This site is optimised for modern browsers. For the best experience, please use Google Chrome, Mozilla Firefox, or Microsoft Edge.

Artificial Intelligence and Racial and Ethnic Inequalities in Health and Care - Shortlisting Minutes

Contents

Minutes of the Artificial Intelligence and Racial and Ethnic Inequalities in Health and Care Shortlisting Meeting.

Date and location

Minutes of meeting held on Wednesday 26 May 2021 via Zoom Virtual Meeting.

In attendance

Funding Panel Members

  • Professor Peter Bannister (Chair)
  • Dr Naomi Lee (Deputy Chair)

AI Programme Management Office

  • Dr Ruth Nebauer (Assistant Director)
  • Dr Andrew Barber
  • Dr Silvana Cossins
  • Dr Michelle Edye
  • Dr Aayesha Hassan
  • Mrs Paula Milton
  • Miss Deimante Normantaite
  • Dr Raffaella Roncone
  • Mr Edwin Tucker

Operations Team

  • Miss Lily Tomlinson
  • Mrs Samantha Wade

NHSX 

  • Ms Brhmie Balaram
  • Mr Laurence Thorne

Health Foundation

  • Mr Tom Hardie
  • Mr Josh Keith
  • Dr Mai Stafford

Observers

  • Ms Ami Hodges (NIHR)
  • Ms Elizabeth Noble (NIHR)
  • Mr Alex Zabala Findlay (NIHR)

Applications to be invited to interview

AI_HI200004: Ethnic fairness in clinical prediction models used for decision support
Outcome: Invite to interview

AI_HI200006: I-SIRch - Using Artificial Intelligence to Improve the Investigation of Factors Contributing to Adverse Maternity Incidents involving Black Mothers and Families
Outcome: Invite to interview

AI_HI200007: Understanding AI and racial inequalities for cardiovascular risk assessment
Outcome: Invite to interview

AI_HI200008: Ethnic differences in performance and perceptions of Artificial Intelligence retinal image analysis systems for the detection of diabetic retinopathy in the NHS Diabetic Eye Screening Programme
Outcome: Invite to interview

AI_HI200014: STANDING together: STANdards for Data INclusivity and Generalisability
Outcome: Invite to interview

AI_HI200017: Mind the Racial Gap(mrGAP): Unravelling Racial and Ethnic Disparities in Development of Multimorbidity
Outcome: Invite to interview

AI_HI200018: RAINBO: An appRoach for modelling observational dAta to Improve ethNicity coding and Build representative models suitable for ethnic groups
Outcome: Invite to interview

AI_HI200021: Fat CAD AI - A Computer-Aided Detection Artificial Intelligence Approach for Routinely Extracting Abdominal Fat from Computed Tomography Images to Reveal and Mitigate Racial Health Inequalities in Metabolic Syndrome
Outcome: Invite to interview

AI_HI200024: What does blood pressure mean to me?
Outcome: Invite to interview

AI_HI200026: Using machine learning to develop a decision support tool for GPs to address the healthcare needs of Black, Asian and other ethnic minority patients in the community
Outcome: Invite to interview

AI_HI200028: Assessing acceptability, Utilisation and Disclosure of health Information to an automated chatbot for advice about sexually Transmitted infections in minoritisED ethnic populations - AUDITED
Outcome: Invite to interview

AI_HI200029: Co-Designing AI Enhanced Wellbeing Technologies to Support Refugee Communities
Outcome: Invite to interview

AI_HI200031: Ethnic Bias AI toolkit, score card, benchmark datasets and their application within the NHS
Outcome: Invite to interview

AI_HI200035: Remote technologies for Parkinson's disease care in Black minority patients
Outcome: Invite to interview

Applications not invited to interview

AI_HI200001: Removing skin colour bias in AI eczema severity scoring
Outcome: Reject

AI_HI200002: Using an AI model to improve self management for BAME citizens
Outcome: Reject

AI_HI200003: Bias and discrimination monitoring and mitigation software for healthcare AI
Outcome: Reject

AI_HI200005: Developing an artificial intelligence-enabled, ethnically sensitive algorithm to assist the identification of avoidable hospitalisations at the end-of-life
Outcome: Reject

AI_HI200009: Can Artificial Intelligence be used to reduce differences in cardiovascular health outcomes for underserved groups?
Outcome: Reject

AI_HI200010: QUMIE: a framework for QUantifying and Mitigating racial InEquality of AI in Medicine
Outcome: Reject

AI_HI200011: Process Factors and Adverse Outcomes for Ethnic Minorities Receiving Care in NHS Trusts
Outcome: Reject

AI_HI200012: A feasibility assessment using AI technologies to detect and categorise early-stage Pressure Ulcers (PU’s) on patients with darker skin
Outcome: Reject

AI_HI200013: Automatic Speech assessment of Cognition in Minority Ethnic groups (ASC-ME)
Outcome: Reject

AI_HI200015: Artificial Intelligence for improving Categorised Ethnicity in anonymised datasets (AICE)
Outcome: Reject

AI_HI200016: Fairness by Design: Governing Data-Driven Artificial Intelligence to Advance Racial Equity in Mental Health
Outcome: Reject

AI_HI200019: Responsible Machine Learning-Based Risk Prediction Modelling for Equitable Health Outcomes
Outcome: Reject

AI_HI200020: Addressing bias in artificial intelligence: development, evaluation and adoption of fair and intelligent clinical decision support systems for infection management
Outcome: Reject

AI_HI200022: Improving AI development of a computer-aided risk scoring system (CARSS) to address racial and ethnic inequalities in hospital care
Outcome: Reject

AI_HI200023: A Holistic approach to data collection and AI-training for the identification of Skin Lesions
Outcome: Reject

AI_HI200025: InVisAI: An AI-backed Visual Analytics System for Bias Tracking in Diabetes Outcome Prediction
Outcome: Reject

AI_HI200027: Using AI to address racial and ethnic inequalities in mental health support during pregnancy and early parenthood
Outcome: Reject

AI_HI200030: Reducing the risk of bias against underrepresented groups by designing software tools as well as non-technical strategies that can be implemented in the entire AI lifecycle
Outcome: Reject

AI_HI200032: Understanding ethnic biases in NHS datasets and developing a machine learning pipeline for building predictive models with better predictive power for ethnic minorities
Outcome: Reject

AI_HI200033: Understanding and detecting unequal representation in medical datasets to inform policy development and best practice (FATEful Study)
Outcome: Reject

AI_HI200034: Developing methods for evaluating the fairness of risk models in healthcare AI and identifying dataset improvements (CIPHER – EQ)
Outcome: Reject

AI_HI200036: Technical Handling of Equality for Minorities as Individuals and Social Societies (THEMIS)
Outcome: Reject

AI_HI200037: Applying user-centred design to support lifestyle behaviour change using conversational AI for a minority ethnic population: mi healthcoach
Outcome: Reject

AI_HI200038: Application of AI safety technology to identify and correct instances of racial bias present in CVD risk prediction algorithms
Outcome: Reject

AI_HI200039: Minimising the impact of biases in the end-to-end development of AI technologies that include modelling the health needs of minority ethnic groups through standardised checklists, sample tools and accessing additional data
Outcome: Reject