A deep learning- and partial least square regression-based model observer for a low-contrast lesion detection task in CT

Hao Gong, Lifeng Yu, Shuai Leng, Samantha K. Dilger, Liqiang Ren, Wei Zhou, Joel Garland Fletcher, Cynthia H McCollough

Research output: Contribution to journalArticle

Abstract

Purpose: This work aims to develop a new framework of image quality assessment using deep learning-based model observer (DL-MO) and to validate it in a low-contrast lesion detection task that involves CT images with patient anatomical background. Methods: The DL-MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS-DA) model and an internal noise component. The CNN was previously trained to achieve the state-of-the-art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL-MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL-MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11 mm) and three contrast levels (15, 20, and 25 HU). The lesions were inserted into patient liver images using a projection-based method. A validated noise insertion tool was used to synthesize CT exams with 50% and 25% of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two-alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL-MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL-MO and HO. Results: A statistically significant positive correlation was observed between DL-MO and HO for the 2AFC low-contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95% confidence interval (0.950, 0.996)]. Bland–Altman agreement analysis did not indicate statistically significant differences. Conclusions: The proposed DL-MO is highly correlated with HO in a low-contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL-MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.

Original languageEnglish (US)
JournalMedical physics
DOIs
StatePublished - Jan 1 2019

Fingerprint

Least-Squares Analysis
Learning
Noise
Liver
Radiation
Discriminant Analysis
Regression Analysis
Databases
Confidence Intervals

Keywords

  • computed tomography
  • CT protocol optimization
  • deep learning
  • image quality assessment
  • model observer

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

@article{45ee8118caf142a588302a138cfd3bcd,
title = "A deep learning- and partial least square regression-based model observer for a low-contrast lesion detection task in CT",
abstract = "Purpose: This work aims to develop a new framework of image quality assessment using deep learning-based model observer (DL-MO) and to validate it in a low-contrast lesion detection task that involves CT images with patient anatomical background. Methods: The DL-MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS-DA) model and an internal noise component. The CNN was previously trained to achieve the state-of-the-art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL-MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL-MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11 mm) and three contrast levels (15, 20, and 25 HU). The lesions were inserted into patient liver images using a projection-based method. A validated noise insertion tool was used to synthesize CT exams with 50{\%} and 25{\%} of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two-alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL-MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL-MO and HO. Results: A statistically significant positive correlation was observed between DL-MO and HO for the 2AFC low-contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95{\%} confidence interval (0.950, 0.996)]. Bland–Altman agreement analysis did not indicate statistically significant differences. Conclusions: The proposed DL-MO is highly correlated with HO in a low-contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL-MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.",
keywords = "computed tomography, CT protocol optimization, deep learning, image quality assessment, model observer",
author = "Hao Gong and Lifeng Yu and Shuai Leng and Dilger, {Samantha K.} and Liqiang Ren and Wei Zhou and Fletcher, {Joel Garland} and McCollough, {Cynthia H}",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/mp.13500",
language = "English (US)",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",

}

TY - JOUR

T1 - A deep learning- and partial least square regression-based model observer for a low-contrast lesion detection task in CT

AU - Gong, Hao

AU - Yu, Lifeng

AU - Leng, Shuai

AU - Dilger, Samantha K.

AU - Ren, Liqiang

AU - Zhou, Wei

AU - Fletcher, Joel Garland

AU - McCollough, Cynthia H

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Purpose: This work aims to develop a new framework of image quality assessment using deep learning-based model observer (DL-MO) and to validate it in a low-contrast lesion detection task that involves CT images with patient anatomical background. Methods: The DL-MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS-DA) model and an internal noise component. The CNN was previously trained to achieve the state-of-the-art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL-MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL-MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11 mm) and three contrast levels (15, 20, and 25 HU). The lesions were inserted into patient liver images using a projection-based method. A validated noise insertion tool was used to synthesize CT exams with 50% and 25% of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two-alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL-MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL-MO and HO. Results: A statistically significant positive correlation was observed between DL-MO and HO for the 2AFC low-contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95% confidence interval (0.950, 0.996)]. Bland–Altman agreement analysis did not indicate statistically significant differences. Conclusions: The proposed DL-MO is highly correlated with HO in a low-contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL-MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.

AB - Purpose: This work aims to develop a new framework of image quality assessment using deep learning-based model observer (DL-MO) and to validate it in a low-contrast lesion detection task that involves CT images with patient anatomical background. Methods: The DL-MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS-DA) model and an internal noise component. The CNN was previously trained to achieve the state-of-the-art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL-MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL-MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11 mm) and three contrast levels (15, 20, and 25 HU). The lesions were inserted into patient liver images using a projection-based method. A validated noise insertion tool was used to synthesize CT exams with 50% and 25% of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two-alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL-MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL-MO and HO. Results: A statistically significant positive correlation was observed between DL-MO and HO for the 2AFC low-contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95% confidence interval (0.950, 0.996)]. Bland–Altman agreement analysis did not indicate statistically significant differences. Conclusions: The proposed DL-MO is highly correlated with HO in a low-contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL-MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.

KW - computed tomography

KW - CT protocol optimization

KW - deep learning

KW - image quality assessment

KW - model observer

UR - http://www.scopus.com/inward/record.url?scp=85063695886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063695886&partnerID=8YFLogxK

U2 - 10.1002/mp.13500

DO - 10.1002/mp.13500

M3 - Article

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

ER -