Working…
COVID-19 is an emerging, rapidly evolving situation.
Get the latest public health information from CDC: https://www.coronavirus.gov.

Get the latest research information from NIH: https://www.nih.gov/coronavirus.
ClinicalTrials.gov
ClinicalTrials.gov Menu

MRI Incidental Finding Deep Learning Identification (MIDI)

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been evaluated by the U.S. Federal Government. Know the risks and potential benefits of clinical studies and talk to your health care provider before participating. Read our disclaimer for details.
 
ClinicalTrials.gov Identifier: NCT04368481
Recruitment Status : Recruiting
First Posted : April 29, 2020
Last Update Posted : April 29, 2020
Sponsor:
Information provided by (Responsible Party):
King's College Hospital NHS Trust

Brief Summary:

Unexpected abnormalities with potential clinical relevance (incidental findings) occur in 2.7% of head MRI studies. There is a wide variation in how incidental findings (IFs) discovered in 'healthy volunteers' are managed. Routine reporting of 'healthy volunteers' by a radiologist is a challenging logistic and financial burden and in a survey of UK institutions performing research imaging, just 14% of institutions had this as policy. It would be valuable to devise automated strategies to ensure that IFs could be reliably and accurately identified which potentially would remove 90% of scans requiring routine radiological review, thereby increasing the feasibility of implementing a routine reporting strategy.

Deep learning is a new technique in computer science that automatically learns hierarchies of relevant features directly from the raw inputs (e.g. MRI) using multilayered neural networks. Previous studies have demonstrated the potential of deep learning methods in the basic interpretation of neuroradiological studies including MRI scans.

A deep learning algorithm will be trained using both retrospective and prospective head MRI studies to recognise studies with abnormalities. Retrospective and prospective data will be obtained from secondary and tertiary NHS centres across the UK. A deep learning algorithm will be trained that classifies a subset of these studies as normal or abnormal. The technique will then be tested on an independent subset to determine its validity.

Retrospective MRI scans will be collected from NHS databases by the direct care team and then anonymised before being transferred to the research team for training and testing. Prospective MRI scans will be obtained by recruiting patients who are undergoing head MRI scans as part of their routine medical care and then anonymised before being transferred to the research team.


Condition or disease Intervention/treatment
Neurological Disorder Diagnostic Test: Deep Learning Algorithm

Layout table for study information
Study Type : Observational
Estimated Enrollment : 30000 participants
Observational Model: Other
Time Perspective: Other
Official Title: Deep Learning for Identification of Incidental Findings on Cranial MRI
Actual Study Start Date : April 1, 2019
Estimated Primary Completion Date : October 20, 2023
Estimated Study Completion Date : November 20, 2023

Intervention Details:
  • Diagnostic Test: Deep Learning Algorithm
    The study involves the development and testing of a diagnostic test. We will develop the computer algorithm using a dataset of retrospective and prospective head MRI scans, to train convolutional networks using software developed at the Department of Biomedical Engineering, King's College London. The reference standard will be consultant radiologist reports of the MRI studies.


Primary Outcome Measures :
  1. Sensitivity and specificity of a deep learning computer algorithm (termed a convolutional neural network) to recognise abnormalities on cranial MRI scans. [ Time Frame: At end of study (5-year study) ]
    Sensitivity, specificity, positive predictive value, and negative predictive values.


Secondary Outcome Measures :
  1. Sensitivity and specificity of a deep learning computer algorithm to broadly categorise abnormalities on cranial MRI scans. [ Time Frame: At end of study (5-year study) ]
    Sensitivity, specificity, positive predictive value, and negative predictive values.



Information from the National Library of Medicine

Choosing to participate in a study is an important personal decision. Talk with your doctor and family members or friends about deciding to join a study. To learn more about this study, you or your doctor may contact the study research staff using the contacts provided below. For general information, Learn About Clinical Studies.


Layout table for eligibility information
Ages Eligible for Study:   18 Years and older   (Adult, Older Adult)
Sexes Eligible for Study:   All
Accepts Healthy Volunteers:   No
Sampling Method:   Non-Probability Sample
Study Population
All adult cranial MRI scan patients presenting at the radiology departments of secondary and tertiary NHS centres across the UK for any indication.
Criteria

Inclusion Criteria:

  • All cranial MRI scans with compatible sequences
  • > 18 years old

Exclusion Criteria:

  • No corresponding radiologist report for the retrospective analyses
  • No consent for future use of the research images held within the historic database stored at The Centre for Neuroimaging Sciences (Kings College London).
  • Poor image quality

Information from the National Library of Medicine

To learn more about this study, you or your doctor may contact the study research staff using the contact information provided by the sponsor.

Please refer to this study by its ClinicalTrials.gov identifier (NCT number): NCT04368481


Contacts
Layout table for location contacts
Contact: Alima Rahman +44(0)20 7848 9670 alima.rahman@kcl.ac.uk
Contact: Thomas Booth tombooth@doctors.org.uk

Locations
Layout table for location information
United Kingdom
Guy's Hospital Recruiting
London, United Kingdom
King's College Hospital Recruiting
London, United Kingdom
Contact: Research Team       research.kch@gmail.com   
St George's Hospital Recruiting
London, United Kingdom
St Thomas' Hospital Recruiting
London, United Kingdom
Sponsors and Collaborators
King's College Hospital NHS Trust
Investigators
Layout table for investigator information
Principal Investigator: Thomas Booth King's College Hospital NHS Trust
Layout table for additonal information
Responsible Party: King's College Hospital NHS Trust
ClinicalTrials.gov Identifier: NCT04368481    
Other Study ID Numbers: KCH18-197
First Posted: April 29, 2020    Key Record Dates
Last Update Posted: April 29, 2020
Last Verified: February 2020

Layout table for additional information
Studies a U.S. FDA-regulated Drug Product: No
Studies a U.S. FDA-regulated Device Product: No
Additional relevant MeSH terms:
Layout table for MeSH terms
Nervous System Diseases