Radiograph Accelerated Detection and Identification of Cancer in the Lung (RADICAL)

RADICAL: A Mixed Methods Study to Assess the Clinical Effectiveness and Acceptability of an Artificial Intelligence Software to Prioritise Chest X-ray (CXR) Interpretation

Lung cancer is the most common cause of cancer death in the UK yet compared to Europe it has low survival rates.The NHS aims to find 75% of cancers at an early stage as this can improve the chances of survival.

To support this target, Qure.ai have developed the UK-approved qXR product, which is a software program that automatically analyses chest x-rays using artificial intelligence to identify features associated with lung cancer, indicative of other diagnoses, or that contain no abnormal features ('normal'). qXR is a class IIb medical device that can be used by radiologists to prioritise reporting based upon the presence or absence of these features. This may improve the accuracy and efficiency of reporting these images.

The project includes different elements including:

i) Clinical effectiveness study across 3 sectors within NHS Greater Glasgow and Clyde (NHSGGC).The primary objective is to assess the clinical effectiveness of qXR to prioritise patients that have suspected lung cancer (identified from AI analysis of a chest x-ray) for follow-on CT.

Primary study outcome measure - Time to 'decision to recommend CT', or to a decision not to undertake CT for CXR acquired with USC (CXR acquired to CXR reported).

Secondary objectives include:

i) To assess the potential utility of qXR within the optimised lung cancer pathway in terms of the impact on both patient treatment and radiological workflow.

ii) A technical evaluation utilising retrospective and prospective cohorts. The technical retrospective study will determine the performance of qXR using a sample of 1000 CXR images from all chest x-ray referral sources across all sectors (this differs from the prospective study, which only examines outpatient referred chest x-rays).

iii) A health economic evaluation. Use of per patient healthcare utilisation costs to model cost benefits of qXR, including implementation of supported reporting of normal CXR.

iv) A qualitative evaluation to assess acceptability and barriers to scale-up and implementation

Study Overview

• Status: Active, not recruiting
• Conditions: Lung Cancer
• Intervention / Treatment: Other: qXR

Detailed Description

A clinical effectiveness study will be conducted in 3 NHS Greater Glasgow and Clyde sectors over a 12-month period.

Sectors will be identified and initiated into the qXR solution with a 30 day implementation period. The order in which sites will receive the qXR intervention will be determined by computer-based randomisation.

The technical retrospective study will determine the performance of qXR using a sample of 1000 CXR images from all chest x-ray referral sources across all sectors (this differs from the prospective study, which only examines outpatient referred chest x-rays). An economic evaluation will be conducted comparing costs and outcomes with and without the introduction of qXR. The software potentially impacts costs via two mechanisms: the identification of normal can enhance efficiency of CXR reporting; and the identification of USCs can support the prioritisation of CXRs that show signs of lung cancer, accelerating the provision of CT, which leads to faster diagnosis and treatment, and ultimately better outcomes.

Qualitative evaluation: To determine acceptability, staff interviews and patient focus groups will be carried out.

Data will be collected by an experienced qualitative researcher using a semi-structured interview guide, developed based on the key constructs of the Theoretical Framework of Acceptability. All interviews will be conducted via Zoom at a mutually agreed upon date and time and are estimated to last, on average, around 45 minutes.

To capture the NHS service user perspective, the investigators will also conduct three online focus groups with approximately 20 NHS service users.

• Study Type: Observational
• Enrollment (Estimated): 60000

Contacts and Locations

Study Locations

• United Kingdom
  • Glasgow, United Kingdom
    • NHS Greater Glasgow and Clyde
  • Glasgow, United Kingdom
    • Glasgow Royal Infirmary (North Sector)
  • Glasgow, United Kingdom
    • Queen Elizabeth University Hosp (South Sector)
  • Paisley, United Kingdom
    • The Royal Alexandra Hospital (Clyde Sector)

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Service deployment within a single health board (NHS Greater Glasgow and Clyde), within the following sectors and hospitals:

Clyde Sector

• The Royal Alexandra Hospital
• Inverclyde Royal Hospital
• Vale of Leven Hospital

South Sector

• Queen Elizabeth University Hospital
• New Victoria Hospital
• Gartnavel General Hospital
• West Glasgow Ambulatory Care Hospital

North Sector

• Glasgow Royal Infirmary
• Stobhill Hospital

At each participating site, the investigators will identify all patients referred through the outpatient pathway (including GP-referrals). This will include those with suspected lung cancer referred for CXR.

Description

Inclusion Criteria:

• Unconsented patients ≧ 18 years old with frontal chest radiograph, acquired consecutively during usual care through the outpatient (including GP) referral pathway only, whose radiograph has not already been reported (applies to clinical effectiveness and health economic evaluation studies).
• Unconsented patients ≧ 18 years old with frontal chest radiograph, sampled from images already acquired and reported in the current or previous calendar year (applies to technical evaluation).
• Key stakeholders such as NHS service users, healthcare staff and NHS management (applies to qualitative evaluation).

Exclusion Criteria:

• Patient has requested that they are removed from the study, or has objected to the use of AI in their routine clinical care and this has been subsequently upheld by the health board (applies to clinical effectiveness study, health economic evaluation and technical evaluation).

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Service deployment; Intervention Chest X-ray received - care team (standard of care) CT scan - care team (standard of care)	Other: qXR; a software product that uses artificial intelligence to triage, prioritise, and (for tuberculosis only) diagnose based upon identified abnormalities within the CXR.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Time to 'decision to recommend CT', or to a decision not to undertake CT for CXR acquired with USC (CXR acquired to CXR reported)	Time to 'decision to recommend CT', or to a decision not to undertake CT for CXR acquired with USC (CXR acquired to CXR reported)	through study completion, an average of 1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Time from acquisition to reporting of all CXRs	Time from acquisition to reporting of all CXRs	through study completion, an average of 1 year
Time to diagnosis of lung cancer	Time to diagnosis of lung cancer	through study completion, an average of 1 year
Time to treatment initiation lung cancer	Time to treatment initiation lung cancer	through study completion, an average of 1 year
Number of hospital visits during screening pathway	Number of hospital visits during screening pathway	through study completion, an average of 1 year
Hospitalisation within 6 and 12 months CXR acquisition	Hospitalisation within 6 and 12 months CXR acquisition	through study completion, an average of 1 year
Death within 6 and 12 months of CXR acquisition	Death within 6 and 12 months of CXR acquisition	through study completion, an average of 1 year
Percentage of CXRs not identified by qXR as suspected lung cancer that the radiologist refers for CT for USC	Percentage of CXRs not identified by qXR as suspected lung cancer that the radiologist refers for CT for USC	through study completion, an average of 1 year
Percentage of non-USC that are referred for CT with subsequent detection of lung cancer	Percentage of non-USC that are referred for CT with subsequent detection of lung cancer	through study completion, an average of 1 year
Model performance e.g. sensitivity, specificity, positive and negative predictive values.	Model performance e.g. sensitivity, specificity, positive and negative predictive values.	through study completion, an average of 1 year

Collaborators and Investigators

• Sponsor: NHS Greater Glasgow and Clyde
• Collaborators: Qure.ai Technologies Pvt. Ltd
• Investigators: Principal Investigator: David Lowe, NHS Greater Glasgow and Clyde Board HQ

Publications and helpful links

General Publications

• Duncan SF, McConnachie A, Blackwood J, Stobo DB, Maclay JD, Wu O, Germeni E, Robert D, Bilgili B, Kumar S, Hall M, Lowe DJ. Radiograph accelerated detection and identification of cancer in the lung (RADICAL): a mixed methods study to assess the clinical effectiveness and acceptability of Qure.ai artificial intelligence software to prioritise chest X-ray (CXR) interpretation. BMJ Open. 2024 Sep 20;14(9):e081062. doi: 10.1136/bmjopen-2023-081062.

Study record dates

Study Major Dates

• Study Start (Actual): December 4, 2023
• Primary Completion (Estimated): November 30, 2025
• Study Completion (Estimated): November 30, 2025

Study Registration Dates

• First Submitted: August 29, 2023
• First Submitted That Met QC Criteria: September 12, 2023
• First Posted (Actual): September 21, 2023

Study Record Updates

• Last Update Posted (Estimated): September 19, 2025
• Last Update Submitted That Met QC Criteria: September 15, 2025
• Last Verified: September 1, 2025

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Respiratory Tract Diseases; Lung Diseases; Respiratory Tract Neoplasms; Thoracic Neoplasms; Lung Neoplasms
• Other Study ID Numbers: INGN23RM028; 23/NW/0211 (Other Identifier: Ethics Committee)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No