Classification of paediatric brain tumours by diffusion weighted imaging and machine learning

Jan Novak; Niloufar Zarinabad; Heather Rose; Theodoros Arvanitis; Lesley MacPherson; Benjamin Pinkey; Adam Oates; Patrick Hales; Richard Grundy; Dorothee Auer; Daniel Rodriguez Gutierrez; Tim Jaspan; Shivaram Avula; Laurence Abernethy; Ramneek Kaur; Darren Hargrave; Dipayan Mitra; Simon Bailey; Nigel Davies; Christopher Clark; Andrew Peet

doi:10.1038/s41598-021-82214-3

Classification of paediatric brain tumours by diffusion weighted imaging and machine learning

Jan Novak
, Niloufar Zarinabad
, Heather Rose
, Theodoros Arvanitis
, Lesley MacPherson
, Benjamin Pinkey
, Adam Oates
, Patrick Hales
, Richard Grundy
, Dorothee Auer
, Daniel Rodriguez Gutierrez
, Tim Jaspan
, Shivaram Avula
, Laurence Abernethy
, Ramneek Kaur
, Darren Hargrave
, Dipayan Mitra
, Simon Bailey
, Nigel Davies
, Christopher Clark

Andrew Peet^*

^*Corresponding author for this work

grid.6572.6 0000 0004 1936 7486 Institute of Cancer and Genomic Sciences, School of Medical and Dental Sciences University of Birmingham Birmingham UK; grid.498025.2 Oncology, Birmingham Women’s and Children’s NHS Foundation Trust Birmingham UK
grid.6572.6 0000 0004 1936 7486 Institute of Cancer and Genomic Sciences, School of Medical and Dental Sciences University of Birmingham Birmingham UK; grid.498025.2 Oncology, Birmingham Women’s and Children’s NHS Foundation Trust Birmingham UK; grid.7372.1 0000 0000 8809 1613 Institute of Digital Healthcare, WMG University of Warwick Coventry UK
grid.498025.2 Radiology, Birmingham Women’s and Children’s NHS Foundation Trust Birmingham UK
grid.83440.3b 0000000121901201 Developmental Imaging & Biophysics Section UCL Great Ormond Street Institute of Child Health WC1N 1EH London UK
grid.4563.4 0000 0004 1936 8868 The Children’s Brain Tumour Research Centre University of Nottingham Nottingham UK
grid.4563.4 0000 0004 1936 8868 Sir Peter Mansfield Imaging Centre University of Nottingham Biomedical Research Centre Nottingham UK; NIHR Nottingham Biomedical Research Centre Nottingham UK
grid.4563.4 0000 0004 1936 8868 The Children’s Brain Tumour Research Centre University of Nottingham Nottingham UK; grid.415598.4 0000 0004 0641 4263 Medical Physics Nottingham University Hospital, Queen’s Medical Centre Nottingham UK
grid.4563.4 0000 0004 1936 8868 The Children’s Brain Tumour Research Centre University of Nottingham Nottingham UK; grid.415598.4 0000 0004 0641 4263 Neuroradiology, Nottingham University Hospital, Queen’s Medical Centre Nottingham UK
grid.413582.9 0000 0001 0503 2798 Department of Radiology Alder Hey Children’s Hospital NHS Foundation Trust Liverpool UK
Haematology and Oncology Department Great Ormond Street Children’s Hospital London UK
grid.420004.2 0000 0004 0444 2244 The Newcastle Upon Tyne Hospitals NHS Foundation Trust Newcastle UK
grid.419334.8 0000 0004 0641 3236 Sir James Spence Institute of Child Health Royal Victoria Infirmary Newcastle upon Tyne UK
grid.6572.6 0000 0004 1936 7486 Institute of Cancer and Genomic Sciences, School of Medical and Dental Sciences University of Birmingham Birmingham UK; grid.498025.2 Oncology, Birmingham Women’s and Children’s NHS Foundation Trust Birmingham UK; grid.412563.7 0000 0004 0376 6589 Radiation Protection Services University Hospitals Birmingham NHS Foundation Trust Birmingham UK

Research output: Contribution to journal › Article › peer-review

52 Citations (Scopus)

24 Downloads (Pure)

Abstract

To determine if apparent diffusion coefficients (ADC) can discriminate between posterior fossa brain tumours on a multicentre basis. A total of 124 paediatric patients with posterior fossa tumours (including 55 Medulloblastomas, 36 Pilocytic Astrocytomas and 26 Ependymomas) were scanned using diffusion weighted imaging across 12 different hospitals using a total of 18 different scanners. Apparent diffusion coefficient maps were produced and histogram data was extracted from tumour regions of interest. Total histograms and histogram metrics (mean, variance, skew, kurtosis and 10th, 20th and 50th quantiles) were used as data input for classifiers with accuracy determined by tenfold cross validation. Mean ADC values from the tumour regions of interest differed between tumour types, (ANOVA P < 0.001). A cut off value for mean ADC between Ependymomas and Medulloblastomas was found to be of 0.984 × 10 -3 mm 2 s -1 with sensitivity 80.8% and specificity 80.0%. Overall classification for the ADC histogram metrics were 85% using Naïve Bayes and 84% for Random Forest classifiers. The most commonly occurring posterior fossa paediatric brain tumours can be classified using Apparent Diffusion Coefficient histogram values to a high accuracy on a multicentre basis.

Original language	English
Article number	2987
Journal	Scientific Reports
Volume	11
Issue number	1
DOIs	https://doi.org/10.1038/s41598-021-82214-3
Publication status	Published - 4 Feb 2021

Bibliographical note

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Funding: We would like to acknowledge funding from the Cancer Research UK and EPSRC Cancer Imaging Programme at the Children’s Cancer and Leukaemia Group (CCLG) in association with the MRC and Department of Health (England) (C7809/A10342), the Cancer Research UK and NIHR Experimental Cancer Medicine Centre Paediatric Network (C8232/A25261), the Medical Research Council –Health Data Research UK Substantive Site, the Children’s Research Fund, Poppyfields and Help Harry Help Others charity. Professor Peet is funded through an NIHR Research Professorship, NIHR-RP-R2-12-019. Professor Theodoros N Arvanitis is partially supported by Health Data Research UK, which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and Wellcome Trust. We would also like to acknowledge the MR radiographers involved in the study.

Access to Document

10.1038/s41598-021-82214-3Licence: CC BY 3.0

Classification of paediatric brain tumours by diffusion weighted imaging and machine learning
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Final published version, 1.86 MBLicence: CC BY 3.0

Cite this

Novak, J., Zarinabad, N., Rose, H., Arvanitis, T., MacPherson, L., Pinkey, B., Oates, A., Hales, P., Grundy, R., Auer, D., Gutierrez, D. R., Jaspan, T., Avula, S., Abernethy, L., Kaur, R., Hargrave, D., Mitra, D., Bailey, S., Davies, N., ... Peet, A. (2021). Classification of paediatric brain tumours by diffusion weighted imaging and machine learning. Scientific Reports, 11(1), Article 2987. https://doi.org/10.1038/s41598-021-82214-3

@article{00a6dd14798844bc9d4c3212c7b294cb,

title = "Classification of paediatric brain tumours by diffusion weighted imaging and machine learning",

abstract = "To determine if apparent diffusion coefficients (ADC) can discriminate between posterior fossa brain tumours on a multicentre basis. A total of 124 paediatric patients with posterior fossa tumours (including 55 Medulloblastomas, 36 Pilocytic Astrocytomas and 26 Ependymomas) were scanned using diffusion weighted imaging across 12 different hospitals using a total of 18 different scanners. Apparent diffusion coefficient maps were produced and histogram data was extracted from tumour regions of interest. Total histograms and histogram metrics (mean, variance, skew, kurtosis and 10th, 20th and 50th quantiles) were used as data input for classifiers with accuracy determined by tenfold cross validation. Mean ADC values from the tumour regions of interest differed between tumour types, (ANOVA P < 0.001). A cut off value for mean ADC between Ependymomas and Medulloblastomas was found to be of 0.984 × 10 -3 mm 2 s -1 with sensitivity 80.8\% and specificity 80.0\%. Overall classification for the ADC histogram metrics were 85\% using Na{\"i}ve Bayes and 84\% for Random Forest classifiers. The most commonly occurring posterior fossa paediatric brain tumours can be classified using Apparent Diffusion Coefficient histogram values to a high accuracy on a multicentre basis. ",

author = "Jan Novak and Niloufar Zarinabad and Heather Rose and Theodoros Arvanitis and Lesley MacPherson and Benjamin Pinkey and Adam Oates and Patrick Hales and Richard Grundy and Dorothee Auer and Gutierrez, \{Daniel Rodriguez\} and Tim Jaspan and Shivaram Avula and Laurence Abernethy and Ramneek Kaur and Darren Hargrave and Dipayan Mitra and Simon Bailey and Nigel Davies and Christopher Clark and Andrew Peet",

note = "This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Funding: We would like to acknowledge funding from the Cancer Research UK and EPSRC Cancer Imaging Programme at the Children{\textquoteright}s Cancer and Leukaemia Group (CCLG) in association with the MRC and Department of Health (England) (C7809/A10342), the Cancer Research UK and NIHR Experimental Cancer Medicine Centre Paediatric Network (C8232/A25261), the Medical Research Council –Health Data Research UK Substantive Site, the Children{\textquoteright}s Research Fund, Poppyfields and Help Harry Help Others charity. Professor Peet is funded through an NIHR Research Professorship, NIHR-RP-R2-12-019. Professor Theodoros N Arvanitis is partially supported by Health Data Research UK, which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and Wellcome Trust. We would also like to acknowledge the MR radiographers involved in the study.",

year = "2021",

month = feb,

day = "4",

doi = "10.1038/s41598-021-82214-3",

language = "English",

volume = "11",

journal = "Scientific Reports",

issn = "2045-2322",

publisher = "Nature Research",

number = "1",

}

Novak, J, Zarinabad, N, Rose, H, Arvanitis, T, MacPherson, L, Pinkey, B, Oates, A, Hales, P, Grundy, R, Auer, D, Gutierrez, DR, Jaspan, T, Avula, S, Abernethy, L, Kaur, R, Hargrave, D, Mitra, D, Bailey, S, Davies, N, Clark, C & Peet, A 2021, 'Classification of paediatric brain tumours by diffusion weighted imaging and machine learning', Scientific Reports, vol. 11, no. 1, 2987. https://doi.org/10.1038/s41598-021-82214-3

TY - JOUR

T1 - Classification of paediatric brain tumours by diffusion weighted imaging and machine learning

AU - Novak, Jan

AU - Zarinabad, Niloufar

AU - Rose, Heather

AU - Arvanitis, Theodoros

AU - MacPherson, Lesley

AU - Pinkey, Benjamin

AU - Oates, Adam

AU - Hales, Patrick

AU - Grundy, Richard

AU - Auer, Dorothee

AU - Gutierrez, Daniel Rodriguez

AU - Jaspan, Tim

AU - Avula, Shivaram

AU - Abernethy, Laurence

AU - Kaur, Ramneek

AU - Hargrave, Darren

AU - Mitra, Dipayan

AU - Bailey, Simon

AU - Davies, Nigel

AU - Clark, Christopher

AU - Peet, Andrew

N1 - This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Funding: We would like to acknowledge funding from the Cancer Research UK and EPSRC Cancer Imaging Programme at the Children’s Cancer and Leukaemia Group (CCLG) in association with the MRC and Department of Health (England) (C7809/A10342), the Cancer Research UK and NIHR Experimental Cancer Medicine Centre Paediatric Network (C8232/A25261), the Medical Research Council –Health Data Research UK Substantive Site, the Children’s Research Fund, Poppyfields and Help Harry Help Others charity. Professor Peet is funded through an NIHR Research Professorship, NIHR-RP-R2-12-019. Professor Theodoros N Arvanitis is partially supported by Health Data Research UK, which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and Wellcome Trust. We would also like to acknowledge the MR radiographers involved in the study.

PY - 2021/2/4

Y1 - 2021/2/4

N2 - To determine if apparent diffusion coefficients (ADC) can discriminate between posterior fossa brain tumours on a multicentre basis. A total of 124 paediatric patients with posterior fossa tumours (including 55 Medulloblastomas, 36 Pilocytic Astrocytomas and 26 Ependymomas) were scanned using diffusion weighted imaging across 12 different hospitals using a total of 18 different scanners. Apparent diffusion coefficient maps were produced and histogram data was extracted from tumour regions of interest. Total histograms and histogram metrics (mean, variance, skew, kurtosis and 10th, 20th and 50th quantiles) were used as data input for classifiers with accuracy determined by tenfold cross validation. Mean ADC values from the tumour regions of interest differed between tumour types, (ANOVA P < 0.001). A cut off value for mean ADC between Ependymomas and Medulloblastomas was found to be of 0.984 × 10 -3 mm 2 s -1 with sensitivity 80.8% and specificity 80.0%. Overall classification for the ADC histogram metrics were 85% using Naïve Bayes and 84% for Random Forest classifiers. The most commonly occurring posterior fossa paediatric brain tumours can be classified using Apparent Diffusion Coefficient histogram values to a high accuracy on a multicentre basis.

AB - To determine if apparent diffusion coefficients (ADC) can discriminate between posterior fossa brain tumours on a multicentre basis. A total of 124 paediatric patients with posterior fossa tumours (including 55 Medulloblastomas, 36 Pilocytic Astrocytomas and 26 Ependymomas) were scanned using diffusion weighted imaging across 12 different hospitals using a total of 18 different scanners. Apparent diffusion coefficient maps were produced and histogram data was extracted from tumour regions of interest. Total histograms and histogram metrics (mean, variance, skew, kurtosis and 10th, 20th and 50th quantiles) were used as data input for classifiers with accuracy determined by tenfold cross validation. Mean ADC values from the tumour regions of interest differed between tumour types, (ANOVA P < 0.001). A cut off value for mean ADC between Ependymomas and Medulloblastomas was found to be of 0.984 × 10 -3 mm 2 s -1 with sensitivity 80.8% and specificity 80.0%. Overall classification for the ADC histogram metrics were 85% using Naïve Bayes and 84% for Random Forest classifiers. The most commonly occurring posterior fossa paediatric brain tumours can be classified using Apparent Diffusion Coefficient histogram values to a high accuracy on a multicentre basis.

UR - https://www.nature.com/articles/s41598-021-82214-3

UR - https://www.scopus.com/pages/publications/85100512745

U2 - 10.1038/s41598-021-82214-3

DO - 10.1038/s41598-021-82214-3

M3 - Article

C2 - 33542327

SN - 2045-2322

VL - 11

JO - Scientific Reports

JF - Scientific Reports

IS - 1

M1 - 2987

ER -

Classification of paediatric brain tumours by diffusion weighted imaging and machine learning

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this