Impact of Deep Learning-based Optimization Algorithm on Image Quality of Low-dose Coronary CT Angiography with Noise Reduction: A Prospective Study

Liu, Peijun; Wang, Man; Yu, Min; Wang, Yun; Liu, Zhuoheng; Li, Yumei; Jin, Zhengyu

doi:10.1016/j.acra.2019.11.010

Cited by 36 publications

(23 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although CCTA has obvious advantages over other methods (22,23), we still strive to reduce the radiation dose and contrast medium dose according to the ALARA principle. At present, the more widely used method is to use low tube voltage scanning to improve contrast enhancement in blood vessels combined with iterative reconstruction (IR) algorithms or deep learning-based image reconstruction algorithms (16,17) to reduce image noise, to ensure adequate contrast-to-noise for blood vessels to meet the diagnostic requirements. For the contrast injection in CCTA, the guidelines suggest 1-2 mL/kg for the contrast medium dose and an injection time of 10s for children (24).…”

Section: Discussionmentioning

confidence: 99%

Performance evaluation of using shorter contrast injection and 70 kVp with deep learning image reconstruction for reduced contrast medium dose and radiation dose in coronary CT angiography for children: a pilot study

Sun¹,

Li²,

Li³

et al. 2021

Quant Imaging Med Surg

View full text Add to dashboard Cite

Background: Iterative reconstruction algorithms are often used to reduce image noise in low-dose coronary computed tomography angiography (CCTA) but encounter limitations. The newly introduced deep learning image reconstruction (DLIR) algorithm may provide new opportunities. We assessed the image quality and diagnostic performance of DLIR in low radiation dose and contrast medium dose CCTA of pediatric patients with 70 kVp and a shortened injection protocol.Methods: This was a prospective study. A total of 27 consecutive arrhythmic pediatric patients were enrolled in the study group and underwent CCTA using a prospective ECG-triggered single-beat protocol: tube voltage 70 kVp, automatic tube current modulation for a noise index (NI) of 22, and contrast dose of 0.4-0.6 mL/kg. Images were reconstructed with DLIR. They were compared with 27 matched patients in the control group scanned with 80 kVp, a lower NI setting (NI =19), and a higher contrast dose (0.8-1.2 mL/kg).The images in the control group were reconstructed using the adaptive statistical iterative reconstruction (ASIR-V) algorithm. The image contrast, image quality, and diagnostic confidence were assessed by 2 experienced radiologists using a 5-point scale (1: nondiagnostic and 5: excellent). The CT value and standard deviation of the aorta and perivascular tissue were measured, and the contrast-to-noise ratio (CNR) for the aorta was calculated. The contrast medium and radiation doses were compared. Results: The study and control groups had similar image contrast scores (4.75±0.57 vs. 4.78±0.42), image quality scores (3.67±0.47 vs. 3.44±0.51), and diagnostic confidence (4.74±0.44 vs. 4.74±0.45) (all P>0.05). There was an adequate enhancement in the aorta (614.74±127.73 vs. 705.89±111.20 HU) and similar CNR (20.34±4.64 vs. 20.99±4.14) in both groups. The image noise of the study group was lower in the aorta (30.61±3.88 vs. 34.77±3.49) and similar in perivascular tissue (27.66±6.24 vs. 27.55±3.33) compared with the control group. The study group reduced the total contrast medium dose by 53% to 15.07±3.68 ml and radiation dose by 36% to 0.57±0.31 mSv. Conclusions:The DLIR algorithm in CCTA for children using 70 kVp tube voltage with a shortened contrast medium injection protocol maintains image quality and diagnostic confidence while significantly reducing contrast medium dose and radiation dose compared with the use of the conventional CCTA protocol.

show abstract

Section: Discussionmentioning

confidence: 99%

Performance evaluation of using shorter contrast injection and 70 kVp with deep learning image reconstruction for reduced contrast medium dose and radiation dose in coronary CT angiography for children: a pilot study

Sun¹,

Li²,

Li³

et al. 2021

Quant Imaging Med Surg

View full text Add to dashboard Cite

show abstract

“…Kim et al compared image quality and noise on a low dose chest CT between ASiR-V and DL-based imaged reconstruction [124]. While maintaining better image quality, DL-based image reconstruction images were found to be less noisy compared to iterative reconstruction with ASiR-V. Other applications of CT screening include, low dose coronary CT angiography, where Liu et al found that DL-based algorithm can decrease image noise and thus improve the image quality [125]. Such DL-based image reconstruction can help further reduce radiation dose and improve image quality for low-dose CT based lung cancer screening.…”

Section: Emerging Technologies and Dl-based Reconstructionmentioning

confidence: 99%

Artificial intelligence in image reconstruction: The change is here

Singh

Wang

et al. 2020

Physica Medica

View full text Add to dashboard Cite

Innovations in CT have been impressive among imaging and medical technologies in both the hardware and software domain. The range and speed of CT scanning improved from the introduction of multidetector-row CT scanners with wide-array detectors and faster gantry rotation speeds. To tackle concerns over rising radiation doses from its increasing use and to improve image quality, CT reconstruction techniques evolved from filtered back projection to commercial release of iterative reconstruction techniques, and recently, of deep learning (DL)based image reconstruction. These newer reconstruction techniques enable improved or retained image quality versus filtered back projection at lower radiation doses. DL can aid in image reconstruction with training data without total reliance on the physical model of the imaging process, unique artifacts of PCD-CT due to charge sharing, K-escape, fluorescence x-ray emission, and pulse pileups can be handled in the data-driven fashion. With sufficiently reconstructed images, a well-designed network can be trained to upgrade image quality over a practical/clinical threshold or define new/killer applications. Besides, the much smaller detector pixel for PCD-CT can lead to huge computational costs with traditional model-based iterative reconstruction methods whereas deep networks can be much faster with training and validation. In this review, we present techniques, applications, uses, and limitations of deep learning-based image reconstruction methods in CT.

show abstract

“…Artificial intelligence is expected to soon play a key role in diagnostic imaging [1,2], radiation therapy [3,4] and medical physics in general. It has already been demonstrated to successfully improve image quality [1], decrease radiation dosage [5,6], assign label types and identify pathology locations [7][8][9][10][11][12][13], create and optimize protocols [14], accurately segment pathology areas and organs [15,16], and optimize technology utilization [17].…”

Section: Introductionmentioning

confidence: 99%

The EU medical device regulation: Implications for artificial intelligence-based medical device software in medical physics

Beckers¹,

Kwade²,

Zanca³

2021

Physica Medica

View full text Add to dashboard Cite

Medical device manufacturers are increasingly applying artificial intelligence (AI) to innovate their products and to improve patient outcomes. Health institutions are also developing their own algorithms, to address specific needs for which no commercial product exists.Although AI-based algorithms offer good prospects for improving patient outcomes, their wide adoption in clinical practice is still limited. The most significant barriers to the trust required for wider implementation are safety and clinical performance assurance .Qualified medical physicist experts (MPEs) play a key role in safety and performance assessment of such tools, before and during integration in clinical practice. As AI methods drive clinical decision-making, their quality should be assured and tested. Occasionally, an MPE may be also involved in the in-house development of such an AI algorithm. It is therefore important for MPEs to be well informed about the current regulatory framework for Medical Devices.The new European Medical Device Regulation (EU MDR), with date of application set for 26 of May 2021, imposes stringent requirements that need to be met before such tools can be applied in clinical practice.The objective of this paper is to give MPEs perspective on how the EU MDR affects the development of AIbased medical device software. We present our perspective regarding how to implement a regulatory roadmap, from early-stage consideration through design and development, regulatory submission, and post-market surveillance. We have further included an explanation of how to set up a compliant quality management system to ensure reliable and consistent product quality, safety, and performance .

show abstract

Impact of Deep Learning-based Optimization Algorithm on Image Quality of Low-dose Coronary CT Angiography with Noise Reduction: A Prospective Study

Cited by 36 publications

References 27 publications

Performance evaluation of using shorter contrast injection and 70 kVp with deep learning image reconstruction for reduced contrast medium dose and radiation dose in coronary CT angiography for children: a pilot study

Performance evaluation of using shorter contrast injection and 70 kVp with deep learning image reconstruction for reduced contrast medium dose and radiation dose in coronary CT angiography for children: a pilot study

Artificial intelligence in image reconstruction: The change is here

The EU medical device regulation: Implications for artificial intelligence-based medical device software in medical physics

Contact Info

Product

Resources

About