Artificial Intelligence Applications in Healthcare

Introduction 

Artificial intelligence (AI) is becoming more present in society, with wide ranging applications from speech recognition to self-driving vehicles.  While there may be debate around the ethical and moral consequences of some AI applications, one area that will undoubtably benefit is healthcare.  AI has the potential to revolutionise healthcare with applications in several areas including diagnostics, predicting treatment responses and therapy research.  The advent of a highly accurate, rapid, on-demand diagnostic assessments, coupled with new targeted delivery of new therapies could help reduce treatment waiting times, detect diseases at an early stage and improve outcomes  

This article will explore the patent landscape surrounding AI in healthcare using PatWorld (global patent search database). 

Method 

A dataset was generated by searching AI keywords (including synonyms such as machine learning, neural networks and deep learning) within the A61 - Medical Or Veterinary Science; Hygiene classification.  Patents/applications were analysed using PatWorld software to generate charts and graphs used in the report.   

Filing trends 

AI use in healthcare is a relatively new, emerging field that has seen nearly exponential growth in recent years (Figure. 1).  Priority filings appear to increase significantly from 2015 (N.B. priority filings from 2021 and 2022 were not included as these figures may be inaccurate due to applications yet to be published claiming priority in these years).   

Figure 1: Earliest priority year vs number of patent families for AI in healthcare. 

The main area of innovation appears to be in diagnostics.  Figure 2 depicts a pie chart of the top CPC classifications in the area.  A61B5/00 - Measuring for diagnostic purposes appears to be the most prevalent classification, followed by G16H50/20 - for computer-aided diagnosis.   Figure 2 also shows A61B5/7267 and /7264, which are related to “Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems” and several G classifications that are related to AI in general. 

There appears to be several areas within diagnostics where AI could be implemented, some of which include: 

• Online diagnostic tool using a patient questionnaire (US10478112B2 – Harvard College)  
• Deep learning based diagnosis of lung disorders from x-ray images (US2021042916A1 – AI Tech and The Regents of The University Of California)  
• Rapid assessment and outcome analysis using machine-learning classifiers (US10825167B2 – Siemens Healthcare)  

Comments ({{count}})

{{comment.user.full_name}}

{{getTime(comment.created_at)}}

{{comment.message}}

Replies: {{comment.comments_count}}

Likes {{comment.likes_count}}

Likes {{comment.likes_count}}

Reply

Close

{{reply.user.full_name}}

{{getTime(reply.created_at)}}

{{reply.message}}

Likes {{reply.likes_count}}

Likes {{reply.likes_count}}

Submit

There are currently no comments. Be the first to comment on this article

Load more +

Want to leave a Comment? Register now.

Register Login