Reducing radiation dose for multi-phase contrast-enhanced dual energy renal CT: pilot study evaluating prior iterative reconstruction

Patrick J. Navin, Bohyun Kim, Michael Wells, Ashish Khandelwal, Ahmed F. Halaweish, Taylor R. Moen, Matthew P. Johnson, Shannon McCollough, Yong Suk Lee, Shuai Leng, Cynthia H McCollough, Joel Garland Fletcher

Research output: Contribution to journalArticle

Abstract

Purpose: Prior iterative reconstruction (PIR) uses spatial information from one phase of enhancement to reduce image noise in other phases. We sought to determine if PIR could reduce radiation dose while preserving observer performance and CT number at multi-phase dual energy (DE) renal CT. Methods: CT projection data from multi-phase DE renal CT examinations were collected. Images corresponding to 40% radiation dose were reconstructed using validated noise insertion and PIR. Three genitourinary radiologists examined routine and 40% dose PIR images. Probability of malignancy was assessed [from 0 to 100] with malignancy assumed at probability ≥ 75. Observer performance was compared on a per patient and per lesion level. CT number accuracy was measured. Results: Twenty-three patients had 49 renal lesions (11 solid renal neoplasms). CT number was nearly identical between techniques (mean CT number difference: unenhanced 2 ± 2 HU; enhanced 4 ± 4 HU). AUC for malignancy was similar between multi-phase routine dose DE and lower dose PIR images [per patient: 0.950 vs. 0.916 (p = 0.356); per lesion: 0.931 vs. 0.884 (p = 0.304)]. Per patient sensitivity was also similar (78% routine dose vs. 82% lower dose [p ≥ 0.99]), as was specificity (91% routine dose vs. 93% lower dose PIR [p > 0.99]), with similar findings on a per lesion level. Subjective image quality was also similar (p = 0.34). Conclusions: Prior iterative reconstruction is a new reconstruction method for multi-phase CT examinations that promises to facilitate radiation dose reduction by over 50% for multi-phase DE renal CT exams without compromising CT number or observer performance.

Original languageEnglish (US)
JournalAbdominal Radiology
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Radiation
Kidney
Computer-Assisted Image Processing
Neoplasms
Kidney Neoplasms
Area Under Curve
Noise
Radiologists

Keywords

  • Computed tomography, x-ray
  • Iterative reconstruction
  • Radiation dosage
  • Renal neoplasms

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Gastroenterology
  • Urology

Cite this

Reducing radiation dose for multi-phase contrast-enhanced dual energy renal CT : pilot study evaluating prior iterative reconstruction. / Navin, Patrick J.; Kim, Bohyun; Wells, Michael; Khandelwal, Ashish; Halaweish, Ahmed F.; Moen, Taylor R.; Johnson, Matthew P.; McCollough, Shannon; Lee, Yong Suk; Leng, Shuai; McCollough, Cynthia H; Fletcher, Joel Garland.

In: Abdominal Radiology, 01.01.2019.

Research output: Contribution to journalArticle

@article{c3a6d3a8fe164b11a8ac306b7d1f778b,
title = "Reducing radiation dose for multi-phase contrast-enhanced dual energy renal CT: pilot study evaluating prior iterative reconstruction",
abstract = "Purpose: Prior iterative reconstruction (PIR) uses spatial information from one phase of enhancement to reduce image noise in other phases. We sought to determine if PIR could reduce radiation dose while preserving observer performance and CT number at multi-phase dual energy (DE) renal CT. Methods: CT projection data from multi-phase DE renal CT examinations were collected. Images corresponding to 40{\%} radiation dose were reconstructed using validated noise insertion and PIR. Three genitourinary radiologists examined routine and 40{\%} dose PIR images. Probability of malignancy was assessed [from 0 to 100] with malignancy assumed at probability ≥ 75. Observer performance was compared on a per patient and per lesion level. CT number accuracy was measured. Results: Twenty-three patients had 49 renal lesions (11 solid renal neoplasms). CT number was nearly identical between techniques (mean CT number difference: unenhanced 2 ± 2 HU; enhanced 4 ± 4 HU). AUC for malignancy was similar between multi-phase routine dose DE and lower dose PIR images [per patient: 0.950 vs. 0.916 (p = 0.356); per lesion: 0.931 vs. 0.884 (p = 0.304)]. Per patient sensitivity was also similar (78{\%} routine dose vs. 82{\%} lower dose [p ≥ 0.99]), as was specificity (91{\%} routine dose vs. 93{\%} lower dose PIR [p > 0.99]), with similar findings on a per lesion level. Subjective image quality was also similar (p = 0.34). Conclusions: Prior iterative reconstruction is a new reconstruction method for multi-phase CT examinations that promises to facilitate radiation dose reduction by over 50{\%} for multi-phase DE renal CT exams without compromising CT number or observer performance.",
keywords = "Computed tomography, x-ray, Iterative reconstruction, Radiation dosage, Renal neoplasms",
author = "Navin, {Patrick J.} and Bohyun Kim and Michael Wells and Ashish Khandelwal and Halaweish, {Ahmed F.} and Moen, {Taylor R.} and Johnson, {Matthew P.} and Shannon McCollough and Lee, {Yong Suk} and Shuai Leng and McCollough, {Cynthia H} and Fletcher, {Joel Garland}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s00261-019-02150-9",
language = "English (US)",
journal = "Abdominal Radiology",
issn = "2366-004X",
publisher = "Springer New York",

}

TY - JOUR

T1 - Reducing radiation dose for multi-phase contrast-enhanced dual energy renal CT

T2 - pilot study evaluating prior iterative reconstruction

AU - Navin, Patrick J.

AU - Kim, Bohyun

AU - Wells, Michael

AU - Khandelwal, Ashish

AU - Halaweish, Ahmed F.

AU - Moen, Taylor R.

AU - Johnson, Matthew P.

AU - McCollough, Shannon

AU - Lee, Yong Suk

AU - Leng, Shuai

AU - McCollough, Cynthia H

AU - Fletcher, Joel Garland

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Purpose: Prior iterative reconstruction (PIR) uses spatial information from one phase of enhancement to reduce image noise in other phases. We sought to determine if PIR could reduce radiation dose while preserving observer performance and CT number at multi-phase dual energy (DE) renal CT. Methods: CT projection data from multi-phase DE renal CT examinations were collected. Images corresponding to 40% radiation dose were reconstructed using validated noise insertion and PIR. Three genitourinary radiologists examined routine and 40% dose PIR images. Probability of malignancy was assessed [from 0 to 100] with malignancy assumed at probability ≥ 75. Observer performance was compared on a per patient and per lesion level. CT number accuracy was measured. Results: Twenty-three patients had 49 renal lesions (11 solid renal neoplasms). CT number was nearly identical between techniques (mean CT number difference: unenhanced 2 ± 2 HU; enhanced 4 ± 4 HU). AUC for malignancy was similar between multi-phase routine dose DE and lower dose PIR images [per patient: 0.950 vs. 0.916 (p = 0.356); per lesion: 0.931 vs. 0.884 (p = 0.304)]. Per patient sensitivity was also similar (78% routine dose vs. 82% lower dose [p ≥ 0.99]), as was specificity (91% routine dose vs. 93% lower dose PIR [p > 0.99]), with similar findings on a per lesion level. Subjective image quality was also similar (p = 0.34). Conclusions: Prior iterative reconstruction is a new reconstruction method for multi-phase CT examinations that promises to facilitate radiation dose reduction by over 50% for multi-phase DE renal CT exams without compromising CT number or observer performance.

AB - Purpose: Prior iterative reconstruction (PIR) uses spatial information from one phase of enhancement to reduce image noise in other phases. We sought to determine if PIR could reduce radiation dose while preserving observer performance and CT number at multi-phase dual energy (DE) renal CT. Methods: CT projection data from multi-phase DE renal CT examinations were collected. Images corresponding to 40% radiation dose were reconstructed using validated noise insertion and PIR. Three genitourinary radiologists examined routine and 40% dose PIR images. Probability of malignancy was assessed [from 0 to 100] with malignancy assumed at probability ≥ 75. Observer performance was compared on a per patient and per lesion level. CT number accuracy was measured. Results: Twenty-three patients had 49 renal lesions (11 solid renal neoplasms). CT number was nearly identical between techniques (mean CT number difference: unenhanced 2 ± 2 HU; enhanced 4 ± 4 HU). AUC for malignancy was similar between multi-phase routine dose DE and lower dose PIR images [per patient: 0.950 vs. 0.916 (p = 0.356); per lesion: 0.931 vs. 0.884 (p = 0.304)]. Per patient sensitivity was also similar (78% routine dose vs. 82% lower dose [p ≥ 0.99]), as was specificity (91% routine dose vs. 93% lower dose PIR [p > 0.99]), with similar findings on a per lesion level. Subjective image quality was also similar (p = 0.34). Conclusions: Prior iterative reconstruction is a new reconstruction method for multi-phase CT examinations that promises to facilitate radiation dose reduction by over 50% for multi-phase DE renal CT exams without compromising CT number or observer performance.

KW - Computed tomography, x-ray

KW - Iterative reconstruction

KW - Radiation dosage

KW - Renal neoplasms

UR - http://www.scopus.com/inward/record.url?scp=85071096274&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071096274&partnerID=8YFLogxK

U2 - 10.1007/s00261-019-02150-9

DO - 10.1007/s00261-019-02150-9

M3 - Article

C2 - 31422439

AN - SCOPUS:85071096274

JO - Abdominal Radiology

JF - Abdominal Radiology

SN - 2366-004X

ER -