AI driven inhalation technique and adherence support device with data aggregation platform for remote monitoring of asthma

The global cost of asthma is high and rising with an expected 400M sufferers by 2025 and 1000 deaths every day. The UK has one of the highest prevalence and mortality rates in Europe. Over 5.4 million people in the UK have this complex condition which is often difficult to manage. Around 1400 people die each year in the UK and up to 70% of these deaths are preventable with better symptom control. In the UK alone costs are over £1.1B.

Uncontrolled asthma remains a stubbornly intractable problem. Outcomes have plateaued in the last 20 years despite major pharma investment in new drugs and costly drug delivery technologies and Metered Dose Inhalers (MDIs) remain the mainstay of treatment. Symptom control by inhaler devices can be adversely impacted by incorrect inhaler technique and poor adherence. Innovation has focused on Apps and 'smart inhalers' that prompt adherence (not inhalation technique), and monitor inhaler usage to try to improve symptom control.

Evidence of benefit from these first generation smart inhalers has been limited and reviews point to the need for a second generation to deliver clinical, physiological and behavioural data insights including the identification of inhalation technique and environmental interactions, to improve self management and ensure optimum symptom control. The latest smart inhalers and platforms deliver _some_ of this functionality but fail to identify, integrate and correct the critical steps of correct inhalation to support self management.

Building on the expertise gained building a previous CE marked inhalation device, our proposal will give detailed, real-time visualisation of each inhalation step, _for the first time._ Our R&D will deliver a cloud based, AI driven data aggregation and analytics platform to permit real time reporting to a new App and healthcare professional (HCP) dashboard, generate risk notifications and test in a synthetic population, ready for UK clinical trials.

The application of Machine Learning algorithms to the aggregated clinical data, air quality measurements, weather and epidemiology data, combined with clinical expertise, will support hyper personalised self management via data insights and identify 'at risk' patients for HCP intervention. Delivering these real-time data insights to users and HCPs will reduce the cost burden of avoidable HCP visits and hospital admissions.

Throughout the duration of the research we will work with closely with Asthma UK, clinicians and user focus groups to refine the design of the product and the app in an iterative way, to ensure user acceptability.