Chetan Kaher, Chief Innovation Officer at Accelerator company Jiva.ai, shares the Jiva.ai story.
Jiva.ai: Multimodality in an easy-to-use AI platform
Back when we were students at Kings College London, Manish Patel (CEO of Jiva.ai) and myself realised we were lacking the ability to suitably integrate related datasets together (i.e. genetics, imaging and biomarkers) to give a better understanding of cancer progression and the therapies it would be amenable to. Indeed, Manish committed his PhD to this very subject: model integration in complex systems. After a pit stop in algorithmic trading for Manish and maxillofacial surgery and dentistry for me, we decided the time for multimodality in an easy-to-use AI platform had come. We were lucky enough to bring in our third co-founder, Sarah D’Souza, and Jiva.ai was no longer a dream!
Our first AI project
We joined the NHS Clinical Entrepreneurship program and were lucky enough to be given our first project in the AI-led diagnosis of prostate cancer from MRI scans, with the fantastic Professor Das Gupta. Back in 2019, NICE stated that the first line of diagnosis for prostate cancer should be an MRI scan. Unfortunately, there are not enough radiologists to fulfil demand, and the existing workforce we have is already stretched, with scans awaiting reading for up to 56 days.
A value proposition session run by the Knowledge Transfer Network highlighted an even bigger problem -the lack of good sensitivity and specificity in reading the scans. This means that there is a risk of false diagnosis and even worse, patients are needlessly sent for invasive biopsies, which have high complication rates (such as infection, erective dysfunction, sepsis). The avoidable cost to the NHS of this is well over £100 million and of course the health costs to the patients can be insurmountable.
A solution to the wider problem
Here at Jiva.ai, we have built an AI-led prostate cancer diagnostic tool (JivaRDX) with a sensitivity and specificity of over 90%. This is currently undergoing clinical trials, with a view to receive FDA and UKCA approval before year-end. A similar methodology was used to show how our underlying AI platform could save up to 80% on time and data science costs, by handing easy-to-use AI tools to the domain experts.
Using multiple datatypes
We have since been accepted onto the KQ labs AI Accelerator at the prestigious Francis Crick Institute and we have been very fortunate to have also won two InnovateUK grants in the earlier detection of Liver disease and Bone Fractures.
This varied work, using multiple datatypes has been fundamental in developing the underlying no code/low code multimodal AI platform, which allows users with no coding experience to build their own AI models with any type of data using a drag and drop interface.
As new models of data that also affect the system become available, these can be added to the platform iteratively and multimodal analysis can be performed.
The Welsh Development bank, Wealth Club and Consilience Ventures have also recently seeded us, and we have been able to grow a brilliant development and commercial team, led by the awesome Jarrod Germano.
Going beyond healthcare
As the platform begins commercialisation, we have gained multiple projects and not just in healthcare! We have now used the platform in debt management services and Marketing optimisation.
This year our objectives are to close a pre-series A round, gain regulatory approval for JivaRDX and add more AI techniques (such as natural language processing, reinforcement learning) into the platform as well as further commercialise the platform in other sectors, whilst developing a library of use cases.
If you are interested in working together with us, please do not hesitate to email us on: email@example.com.
DigitalHealth.London is delighted to publish blogs by the NHS staff and digital health companies we support through our programmes, as well as sector thought-leaders, experts and academics. Any opinions expressed within blogs published on our website are those of the author and not necessarily held by DigitalHealth.London. For more information, or if you would like to write a blog for our website, please email info@digitalHealth.london.
Jiva.ai is part of the sixth cohort of the DigitalHealth.London Accelerator programme.
The DigitalHealth.London Accelerator is a collaborative programme funded by two of London’s Academic Health Science Networks – UCL Partners and the Health Innovation Network, MedCity, CW+ and receives match funding from the European Regional Development Fund.