Connect with us


Diagnosis and chill? The rise of AI health tech subscription services

By HGF Patent Director, Janine Swarbrick



Artificial intelligence (AI) looks set to revolutionise the healthcare industry.

By automating tedious and time-consuming tasks, and having the capability to identify trends and connections in vast amounts of patient data, AI has the potential for good, to speed up diagnosis, improve patient outcomes, free up clinician time, and reduce costs.

Healthcare providers and HealthTech companies are increasingly turning to AI to improve the accuracy and efficiency of diagnosis and treatment.

This article explores some of the current trends in using AI in healthcare and how subscription services are changing the way healthcare organisations access and use AI.

Medical image analysis is well suited to the application of AI. Medical imaging is a critical tool for diagnosis and treatment of a wide range of medical conditions, but analysis can be a time-consuming and labour-intensive process, often requiring input from experienced clinicians who are already stretched to draw meaningful conclusions from the images.

AI can be deployed to automate the analysis of medical images, freeing up healthcare professionals to focus on other tasks and potentially improving the accuracy of diagnosis and even diagnosing conditions much earlier than can be reliably detected by human analysis.

AI is also capable of drawing trends and identifying correlations between vast amounts of data, whether that is patient health records, test results, environmental data or sensor data.

This provides ways for healthcare providers to extract valuable insights and patterns in data which may otherwise go unnoticed, take far longer to identify, or simply involve too much data for humans to meaningfully process in a timely way.

By uncovering hidden relationships and trends within the data, AI can aid in the identification of disease risk factors, early detection of medical conditions, and the development of personalised treatment plans, based in an informed way on the evidence available.

The potential for AI to contribute to preventive measures in health care, and enable proactive early stage intervention, can lead to much better health outcomes and crucially in the current economic climate, reduced healthcare costs

There are seemingly an ever increasing number of AI-based companies and SMEs working in the healthcare sector.

One theme which stands out from is the provision of AI services as subscription-based services.

The use of subscription services is changing the way healthcare organisations access and use AI based tools.

Traditionally, healthcare providers may have had to purchase expensive software licences to use a company’s software, and challenges in making the licensed software available to the right people in the clinical setting, as well as the administrative burden of maintaining the most recent release of the software, can act as a deterrent to use.

However, subscription services are now becoming increasingly popular.

Such services allow healthcare organisations to access AI technology from the provider on a pay-as-you-go basis, per patient or per location for example, making it more affordable and flexible to access the technology.

The end user has the freedom to access the software from different locations, devices, how and when they want to.

A business model maintaining the software at the provider side can also be attractive for the provider company who can retain greater control over how and by whom their software is used.

Providing AI based, or other software, services can provide challenges when managing the intellectual property residing in the software.

Making the software available to the end user can be undesirable, as there is a risk of a competitor being able to access the code and re-code the functionality themselves making it particularly challenging to argue your copyright in the code is infringed.

Perhaps more serious is the possibility of a competitor reverse engineering your code to access the know-how built up by your company and embodied in the software functionality, essentially giving away your companies hard-earned market advantage for free.

Importantly, allowing a business to retain control over the software at their side, rather than providing it to the healthcare provider or end user, facilitates greater control over the IP in the software because of the limited distribution of the software to the wider population, giving companies the confidence that their valuable business know-how in the software is less likely to be taken and copied.

This can be a particular concern with AI based software, whereby the software is often trained for a particular task by the company using a training data set obtained or curated by the company, thereby also holding valuable company know-how in the training data as well as the end AI model.

Retaining the software at the company side and providing subscription access to customers prevents distribution not only of the functional software used to obtain results, but potentially also allows for the training data set to be more securely held by the company without distribution to customers.

Software, including AI software, can often be patent protected in many jurisdictions if it provides a “technical effect”, which can often be demonstrated in the field of healthcare with careful patent application drafting.

Patents can provide a powerful business asset in protecting your innovation and preventing others from providing technology within the scope of your granted claims.

Some companies are hesitant to consider patent protection for their AI models, as they would rather not disclose what their code does or how it works in a patent disclosure, even for a monopoly over that technology for the life of the patent, or would not want to enter into litigation or negotiations if the patent was felt to be infringed by a competitor.

It is important to remember that it is often not necessary to disclose the actual code in a patent application nor the data used as training data. It may not even be necessary to disclose the particular algorithm found to provide the best results if one variation of a model works especially well, from a family of related models which all provide the technical effect claimed in the patent application.

What a patent application requires is disclosure of sufficient information for a “person skilled in the art” – for example an AI software engineer, or a team of a clinician and computer programmer – to be able to reproduce the invention without undue burden.

In short, offering AI powered healthcare software as a subscription service can offer strategic advantages, including both intellectual property and business model benefits.

From an IP viewpoint, a subscription model can allow a healthcare company to retain ownership and control over their proprietary algorithms and trained models, while providing access to the latest advancements in AI powered healthcare.

Furthermore, the subscription based business model allows for companies to offer a flexible and scalable approach for customers, eliminating the need for significant upfront investment in software licences or infrastructure.

By allowing subscription to healthcare software on a pay-as-you-go or tailored subscription package basis, the customer can access the company’s technology in a way meeting their specific needs.

By combining the IP advantages with the flexibility allowed via offering the software as a subscription service, healthcare companies can offer AI-driven innovations in an attractive way both for the provider and the customer, to potentially transform the landscape of healthcare delivery for patients, clinicians and companies alike.

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Trending stories