IDEAL: Artificial Intelligence and Big Data for Early Lung Cancer Diagnosis Study

NCT03753724 | Completed

• 1 other identifier: 13489
• Study Type: observational
• Target: 1,293
• Locations: 1 country
• Sites: 4

Brief Summary

This study aims to test the use of novel CT image analysis techniques to enable a better characterisation of small pulmonary nodules. The study will incorporate solid and predominantly solid nodules of 5-15 mm scanned using a variety of scanner types, imaging protocols and patient populations. The investigators hope that the new image processing techniques will improve the accuracy of lung nodule analysis which will in turn reduce the number of unnecessary investigations for benign nodules and may increase the accuracy of the early diagnosis of lung cancer in malignant nodules. This study aims to test this novel analysis software to subsequently allow validation.

Conditions

Pulmonary Nodule, Solitary; Pulmonary Nodule, Multiple; Lung Cancer

Keywords

pulmonary nodules; lung cancer; artificial intelligence; CAP model

Outcome Measures

Primary Outcomes (1)

• The overall diagnostic performance of a new computer aided prediction (CAP) model for malignancy in small pulmonary nodules (% diagnostic accuracy).

  Area Under the Receiver Operator Characteristic Curve (AUC).

  Up to 1 year.

Secondary Outcomes (2)

• The health economic benefits of the CAP model.

  At 2 weeks, 3 months (group 2 & 3 only) and year 1.

• The diagnostic performance of the CAP model for malignancy in small pulmonary nodules at a specific operating point relevant to clinical practice.

  Up to 1 year.

Study Arms (3)

Group 1

On review of the scans by an expert, no further follow up or investigation is required, as the nodule(s) has been categorised as benign. As the patient was unaware the scan was being reviewed and no further investigations are required. Patient is invited to take part in the study (by telephone).

Group 2

On review, the nodule is indeterminate and further scanning at a later date is required. The patient is informed of this, usually via a telephone call from either the doctor or nurse specialist working in the Lung Nodule Clinic (LNC) or site equivalent. This LNC is usually a virtual clinic - no physical interaction with the patient - and the follow up scan is reviewed and the patient contacted again via the virtual clinic. Patient is invited to take part in the study (by telephone, by post or in clinic if appropriate).

Group 3

On review, the nodule is regarded as potentially malignant and further scans and a clinic appointment is made for the patient. Patient is invited to take part in the study (in clinic if appropriate).

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Patients aged 18 years old or older with CT scans reported as having pulmonary nodule(s) of 5-15mm who meet the inclusion criteria and who can be placed into groups 1, 2 or 3.

You may qualify if:

• Male or Female, aged 18 years or above
• CT scans identified as having pulmonary nodule(s) of 5-15mm
• Patients with solid or predominantly solid nodules referred to the pulmonary nodule clinic or for CT scan review by a specialist
• CT scan section thickness of 3mm and less

You may not qualify if:

• The CT scans are technically inadequate
• Having received treatment for cancer in the last 5 years
• Patient has more than five reported qualifying nodules

Sponsors & Collaborators

• University of Oxford: lead
• Optellum Ltd.: collaborator

Study Sites (4)

Royal Berkshire Nhs Foundation Trust

Reading, Berkshire, RG1 5AN, United Kingdom

Location

Nottingham University Hospitals Nhs Trust

Nottingham, Nottinghamshire, NG7 2UH, United Kingdom

Location

Oxford University Hospitals NHS Foundation Trust

Oxford, Oxfordshire, OX3 7LE, United Kingdom

Location

Leeds Teaching Hospitals Nhs Trust

Leeds, West Yorkshire, LS9 7TF, United Kingdom

Location

MeSH Terms

Conditions

Solitary Pulmonary Nodule; Multiple Pulmonary Nodules; Lung Neoplasms

Condition Hierarchy (Ancestors)

Lung Diseases; Respiratory Tract Diseases; Respiratory Tract Neoplasms; Thoracic Neoplasms; Neoplasms by Site; Neoplasms

Study Officials

• Fergus Gleeson, Prof

  University of Oxford/Oxford University Hospitals NHS Foundation Trust

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: PROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: August 15, 2018
• First Posted: November 27, 2018
• Study Start: August 29, 2018
• Primary Completion: March 31, 2022
• Study Completion: March 31, 2022
• Last Updated: July 29, 2022

Record last verified: 2022-07

Locations

GB (4)