We're pursuing MDR Class IIb certification

We haven't written publicly about this before, so consider this an introduction and a look at where Ask Aletta is headed.

Mediocre certified beats great uncertified

Here is an uncomfortable truth about healthcare software: a mediocre certified product will beat a great uncertified one almost every time.

Certification does not automatically make a product good, but it does show that the manufacturer has met defined requirements for safety, quality and risk management. In healthcare, where patient safety is on the line, those assurances matter. For regulated clinical uses, certification is therefore not simply a preference but a condition of entry. Hospitals, GP practices and procurement departments cannot adopt an uncertified product for an intended use that requires compliance with the Medical Device Regulation.

From the beginning, we set out to build a product that is both great and certified.

Our first foundation: ISO 27001

We took an important first step by becoming ISO 27001 certified. Information security is fundamental to software that healthcare professionals must be able to rely on, particularly when that software will eventually work with clinical context and patient information.

Rather than adding security processes later, we wanted this foundation in place early, before beginning the next stage of our regulatory journey.

Where we are today: reactive

Today, Ask Aletta is a reactive AI knowledge platform: healthcare professionals ask a question, and Ask Aletta helps them find an answer.

They use the platform to search clinical guidelines, summarise documents, draft patient communication and support referral letters. All of this happens in the browser and begins with a request from the user. It is already very useful, as our users tell us every day, but it is not yet the product we ultimately set out to build.

Where we're going: proactive

The future of Ask Aletta is a proactive AI assistant that works inside a healthcare professional's HIS, EHR or patient communication platform. It will be embedded in the clinical workflow, with an understanding of the context in which the healthcare professional is working and the history of the patient in front of them.

With that context, Ask Aletta will help healthcare professionals prepare for consultations, support clinical decisions, draft documents and surface relevant information before they have thought to ask for it. This is how we want Ask Aletta to develop from a reactive knowledge platform into a proactive assistant within the care process.

Why that requires MDR certification

Whether software qualifies as a medical device depends on its intended purpose. The product we intend to build will provide information used to support diagnostic and therapeutic decisions for individual patients, bringing it within the scope of the EU Medical Device Regulation.

Under Rule 11 of the MDR, software that informs diagnostic or therapeutic decisions may be classified as Class IIb when those decisions could result in a serious deterioration in a person's health or a surgical intervention. Based on the intended purpose and risk profile of the product we are building, Class IIb is the pathway we are pursuing.

This requires us to demonstrate far more than technical performance. The process includes systematic risk management, clinical evaluation, quality management, verification, validation and continuous monitoring after the product enters the market.

For the product we envision, there is no responsible shortcut around that work. If we want Ask Aletta to operate inside the clinical workflow, at the moment important decisions are made, we must meet the standards that decisions of that weight demand. That is why we are pursuing MDR Class IIb certification and why we are doing the work now.

Why we think healthcare needs this

This is more than a product decision; it is a bet on what healthcare needs. Healthcare costs are rising as the population ages, while fewer healthcare professionals are being asked to care for more patients.

We believe AI can make a meaningful contribution in two areas:

• Quality of care. Making it easier for care to follow the guidelines and evidence that healthcare professionals have collectively agreed represent best practice.
• Efficiency. Supporting healthcare professionals directly inside their existing workflow, rather than asking them to interrupt that workflow to use another tool.

AI will not solve the pressures facing healthcare on its own. Applied responsibly, in the right place and with the right safeguards, however, it can help healthcare professionals spend more of their time on the work that only they can do.

Closer than you might think

This is not a distant roadmap item. We are already well into the process and aim to complete the Class IIb certification process within the next few months.

There is still important work ahead, and certification is never something to take for granted, but the direction is clear. We will share more about the process as we progress. For now, we wanted you to know what we are building towards and why.