King's College London

Research portal

Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists

Research output: Contribution to journalArticle

Standard

Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists. / Antonelli, Michela; Johnston, Edward W.; Dikaios, Nikolaos; Cheung, King K.; Sidhu, Harbir S.; Appayya, Mrishta B.; Giganti, Francesco; Simmons, Lucy A.M.; Freeman, Alex; Allen, Clare; Ahmed, Hashim U.; Atkinson, David; Ourselin, Sebastien; Punwani, Shonit.

In: European Radiology, Vol. 29, No. 9, 01.09.2019, p. 4754-4764.

Research output: Contribution to journalArticle

Harvard

Antonelli, M, Johnston, EW, Dikaios, N, Cheung, KK, Sidhu, HS, Appayya, MB, Giganti, F, Simmons, LAM, Freeman, A, Allen, C, Ahmed, HU, Atkinson, D, Ourselin, S & Punwani, S 2019, 'Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists', European Radiology, vol. 29, no. 9, pp. 4754-4764. https://doi.org/10.1007/s00330-019-06244-2

APA

Antonelli, M., Johnston, E. W., Dikaios, N., Cheung, K. K., Sidhu, H. S., Appayya, M. B., ... Punwani, S. (2019). Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists. European Radiology, 29(9), 4754-4764. https://doi.org/10.1007/s00330-019-06244-2

Vancouver

Antonelli M, Johnston EW, Dikaios N, Cheung KK, Sidhu HS, Appayya MB et al. Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists. European Radiology. 2019 Sep 1;29(9):4754-4764. https://doi.org/10.1007/s00330-019-06244-2

Author

Antonelli, Michela ; Johnston, Edward W. ; Dikaios, Nikolaos ; Cheung, King K. ; Sidhu, Harbir S. ; Appayya, Mrishta B. ; Giganti, Francesco ; Simmons, Lucy A.M. ; Freeman, Alex ; Allen, Clare ; Ahmed, Hashim U. ; Atkinson, David ; Ourselin, Sebastien ; Punwani, Shonit. / Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists. In: European Radiology. 2019 ; Vol. 29, No. 9. pp. 4754-4764.

Bibtex Download

@article{5b31566ac0d344c7a2b30325d2c0bd05,
title = "Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists",
abstract = "Objective: The purpose of this study was: To test whether machine learning classifiers for transition zone (TZ) and peripheral zone (PZ) can correctly classify prostate tumors into those with/without a Gleason 4 component, and to compare the performance of the best performing classifiers against the opinion of three board-certified radiologists. Methods: A retrospective analysis of prospectively acquired data was performed at a single center between 2012 and 2015. Inclusion criteria were (i) 3-T mp-MRI compliant with international guidelines, (ii) Likert ≥ 3/5 lesion, (iii) transperineal template ± targeted index lesion biopsy confirming cancer ≥ Gleason 3 + 3. Index lesions from 164 men were analyzed (119 PZ, 45 TZ). Quantitative MRI and clinical features were used and zone-specific machine learning classifiers were constructed. Models were validated using a fivefold cross-validation and a temporally separated patient cohort. Classifier performance was compared against the opinion of three board-certified radiologists. Results: The best PZ classifier trained with prostate-specific antigen density, apparent diffusion coefficient (ADC), and maximum enhancement (ME) on DCE-MRI obtained a ROC area under the curve (AUC) of 0.83 following fivefold cross-validation. Diagnostic sensitivity at 50{\%} threshold of specificity was higher for the best PZ model (0.93) when compared with the mean sensitivity of the three radiologists (0.72). The best TZ model used ADC and ME to obtain an AUC of 0.75 following fivefold cross-validation. This achieved higher diagnostic sensitivity at 50{\%} threshold of specificity (0.88) than the mean sensitivity of the three radiologists (0.82). Conclusions: Machine learning classifiers predict Gleason pattern 4 in prostate tumors better than radiologists. Key Points: • Predictive models developed from quantitative multiparametric magnetic resonance imaging regarding the characterization of prostate cancer grade should be zone-specific. • Classifiers trained differently for peripheral and transition zone can predict a Gleason 4 component with a higher performance than the subjective opinion of experienced radiologists. • Classifiers would be particularly useful in the context of active surveillance, whereby decisions regarding whether to biopsy are necessitated.",
keywords = "Diagnosis, computer-assisted, Gleason score, Machine learning, Magnetic resonance imaging, Prostate cancer",
author = "Michela Antonelli and Johnston, {Edward W.} and Nikolaos Dikaios and Cheung, {King K.} and Sidhu, {Harbir S.} and Appayya, {Mrishta B.} and Francesco Giganti and Simmons, {Lucy A.M.} and Alex Freeman and Clare Allen and Ahmed, {Hashim U.} and David Atkinson and Sebastien Ourselin and Shonit Punwani",
year = "2019",
month = "9",
day = "1",
doi = "10.1007/s00330-019-06244-2",
language = "English",
volume = "29",
pages = "4754--4764",
journal = "European Radiology",
issn = "0938-7994",
publisher = "Springer Verlag",
number = "9",

}

RIS (suitable for import to EndNote) Download

TY - JOUR

T1 - Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists

AU - Antonelli, Michela

AU - Johnston, Edward W.

AU - Dikaios, Nikolaos

AU - Cheung, King K.

AU - Sidhu, Harbir S.

AU - Appayya, Mrishta B.

AU - Giganti, Francesco

AU - Simmons, Lucy A.M.

AU - Freeman, Alex

AU - Allen, Clare

AU - Ahmed, Hashim U.

AU - Atkinson, David

AU - Ourselin, Sebastien

AU - Punwani, Shonit

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Objective: The purpose of this study was: To test whether machine learning classifiers for transition zone (TZ) and peripheral zone (PZ) can correctly classify prostate tumors into those with/without a Gleason 4 component, and to compare the performance of the best performing classifiers against the opinion of three board-certified radiologists. Methods: A retrospective analysis of prospectively acquired data was performed at a single center between 2012 and 2015. Inclusion criteria were (i) 3-T mp-MRI compliant with international guidelines, (ii) Likert ≥ 3/5 lesion, (iii) transperineal template ± targeted index lesion biopsy confirming cancer ≥ Gleason 3 + 3. Index lesions from 164 men were analyzed (119 PZ, 45 TZ). Quantitative MRI and clinical features were used and zone-specific machine learning classifiers were constructed. Models were validated using a fivefold cross-validation and a temporally separated patient cohort. Classifier performance was compared against the opinion of three board-certified radiologists. Results: The best PZ classifier trained with prostate-specific antigen density, apparent diffusion coefficient (ADC), and maximum enhancement (ME) on DCE-MRI obtained a ROC area under the curve (AUC) of 0.83 following fivefold cross-validation. Diagnostic sensitivity at 50% threshold of specificity was higher for the best PZ model (0.93) when compared with the mean sensitivity of the three radiologists (0.72). The best TZ model used ADC and ME to obtain an AUC of 0.75 following fivefold cross-validation. This achieved higher diagnostic sensitivity at 50% threshold of specificity (0.88) than the mean sensitivity of the three radiologists (0.82). Conclusions: Machine learning classifiers predict Gleason pattern 4 in prostate tumors better than radiologists. Key Points: • Predictive models developed from quantitative multiparametric magnetic resonance imaging regarding the characterization of prostate cancer grade should be zone-specific. • Classifiers trained differently for peripheral and transition zone can predict a Gleason 4 component with a higher performance than the subjective opinion of experienced radiologists. • Classifiers would be particularly useful in the context of active surveillance, whereby decisions regarding whether to biopsy are necessitated.

AB - Objective: The purpose of this study was: To test whether machine learning classifiers for transition zone (TZ) and peripheral zone (PZ) can correctly classify prostate tumors into those with/without a Gleason 4 component, and to compare the performance of the best performing classifiers against the opinion of three board-certified radiologists. Methods: A retrospective analysis of prospectively acquired data was performed at a single center between 2012 and 2015. Inclusion criteria were (i) 3-T mp-MRI compliant with international guidelines, (ii) Likert ≥ 3/5 lesion, (iii) transperineal template ± targeted index lesion biopsy confirming cancer ≥ Gleason 3 + 3. Index lesions from 164 men were analyzed (119 PZ, 45 TZ). Quantitative MRI and clinical features were used and zone-specific machine learning classifiers were constructed. Models were validated using a fivefold cross-validation and a temporally separated patient cohort. Classifier performance was compared against the opinion of three board-certified radiologists. Results: The best PZ classifier trained with prostate-specific antigen density, apparent diffusion coefficient (ADC), and maximum enhancement (ME) on DCE-MRI obtained a ROC area under the curve (AUC) of 0.83 following fivefold cross-validation. Diagnostic sensitivity at 50% threshold of specificity was higher for the best PZ model (0.93) when compared with the mean sensitivity of the three radiologists (0.72). The best TZ model used ADC and ME to obtain an AUC of 0.75 following fivefold cross-validation. This achieved higher diagnostic sensitivity at 50% threshold of specificity (0.88) than the mean sensitivity of the three radiologists (0.82). Conclusions: Machine learning classifiers predict Gleason pattern 4 in prostate tumors better than radiologists. Key Points: • Predictive models developed from quantitative multiparametric magnetic resonance imaging regarding the characterization of prostate cancer grade should be zone-specific. • Classifiers trained differently for peripheral and transition zone can predict a Gleason 4 component with a higher performance than the subjective opinion of experienced radiologists. • Classifiers would be particularly useful in the context of active surveillance, whereby decisions regarding whether to biopsy are necessitated.

KW - Diagnosis, computer-assisted

KW - Gleason score

KW - Machine learning

KW - Magnetic resonance imaging

KW - Prostate cancer

UR - http://www.scopus.com/inward/record.url?scp=85067281812&partnerID=8YFLogxK

U2 - 10.1007/s00330-019-06244-2

DO - 10.1007/s00330-019-06244-2

M3 - Article

AN - SCOPUS:85067281812

VL - 29

SP - 4754

EP - 4764

JO - European Radiology

JF - European Radiology

SN - 0938-7994

IS - 9

ER -

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454