The NHS AI Lab

External Resource

The Guide to Good Practice for data-driven health and care technology

The guide sets out the behaviours the government expects from those deploying and using data-driven technologies in health and care

External Resource

Explaining decisions made with AI

This guidance by the ICO and The Alan Turing Institute aims to give organisations practical advice to help explain the processes, services and decisions delivered or assisted by AI, to the individuals affected by them

External Resource

Explainability in data-driven health and care technology

Future Advocacy was commissioned by NHSX to produce this guidance on Principle 7 of the NHS Code of Conduct on data-driven health and care technology, which relates to the explainability of algorithms

External Resource

Evidence standards framework for digital health technologies

NICE’s standards aim to make it easier for innovators and commissioners to understand what good levels of evidence for digital healthcare technologies look like

Case Study

Data Lens: a fast-access data search in multiple languages

A pilot project for the NHS AI Lab Skunkworks team, Data Lens brings together information about multiple databases, providing a fast-access search in multiple languages

Case Study

Using machine learning to identify patients at risk of long term hospital stays

Case study: Investigating the benefits of artificial intelligence applied to hospital admissions data

Report

AI Award research contract

This document provides the NHSX funding agreement for AI in Health and Care Award Phases 3 and 4.

Guidance

AI regulation guide: considerations when developing AI products and tools

A guide from the the AI and digital regulations service on the importance of the value proposition, determining a product’s intended use, and planning ahead to navigate the path to market.

Guidance

AI regulation guide: using PICO to generate evidence for AI development

A guide from the AI and digital regulations service on generating the right evidence for your AI product for health and care, and mastering the PICO statement.

Digital Technology Assessment Criteria (DTAC)

The DTAC brings together legislation and good practice. The DTAC helps healthcare organisations assess suppliers to make sure new digital technologies meet our standards. For developers, it sets out what is expected for entry into the NHS

Case Study

Using AI to identify tissue growth from CT scans

This 12-week artificial intelligence project investigated whether it was possible to identify organs and growths from CT scans and report on any changes to support radiologists.

External Resource

Blog: Lessons learned from the multi-agency advisory service – helping developers of AI and data-driven tech to navigate the regulatory pathway

This blog post from NICE outlines the common challenges that developers of artificial intelligence tech face when navigating their way to the health and care market.

Case Study

Improving hospital bed allocation using AI

This 14-week project investigated the AI techniques that could be used to generate options for moving patients in a way that supports the human team to make the best decisions. The project aimed to provide a proof of concept tool.

Resource Collection

AI in imaging: resource collection

A collection of resources about the use of medical imaging data in artificial intelligence tools for health and social care.

External Resource

An overview of the National COVID-19 Chest Imaging Database: data quality and cohort analysis

An academic paper published 25 November 2021

External Resource

Towards nationally curated data archives for clinical radiology image analysis at scale

Academic paper giving learnings from national data collection in response to the COVID-19 pandemic. Published 23 November 2021.

External Resource

Algorithmic impact assessment: a case study in healthcare

This Ada Lovelace Institute report sets out the first-known detailed proposal for the use of an algorithmic impact assessment for data access in a healthcare context.

External Resource

Generating high-fidelity synthetic patient data for assessing machine learning healthcare software

An article exploring the production of realistic synthetic data based on UK primary care patient data. Published in Nature.com, November 2020.

External Resource

Practical lessons from generating synthetic healthcare data with Bayesian networks

Academic paper from the MHRA and the Intelligent Data Analysis Group at Brunel University.

External Resource

Good machine learning practice for medical device development: guiding principles

These guiding principles from the MHRA will help promote safe, effective, and high-quality medical devices that use artificial intelligence and machine learning.

External Resource

Generating and evaluating cross-sectional synthetic electronic healthcare data

Academic paper from the MHRA and Brunel University on preserving data utility and patient privacy with synthetic data.

External Resource

BayesBoost: Identifying and handling bias using synthetic data generators

External research paper from MHRA and Brunel University exploring attempts to create synthetic data that lead to predictive models with better performance.

Case Study

NCCID case study: Setting standards for testing Artificial Intelligence

This blueprint assessment process is a valuable example of the quality testing that needs to take place for all AI algorithms designed for health and care.

Case Study

Exploring how to create mock patient data (synthetic data) from real patient data

This case study looks at one process for creating “synthetic data” - where new “fake” data is generated from real patient data using a specifically designed model, in a way that maintains several characteristics of the original data.

External Resource

Evidence Standards Framework

NICE's Evidence Standards Framework for Digital Health Technologies. It offers support and guidance to developers so they can generate high quality evidence. It is closely aligned to regulatory requirements and is easy to use.

Case Study

Using AI to find optimal placement schedules for nursing students

This case study looked at whether AI could generate student nurse placement schedules that provide the requirements needed by the hospital, while also giving students diverse range of placements.

Case Study

Using deep learning to detect adrenal lesions in CT scans

Deep learning to detect adrenal lesions in CT scans

Case Study

Examining whether recruitment data can, and should, be used to train AI models for shortlisting interview candidates

Examining whether recruitment data can, and should, be used to train AI models for shortlisting interview candidates

Case Study

Identifying and quantifying Parkinson’s Disease using AI on brain slices

Identifying and quantifying Parkinson’s Disease using AI on brain slices

Case Study

Working with a Trusted Research Environment

Partnership project to gain experience of using a TRE to answer key and important questions for a defined project. OpenSAFELY, an open-source software platform for analysis of electronic health care records data, was used for COVID-19 research.