Kernel Flow and the need for neuroimaging


Kernel Flow

At PSYCH Symposium: London 2022, Kernels Chief Technology Officer, Ryan Field, delivered a presentation on the use of the company’s Kernel Flow device in a Cybin-sponsored feasibility study.

Kernel Flow is a headset that measures brain activity before, during and after the administration of psychedelic medicine, by recording localised changes in blood oxygenation.

Akin to a wearable fMRI scanner, Kernel Flow provides quantifiable neuroimaging data in real-time. To discuss the company’s innovation and how the scanner can be used to accelerate clinical trials, PSYCH sat down with Kernel’s Chief Technology Officer.

Kernel was started by Bryan Johnson, the company’s CEO, who recognised there was an unmet need to measure brain activity,’ explained Field. ‘How the brain and mind works is central to everything, yet we lack the tools to measure it in a meaningful way at scale.

‘With Kernel Flow we can now say with confidence that we have a technology that provides an accessible, scalable and affordable way to measure brain activity. The next step is identifying where Kernel Flow can be applied to address unmet needs, which is where we have these verticals forming and a big thrust initially towards psychedelic research and then endpoints.’ 

Kernel Flow

Kernel sits at the intersection of technology and psychedelic healthcare, and PSYCH was keen to learn about the importance of quantifiable data in clinical trials.

‘Drug development, both in psychedelic and traditional pharmaceutical research, often relies on the collection of subjective data,’ said Field. 

‘A lot of clinical endpoints are driven by patient reported outcomes, such as how they feel and if there was a perceived benefit, which makes it difficult to quantify outcomes precisely over time. 

‘At PSYCH Symposium: London 2022, we demonstrated that Kernel Flow has the capability to measure an individual’s brain state before, during and after treatment with a psychedelic. 

‘This enables us to identify how a person’s state changes over time and could someday predict when another treatment is due and the appropriate personalised dosing protocol. Hopefully they are not returning to where they were before the treatment, but there may be some regression over time, and this could help companies identify and create the ideal personalised protocol for a particular ailment. 

‘Kernel Flow can be particularly beneficial in early-phase trials, in addition to eventually guiding clinicians on the best course of action – using data from the individual for a more personalised approach to medicine. 

‘The hypothesis is that we can accelerate drug trials by providing quantifiable data, as error bars should be lower with direct measurements as opposed to subjective patient reported outcomes. 

‘Currently, subjective measurements are mediated through a patient’s ability to sense what they are feeling and verbalise it, or through a survey where things tend to get lost in translation. With Kernel Flow’s quantifiable data, there is the ability to get faster results with greater precision in measuring brain states.’ 

Drug developers monitor a number of metrics to assess the level of psychedelic experience instigated by a particular medicine. PSYCH asked Field whether Kernel’s headsets were being used in conjunction with other monitoring technologies.

‘Kernel Flow has the ability to measure types of patterns, which would normally require individuals to be in an fMRI tube,’ commented Field. ‘In these large scanners, it can be difficult to add additional technologies that measure blood plasma levels or other metrics crucial to clinical trials. However, a system like Kernel Flow can be used anywhere, allowing clinicians greater access to patients for additional monitoring.’

Big data has revolutionised industries across the board, with significant inroads in healthcare. PSYCH was eager to learn more about the marriage of big data and neuroimaging.

‘I would argue that big data has not existed in brain imaging before, although efforts have been made – such as with the Human Connectome Project,’ declared Field. ‘These have leveraged MRI data from thousands of patients, but, as the data has been collected on different scanners, extensive work is needed to harmonise it.

‘With Kernel Flow all the devices work exactly the same way, so no calibration is needed. Without the need to translate different types of data, we can start to build massive neuroimaging data sets. From there you can start to pick out patterns and build machine-learning-based algorithms, which will be hugely beneficial to understanding brain activity.

‘We designed and started building the prototype systems last July, and will release the initial lot of 100 production devices this fall. These are the systems we have designed to be able to scale to large numbers as we establish markets with a need for functional neuroimaging data. There is no limit to how many we can produce. We just need to identify the partners who want to use neuroimaging as a core part of their product delivery or research and development.

‘Building a million units is a feasible proposition, which will reduce the cost of each system. One of our stated objectives is that by 2033 we envision a world with a Kernel Flow device in every home – for personal use. In the short term, we will continue to work with partners to develop better treatments or tools for clinicians.’

Having conducted a Cybin-sponsored study with ketamine, PSYCH was interested to learn whether the technology had been used with other psychedelic medicines.

‘Not yet, but we would love to,’ revealed Kernel’s CTO. ‘We have been in talks with a lot of companies we met at PSYCH Symposium and once we have finished the ketamine study with Cybin, which demonstrates our proof of concept, we would like to follow that with studies using other psychedelic molecules. 

‘We expect to finish data collection from Cybin’s ketamine study in the next couple of weeks, with the results published late August early September.’

Understanding if drugs work is important, but often more beneficial is understanding how they work. PSYCH asked Field about the capabilities of Kernel Flow in regards to identifying mechanisms of action.

‘That is a good question. With ketamine we saw changes in specific functional brain networks, so while we don’t know the underlying mechanisms of action, we can see a functional response to the drug,’ he explained. 

‘We know that some mental health issues are associated with connectivity in certain networks and that these networks can be modulated by different drugs, with large fluctuations in default mode network connectivity a signature observed with our  ketamine trials.’

In an increasingly digitised world, data can be of immense value – particularly when demonstrating efficacy in clinical trials. PSYCH asked about Kernel’s business model and the collection of data.

‘Typically, whoever we are partnering with would have access to the data as well as Kernel,’ clarified Field. ‘We try to set it up so that both sides can benefit from the data, with the stance that, ultimately, control of the data lies with the individual.

‘When a person signs up to use the Kernel Flow device, the data collected is associated with the individual and they can access it through the Kernel app. We believe that measuring brain activity is very personal and that people should know what their data is being used for. We have added a consent mechanism with regards to sharing data, so we can aggregate data from multiple trials while ensuring we work fairly with our partners. This ensures everyone benefits to move the field forward.

‘Our business model is centered around the sale of the Kernel Flow device and associated services, working with our partners to ensure they achieve their targeted endpoints. However, we do think that having a database of brain activity from healthy individuals, as well as from those with clinical indications, will be of immense value. From this data set, predictive models can be built to identify who may or may not respond to a particular treatment. 

‘The drive is finding the right drug for the right person, and we see a clear path to having a system in every doctor’s office around the world.’

The inaugural PSYCH Symposium was a sold out event, with decision makers converging to unlock the industry’s potential. To empower investor to make informed decisions, The Psychedelics as Medicine Report: Third Edition contains 200+ pages of industry data and investment insights. Download a complimentary copy:

Related posts