remote stroke intervention 

Credit: Shutterstock/Pop Tika

Nanoflex Robotics AG, a Swiss company based in Zurich, is gearing up this year to launch a robotic endoscopic navigation platform designed to perform mechanical thrombectomies on stroke patients, in what would be a world first. Matt Curran, co-founder and CEO of Nanoflex Robotics, confidently stated: ‘We are transforming mechanical thrombectomies into reality through remote robotic surgery, thereby reducing treatment time and preserving the quality of life for stroke patients’.  This significant advance promises to redefine standards in stroke treatment, bringing forth an innovative and technologically advanced approach for acute ischaemic stroke patients, for whom timely intervention is critical.

According to the World Health Organization (WHO), 15 million people have a stroke every year, of whom 6.5 million will die, and 5 million will be left permanently disabled. In total, one in four adults over the age of 25 will have a stroke during their lifetime. This makes stroke the second most common cause of death and the primary cause of long-term disability worldwide. The primary techniques for mechanical thrombectomy in ischaemic stroke, namely use of a stent retriever or aspiration, show recanalisation rates of up to 88%. But there remains a mortality rate of approximately 15% to 20% due to the risks of complications, vascular lesions and the complexity of the thrombectomy procedure. As Bradley Nelson, professor of robotics and intelligent systems at the Swiss Federal Institute of Technology (ETH) Zurich, and a co-founder of and mentor to Nanoflex Robotics, emphasises, ‘physically removing the blood clot from the patient’s brain involves highly specialised, challenging, and delicate procedures.’

The three co-founders of Nanoflex – Bradley Nelson, Christophe Chautems and Matt Curran – came up with the idea of developing robotic-assisted neurosurgery based on a soft robotic system capable of precisely inserting specially designed catheters deep into the brain. These are intended to harness the sub-millimetre precision of robotic systems to accurately target the thrombus. 

Established in November 2021 as a spin-off from the ETH Multi-Scale Robotics Lab, Nanoflex Robotics is focused on developing a multi-patented endoluminal soft robotic system, which represents the culmination of over 20 years of collaborative studies by robotics engineers, clinical specialists and senior scientists at ETH. The start-up has pioneered magnetic navigation that facilitates the precise insertion of specially made guidewires and catheters deep into the brain.

With a team of 15 robotics engineers, Nanoflex Robotics has successfully engineered a compact and portable generator of precise electromagnetic fields. This generator, coupled with advanced algorithms and mathematical modelling, makes it possible to move medical devices containing small magnets with remarkable precision within a three-dimensional space. Surgeons can then exert precise control over their position using intuitive user interfaces.

‘We not only offer our patented magnetic navigation system but also provide the essential surgical devices’, said Matt Curran. The company has developed innovative soft, ultra-flexible and steerable catheters and guidewires tailored to reach the distal vasculature within the brain.  

In January 2024, Nanoflex Robotics received ISO 13485 certification. The acquisition of this global certification for quality management systems in designing and manufacturing remote robotic devices for endovascular interventions is ‘a step forward in bringing to market innovative remote robotic solutions that are designed to increase access to critical care procedures, such as the treatment of acute ischaemic stroke,’ said Christophe Chautems, CTO of Nanoflex Robotics. 

The Swiss start-up, supported by the Swiss Innovation Agency, has secured €18.5 million since its inception. The Nanoflex Robotics team is focusing on developing entirely remote mechanical thrombectomies for acute ischaemic stroke patients. ‘Physicians operating this soft robotic system are not required to be present with the patient during the procedure, significantly reducing their X-ray exposure,’ stated Curran. This leap forward will enable physicians to treat patients with remote control of the robot. ‘As a result, patients can receive treatment in local hospitals rather than being transferred to larger, specialized centers. This is particularly crucial for ischemic stroke treatment, as the longer it takes to remove the clot after formation, the more damage may be caused to brain tissue,’ said Nelson.

In January, Nanoflex joined forces with Brainomix Ltd., based in Oxford, UK, as part of a UK-Switzerland bilateral collaborative R&D program that has secured €892,618.25 in funding. Brainomix specialises in designing AI-based medical imaging software solutions to enhance the diagnosis and treatment of acute stroke. Having obtained CE certification in 2018 for its e-CTA software dedicated to stroke diagnosis and treatment, Brainomix achieved a significant milestone in March 2023 by securing US Food and Drug Administration (FDA) clearance for its flagship stroke AI imaging software, Brainomix 360 e-ASPECTS (Automated Scoring of Early CT Scans for Stroke). This is an advanced, fully automated AI-based imaging tool for comprehensive stroke assessment. Brainomix’s AI stroke software is currently in use in over 30 countries.

Through this 2024 collaborative research partnership, Nanoflex Robotics and Brainomix are currently working on integrating their technologies to create an assisted navigation tool for devices such as catheters as they move through the vascular system. ‘The project aims to leverage the power of artificial intelligence to enhance Nanoflex Robotics’ cutting-edge magnetic robotic navigation system to be the first system worldwide to enable remote thrombectomy’, stated George Harston, chief medical and innovation officer at Brainomix.

Nanoflex Robotics’ technology, initially developed for minimally invasive surgery, opens up new possibilities for remote telesurgery in a variety of surgical indications. In addition to stroke, this technology could potentially be applied in areas such as abdominal surgery, cardiac surgery and thoracic surgery. Currently, Nanoflex is targeting the vast market of neurovascular devices and interventional neurology.

Nanoflex is up against ten or so major competitors, including the giants and industry leaders Medtronic plc, Johnson & Johnson Services Inc., Stryker Corp., Penumbra Inc., Microport Scientific Corp., Terumo Corp., Merit Medical Systems, Inc. and Asahi Intecc Co., Ltd. 

The paper showcased attempts to make GPT-4 produce data that supported an unscientific conclusion – in this case, that penetrating keratoplasty had worse patient outcomes than deep anterior lamellar keratoplasty for sufferers of keratoconus, a condition that causes the cornea to thin which can impair vision. Once the desired values were given, the LLM dutifully compiled a database that to an untrained eye would appear perfectly plausible.

Taloni explained that, while the data would fall apart under statistical scrutiny, it didn’t even push the limits of what Chat-GPT can do. “We made a simple prompt […] The reality is that if someone was to create a fake data set, it is unlikely that they would use just one prompt. [If] they find an issue with the data set, they could fix it with consecutive prompts and that is a real problem. 

“There is this sort of tug of war between those who will inevitably try to generate fake data and all of our defensive mechanisms, including statistical tests and possibly software trained by AI.”

The issue will only worsen as the technology becomes more widely adopted too. Indeed, a recent GlobalData survey found that while only 16.1% of respondents from its Hospital Management industry website reported that they were actively using the technology, a further 26.8% said either that they had plans to use it or were exploring its potential use.

Nature worked with two researchers, Jack Wilkinson and Zewen Lu, to examine the dataset using techniques that would commonly be used to screen for authenticity. They found a number of errors including a mismatch of names and sexes of ‘patients’ and lack of a link between pre- and post-operative vision capacity. 

In light of this, Wilkinson, senior lecturer in Biostatistics at the University of Manchester, explained in an interview with Medical Device Network that he was less concerned by AI’s potential to increase fraud.

“I started asking people to generate datasets using GPT and having a look at them to see if they could pass my checks,” he said. “So far, every one I’ve looked at has been pretty poor. To be honest [they] would fall down under even modest scrutiny.” 

He acknowledged fears like those raised by Dr. Taloni about future improvements in AI-generated datasets but ultimately noted that most data fraud is currently done by “low-skill fabricators,” and that “if those people don’t have that knowledge, they don’t know how to prompt Chat-GPT to have it either.”

The problem for Wilkinson is how widespread falsification already is, even without generative AI.