Comparison of deep learning and human observer performance for detection and characterization of simulated lesions

Ruben De Man, Jianan Gang, Xin Li, Ge Wang

Research output: Contribution to journalArticle

Abstract

Detection and characterization of abnormalities in clinical imaging are of utmost importance for patient diagnosis and treatment. We present a comparison of convolutional neural network (CNN) and human observer performance on a simulated lesion detection and characterization task. We apply both conventional performance metrics, including accuracy and nonconventional metrics such as lift charts to perform qualitative and quantitative comparisons of each type of observer. It is determined that the CNN generally outperforms the human observers, particularly at high noise levels. However, high noise correlation reduces the relative performance of the CNN, and human observer performance is comparable to CNN under these conditions. These findings extend into the field of diagnostic radiology, where the adoption of deep learning is starting to become widespread. Consideration of the applications for which deep learning is most effective is of critical importance to this development.

Original languageEnglish (US)
Article number025503
JournalJournal of Medical Imaging
Volume6
Issue number2
DOIs
StatePublished - Apr 1 2019

Fingerprint

Learning
Noise
Radiology
Therapeutics

Keywords

  • artificial intelligence
  • detection
  • image analysis
  • image quality
  • noise

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Comparison of deep learning and human observer performance for detection and characterization of simulated lesions. / De Man, Ruben; Gang, Jianan; Li, Xin; Wang, Ge.

In: Journal of Medical Imaging, Vol. 6, No. 2, 025503, 01.04.2019.

Research output: Contribution to journalArticle

@article{b0a042e95a274439957fdfab4e644dc7,
title = "Comparison of deep learning and human observer performance for detection and characterization of simulated lesions",
abstract = "Detection and characterization of abnormalities in clinical imaging are of utmost importance for patient diagnosis and treatment. We present a comparison of convolutional neural network (CNN) and human observer performance on a simulated lesion detection and characterization task. We apply both conventional performance metrics, including accuracy and nonconventional metrics such as lift charts to perform qualitative and quantitative comparisons of each type of observer. It is determined that the CNN generally outperforms the human observers, particularly at high noise levels. However, high noise correlation reduces the relative performance of the CNN, and human observer performance is comparable to CNN under these conditions. These findings extend into the field of diagnostic radiology, where the adoption of deep learning is starting to become widespread. Consideration of the applications for which deep learning is most effective is of critical importance to this development.",
keywords = "artificial intelligence, detection, image analysis, image quality, noise",
author = "{De Man}, Ruben and Jianan Gang and Xin Li and Ge Wang",
year = "2019",
month = "4",
day = "1",
doi = "10.1117/1.JMI.6.2.025503",
language = "English (US)",
volume = "6",
journal = "Journal of Medical Imaging",
issn = "0720-048X",
publisher = "Elsevier Ireland Ltd",
number = "2",

}

TY - JOUR

T1 - Comparison of deep learning and human observer performance for detection and characterization of simulated lesions

AU - De Man, Ruben

AU - Gang, Jianan

AU - Li, Xin

AU - Wang, Ge

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Detection and characterization of abnormalities in clinical imaging are of utmost importance for patient diagnosis and treatment. We present a comparison of convolutional neural network (CNN) and human observer performance on a simulated lesion detection and characterization task. We apply both conventional performance metrics, including accuracy and nonconventional metrics such as lift charts to perform qualitative and quantitative comparisons of each type of observer. It is determined that the CNN generally outperforms the human observers, particularly at high noise levels. However, high noise correlation reduces the relative performance of the CNN, and human observer performance is comparable to CNN under these conditions. These findings extend into the field of diagnostic radiology, where the adoption of deep learning is starting to become widespread. Consideration of the applications for which deep learning is most effective is of critical importance to this development.

AB - Detection and characterization of abnormalities in clinical imaging are of utmost importance for patient diagnosis and treatment. We present a comparison of convolutional neural network (CNN) and human observer performance on a simulated lesion detection and characterization task. We apply both conventional performance metrics, including accuracy and nonconventional metrics such as lift charts to perform qualitative and quantitative comparisons of each type of observer. It is determined that the CNN generally outperforms the human observers, particularly at high noise levels. However, high noise correlation reduces the relative performance of the CNN, and human observer performance is comparable to CNN under these conditions. These findings extend into the field of diagnostic radiology, where the adoption of deep learning is starting to become widespread. Consideration of the applications for which deep learning is most effective is of critical importance to this development.

KW - artificial intelligence

KW - detection

KW - image analysis

KW - image quality

KW - noise

UR - http://www.scopus.com/inward/record.url?scp=85069487920&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069487920&partnerID=8YFLogxK

U2 - 10.1117/1.JMI.6.2.025503

DO - 10.1117/1.JMI.6.2.025503

M3 - Article

C2 - 31263738

AN - SCOPUS:85069487920

VL - 6

JO - Journal of Medical Imaging

JF - Journal of Medical Imaging

SN - 0720-048X

IS - 2

M1 - 025503

ER -