Skip to main content

Overview

The 2020s have ushered in a revolutionary era of artificial intelligence (AI), transforming various sectors, including healthcare. For the NHS, AI tools offer immense potential to enhance productivity, improve decision-making, and ultimately elevate the quality of patient care.

In October 2023, the NHS announced a £21 million investment in AI tools for lung cancer diagnosis, demonstrating a commitment to leveraging AI technology1. This investment aims to support faster diagnosis by making clinical review more efficient, potentially raising the rate of meeting the two-week diagnosis target from the current 75% back to pre-pandemic levels of over 90%.

The NHS AI Lab, created to address the challenges of implementing AI in healthcare, brings together government, health and care providers, academics, and technology companies. The lab focuses on several key areas:

  1. Developing imaging technology
  2. Safely collecting and sharing data
  3. Validating AI imaging software

However, it's important to note that the implementation of AI in the NHS faces several challenges. Dr. Farzana Rahman, a radiologist, points out that many radiology departments still use paper-based or telephone-based systems with siloed Picture Archiving and Communications Systems (PACS) and Radiology Information Systems (RIS). This lack of digital infrastructure makes it difficult to fully utilize AI tools. Dr. Rahman suggests that digitizing the NHS's diagnostic services to make them AI-ready might have been a better initial use of government funding2.

Understanding AI​

Artificial Intelligence (AI) refers to the simulation of human intelligence in machines designed to think and learn like humans. AI encompasses a range of technologies, including machine learning (ML), natural language processing (NLP), and large language models (LLMs)3.

Machine Learning​

Machine Learning is a subset of AI that involves the development of algorithms that enable computers to learn from and make decisions based on data. Unlike traditional programming, where explicit instructions are provided, ML algorithms improve their performance over time as they are exposed to more data4. In healthcare, ML can be used for predictive analytics, such as predicting patient outcomes, identifying risk factors for diseases, and optimizing treatment plans.

Natural Language Processing (NLP)​

Natural Language Processing is a field of AI that focuses on the interaction between computers and human language. NLP enables machines to understand, interpret, and generate human language in a way that is both meaningful and useful5. Applications of NLP in healthcare include voice recognition for transcribing medical notes, sentiment analysis for patient feedback, and automated summarization of medical documents.

Large Language Models (LLMs)​

Large Language Models, such as GPT-4, are advanced AI systems trained on vast amounts of textual data. These models can understand and generate human-like text, making them capable of performing a variety of tasks, including language translation, text summarization, and creative writing6. LLMs are particularly useful in healthcare for generating patient reports, providing diagnostic assistance, and synthesizing medical research.

The Hype Around AI​

The excitement around AI and LLMs stems from their ability to automate complex tasks, provide data-driven insights, and significantly reduce the time and effort required for various processes7. In healthcare, these technologies can revolutionize how information is accessed, analyzed, and utilized, leading to more efficient and effective patient care.

In medical imaging specifically, AI has demonstrated remarkable abilities in detecting and diagnosing diseases, often matching or even surpassing human experts in certain tasks. For example:

  • AI algorithms have shown high accuracy in detecting lung abnormalities, including potential cancers, from chest X-rays and CT scans. In some studies, AI systems have outperformed radiologists in identifying early-stage lung cancers8.
  • In ophthalmology, AI-powered systems like Google's Automated Retinal Disease Assessment (ARDA) have been successful in detecting diabetic retinopathy, a leading cause of blindness. This technology has shown promise in assisting doctors to prevent vision loss in millions of patients with diabetes9.
  • AI has also been applied to dermatology, helping to improve access to skin disease information and potentially assisting in the early detection of skin cancers9.
  • In breast cancer screening, AI tools like Mia have been developed to analyze mammograms more efficiently, potentially reducing the workload of radiologists and improving the accuracy of diagnoses10.

However, it's important to note that while AI shows great promise, it is not intended to replace healthcare professionals. Instead, the goal is to augment and assist medical practitioners in providing more accurate, efficient, and personalized care to patients11.

AI Powered Tools​

AI-powered tools can significantly boost efficiency within the NHS by streamlining workflows and reducing administrative burdens. Here’s how:

  • Perplexity AI is a sophisticated search assistant capable of quickly synthesizing information from multiple sources. This tool helps to find relevant research, guidelines, and medical literature more efficiently.
  • Fireflies.ai serves as an AI notetaker to help transcribe, summarize, search, and analyze voice conversations and meetings. This tool can be particularly useful for documenting team meetings, and is able to integrate with Microsoft Teams.
  • Krisp is an AI tool that improves voice audio quality and can transcribe and summarize meetings.
  • Folk is an AI tool for managing contacts and relationships across multiple platforms such as Outlook, Gmail, and LinkedIn.
  • Gamma App assists in creating presentations and other visual content. It can help to produce high-quality presentations more efficiently by automating the process of data visualization and presentation design, allowing for more focus on delivering content rather than spending excessive time on formatting.
  • Phind is an AI-assisted search engine and assistant for programmers, similar to Perplexity.AI.
  • Otter.ai for real-time transcription and collaborative note-taking during meetings.
  • Notion AI for organizing and managing complex projects, workflows, and knowledge bases within teams.

Footnotes​

  1. Woolley, Nicholas. Caldwell-Jones, Gabriela. Bar-Katz, Vladislava. "The NHS has spent £21m on AI tools. Will they be worth it?" Frontier Economics, 2024, Link ↩

  2. Rahman, Farzana. "A reality check for AI tools in the NHS" Open Access Government, 2024, Link ↩

  3. Anonymous. "What is artificial intelligence (AI)?" IBM, Link ↩

  4. Anonymous. "What is machine learning (ML)?" IBM, Link ↩

  5. Holdsworth, Jim. "What is NLP?" IBM, 2024, Link ↩

  6. Anonymous. "What are LLMs?" IBM, Link ↩

  7. Anonymous. " Using AI to improve back office efficiency in the NHS" NHS England, 2022, Link ↩

  8. Ghaffar Nia, Nafiseh. Kaplanoglu, Erkan. Nasab, Ahad. "Evaluation of artificial intelligence techniques in disease diagnosis and prediction" National Library of Medicine, 2023, Link ↩

  9. Anonymous. "AI-enabled imaging and diagnostics previously thought impossible" Google Health, Link ↩ ↩2

  10. Anonymous. "Mia mammography intelligent assessment" NHS England, 2021, Link ↩

  11. Anonymous. "How AI-Powered Medical Imaging is Transforming Healthcare" Onix, Link ↩