1
|
Delbarre DJ, Santos L, Ganjgahi H, Horner N, McCoy A, Westerberg H, Häring DA, Nichols TE, Mallon AM. Application of a convolutional neural network to the quality control of MRI defacing. Comput Biol Med 2022; 151:106211. [PMID: 36327884 DOI: 10.1016/j.compbiomed.2022.106211] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/07/2022] [Revised: 08/26/2022] [Accepted: 10/09/2022] [Indexed: 12/27/2022]
Abstract
Large-scale neuroimaging datasets present unique challenges for automated processing pipelines. Motivated by a large clinical trials dataset with over 235,000 MRI scans, we consider the challenge of defacing - anonymisation to remove identifying facial features. The defacing process must undergo quality control (QC) checks to ensure that the facial features have been removed and that the brain tissue is left intact. Visual QC checks are time-consuming and can cause delays in preparing data. We have developed a convolutional neural network (CNN) that can assist with the QC of the application of MRI defacing; our CNN is able to distinguish between scans that are correctly defaced and can classify defacing failures into three sub-types to facilitate parameter tuning during remedial re-defacing. Since integrating the CNN into our anonymisation pipeline, over 75,000 scans have been processed. Strict thresholds have been applied so that ambiguous classifications are referred for visual QC checks, however all scans still undergo an efficient verification check before being marked as passed. After applying the thresholds, our network is 92% accurate and can classify nearly half of the scans without the need for protracted manual checks. Our model can generalise across MRI modalities and has comparable performance when tested on an independent dataset. Even with the introduction of the verification checks, incorporation of the CNN has reduced the time spent undertaking QC checks by 42% during initial defacing, and by 35% overall. With the help of the CNN, we have been able to successfully deface 96% of the scans in the project whilst maintaining high QC standards. In a similarly sized new project, we would expect the model to reduce the time spent on manual QC checks by 125 h. Our approach is applicable to other projects with the potential to greatly improve the efficiency of imaging anonymisation pipelines.
Collapse
Affiliation(s)
- Daniel J Delbarre
- MRC Harwell Institute, Harwell Campus, Oxfordshire, OX11 0RD, United Kingdom; The Alan Turing Institute, British Library, 96 Euston Road, London, NW1 2DB, United Kingdom.
| | - Luis Santos
- MRC Harwell Institute, Harwell Campus, Oxfordshire, OX11 0RD, United Kingdom; The Alan Turing Institute, British Library, 96 Euston Road, London, NW1 2DB, United Kingdom
| | - Habib Ganjgahi
- Department of Statistics, University of Oxford, 24-29 St Giles', Oxford, OX1 3LB, United Kingdom
| | - Neil Horner
- MRC Harwell Institute, Harwell Campus, Oxfordshire, OX11 0RD, United Kingdom
| | - Aaron McCoy
- MRC Harwell Institute, Harwell Campus, Oxfordshire, OX11 0RD, United Kingdom
| | - Henrik Westerberg
- MRC Harwell Institute, Harwell Campus, Oxfordshire, OX11 0RD, United Kingdom
| | | | - Thomas E Nichols
- Big Data Institute, Li Ka Shing Centre for Health Information and Discovery, Nuffield Department of Population Health, University of Oxford, Old Road Campus, Oxford, OX3 7LF, United Kingdom; Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, OX3 9DU, United Kingdom
| | - Ann-Marie Mallon
- MRC Harwell Institute, Harwell Campus, Oxfordshire, OX11 0RD, United Kingdom; The Alan Turing Institute, British Library, 96 Euston Road, London, NW1 2DB, United Kingdom
| |
Collapse
|
2
|
Wotton JM, Peterson E, Flenniken AM, Bains RS, Veeraragavan S, Bower LR, Bubier JA, Parisien M, Bezginov A, Haselimashhadi H, Mason J, Moore MA, Stewart ME, Clary DA, Delbarre DJ, Anderson LC, D'Souza A, Goodwin LO, Harrison ME, Huang Z, Mckay M, Qu D, Santos L, Srinivasan S, Urban R, Vukobradovic I, Ward CS, Willett AM, Braun RE, Brown SD, Dickinson ME, Heaney JD, Kumar V, Lloyd KK, Mallon AM, McKerlie C, Murray SA, Nutter LM, Parkinson H, Seavitt JR, Wells S, Samaco RC, Chesler EJ, Smedley D, Diatchenko L, Baumbauer KM, Young EE, Bonin RP, Mandillo S, White JK. Identifying genetic determinants of inflammatory pain in mice using a large-scale gene-targeted screen. Pain 2022; 163:1139-1157. [PMID: 35552317 PMCID: PMC9100450 DOI: 10.1097/j.pain.0000000000002481] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/16/2021] [Revised: 08/17/2021] [Accepted: 09/07/2021] [Indexed: 02/03/2023]
Abstract
ABSTRACT Identifying the genetic determinants of pain is a scientific imperative given the magnitude of the global health burden that pain causes. Here, we report a genetic screen for nociception, performed under the auspices of the International Mouse Phenotyping Consortium. A biased set of 110 single-gene knockout mouse strains was screened for 1 or more nociception and hypersensitivity assays, including chemical nociception (formalin) and mechanical and thermal nociception (von Frey filaments and Hargreaves tests, respectively), with or without an inflammatory agent (complete Freund's adjuvant). We identified 13 single-gene knockout strains with altered nocifensive behavior in 1 or more assays. All these novel mouse models are openly available to the scientific community to study gene function. Two of the 13 genes (Gria1 and Htr3a) have been previously reported with nociception-related phenotypes in genetically engineered mouse strains and represent useful benchmarking standards. One of the 13 genes (Cnrip1) is known from human studies to play a role in pain modulation and the knockout mouse reported herein can be used to explore this function further. The remaining 10 genes (Abhd13, Alg6, BC048562, Cgnl1, Cp, Mmp16, Oxa1l, Tecpr2, Trim14, and Trim2) reveal novel pathways involved in nociception and may provide new knowledge to better understand genetic mechanisms of inflammatory pain and to serve as models for therapeutic target validation and drug development.
Collapse
Affiliation(s)
| | - Emma Peterson
- The Jackson Laboratory, Bar Harbor, ME, United States
| | - Ann M. Flenniken
- The Centre for Phenogenomics, Toronto, ON, Canada
- Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, ON, Canada
| | - Rasneer S. Bains
- The Mary Lyon Centre, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Surabi Veeraragavan
- Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, United States
- Jan and Dan Duncan Neurological Research Institute, Texas Children's Hospital, Houston, TX, United States
| | - Lynette R. Bower
- Mouse Biology Program, University of California-Davis, Davis, CA, United States
| | | | - Marc Parisien
- Department of Anesthesia, Faculty of Medicine, Faculty of Dentistry, McGill University, Genome Building, Montreal, QC, Canada
| | - Alexandr Bezginov
- The Centre for Phenogenomics, Toronto, ON, Canada
- The Hospital for Sick Children, Toronto, ON, Canada
| | - Hamed Haselimashhadi
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, Cambridgeshire, United Kingdom
| | - Jeremy Mason
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, Cambridgeshire, United Kingdom
| | | | - Michelle E. Stewart
- The Mary Lyon Centre, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Dave A. Clary
- Mouse Biology Program, University of California-Davis, Davis, CA, United States
| | - Daniel J. Delbarre
- Mammalian Genetics Unit, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | | | - Abigail D'Souza
- The Centre for Phenogenomics, Toronto, ON, Canada
- Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, ON, Canada
| | | | - Mark E. Harrison
- The Mary Lyon Centre, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Ziyue Huang
- The Centre for Phenogenomics, Toronto, ON, Canada
- Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, ON, Canada
| | - Matthew Mckay
- The Jackson Laboratory, Bar Harbor, ME, United States
| | - Dawei Qu
- The Centre for Phenogenomics, Toronto, ON, Canada
- Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, ON, Canada
| | - Luis Santos
- Mammalian Genetics Unit, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Subhiksha Srinivasan
- Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, United States
| | - Rachel Urban
- The Jackson Laboratory, Bar Harbor, ME, United States
| | - Igor Vukobradovic
- The Centre for Phenogenomics, Toronto, ON, Canada
- Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, ON, Canada
| | - Christopher S. Ward
- Department of Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, TX, United States
| | | | | | - Steve D.M. Brown
- Mammalian Genetics Unit, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Mary E. Dickinson
- Department of Molecular Physiology and Biophysics, Baylor College of Medicine, Houston, TX, United States
| | - Jason D. Heaney
- Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, United States
| | - Vivek Kumar
- The Jackson Laboratory, Bar Harbor, ME, United States
| | - K.C. Kent Lloyd
- Mouse Biology Program, University of California-Davis, Davis, CA, United States
- Department of Surgery, School of Medicine, University of California-Davis, Davis, CA, United States
| | - Ann-Marie Mallon
- Mammalian Genetics Unit, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Colin McKerlie
- Lunenfeld-Tanenbaum Research Institute, Sinai Health, Toronto, ON, Canada
- The Hospital for Sick Children, Toronto, ON, Canada
| | | | - Lauryl M.J. Nutter
- The Centre for Phenogenomics, Toronto, ON, Canada
- The Hospital for Sick Children, Toronto, ON, Canada
| | - Helen Parkinson
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, Cambridgeshire, United Kingdom
| | - John R. Seavitt
- Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, United States
| | - Sara Wells
- The Mary Lyon Centre, MRC Harwell Institute, Didcot, Oxfordshire, United Kingdom
| | - Rodney C. Samaco
- Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, United States
- Jan and Dan Duncan Neurological Research Institute, Texas Children's Hospital, Houston, TX, United States
| | | | - Damian Smedley
- William Harvey Research Institute, Charterhouse Square, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, United Kingdom
| | - Luda Diatchenko
- Department of Anesthesia, Faculty of Medicine, Faculty of Dentistry, McGill University, Genome Building, Montreal, QC, Canada
| | | | - Erin E. Young
- Anesthesiology, University of Kansas School of Medicine, KU Medical Center, Kansas City, KS, United States
| | - Robert P. Bonin
- Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON, Canada
| | - Silvia Mandillo
- Institute of Biochemistry and Cell Biology-National Research Council, IBBC-CNR, Monterotondo (RM), Italy
| | | |
Collapse
|
3
|
Cooper RL, Thiery AP, Fletcher AG, Delbarre DJ, Rasch LJ, Fraser GJ. An ancient Turing-like patterning mechanism regulates skin denticle development in sharks. Sci Adv 2018; 4:eaau5484. [PMID: 30417097 PMCID: PMC6221541 DOI: 10.1126/sciadv.aau5484] [Citation(s) in RCA: 37] [Impact Index Per Article: 6.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/22/2018] [Accepted: 10/12/2018] [Indexed: 05/02/2023]
Abstract
Vertebrates have a vast array of epithelial appendages, including scales, feathers, and hair. The developmental patterning of these diverse structures can be theoretically explained by Alan Turing's reaction-diffusion system. However, the role of this system in epithelial appendage patterning of early diverging lineages (compared to tetrapods), such as the cartilaginous fishes, is poorly understood. We investigate patterning of the unique tooth-like skin denticles of sharks, which closely relates to their hydrodynamic and protective functions. We demonstrate through simulation models that a Turing-like mechanism can explain shark denticle patterning and verify this system using gene expression analysis and gene pathway inhibition experiments. This mechanism bears remarkable similarity to avian feather patterning, suggesting deep homology of the system. We propose that a diverse range of vertebrate appendages, from shark denticles to avian feathers and mammalian hair, use this ancient and conserved system, with slight genetic modulation accounting for broad variations in patterning.
Collapse
Affiliation(s)
- Rory L. Cooper
- Department of Animal and Plant Sciences, University of Sheffield, Sheffield, UK
| | - Alexandre P. Thiery
- Department of Animal and Plant Sciences, University of Sheffield, Sheffield, UK
| | | | | | - Liam J. Rasch
- Department of Animal and Plant Sciences, University of Sheffield, Sheffield, UK
- Human Developmental Biology Resource, Institute of Child Health, University College, London, UK
| | - Gareth J. Fraser
- Department of Animal and Plant Sciences, University of Sheffield, Sheffield, UK
- Department of Biology, University of Florida, Gainesville, FL, USA
| |
Collapse
|