Evaluation of principal component analysis-based data-driven respiratory gating for positron emission tomography

Walker, Matthew; Bradley, Kevin M.; McGowan, Daniel R.

doi:10.1259/bjr.20170793

Cited by 30 publications

(23 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The observed increase due to respiratory gating was both expected and similar to that reported by others [26]. A comparison against the RPM device was made previously using anthropomorphic phantom data, for which the true driving waveform was known [12]. Although the similarities between waveforms from the PCA-DDG algorithm and the RPM system have been presented for several patients [7, 13], the benefit from such a comparison is limited by the lack of a gold standard.…”

Section: Discussionsupporting

confidence: 83%

“…The variation of the weighting factor for this component over time provides a respiratory waveform similar to that provided by the external respiratory gating equipment. The method has been recently validated in phantom studies, and several patient examples have been presented [11, 12]. The commercial implementation is built on the work of Thielmans et al [7, 13, 14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of data-driven respiratory gating waveforms for clinical PET imaging

et al. 2019

View full text Add to dashboard Cite

BackgroundWe aimed to evaluate the clinical robustness of a commercially developed data-driven respiratory gating algorithm based on principal component analysis, for use in routine PET imaging.MethodsOne hundred fifty-seven adult FDG PET examinations comprising a total of 1149 acquired bed positions were used for the assessment. These data are representative of FDG scans currently performed at our institution. Data were acquired for 4 min/bed position (3 min/bed for legs). The data-driven gating (DDG) algorithm was applied to each bed position, including those where minimal respiratory motion was expected. The algorithm provided a signal-to-noise measure of respiratory-like frequencies within the data, denoted as R. Qualitative evaluation was performed by visual examination of the waveforms, with each waveform scored on a 3-point scale by two readers and then averaged (score S of 0 = no respiratory signal, 1 = some respiratory-like signal but indeterminate, 2 = acceptable signal considered to be respiratory). Images were reconstructed using quiescent period gating and compared with non-gated images reconstructed with a matched number of coincidences. If present, the SUVmax of a well-defined lesion in the thorax or abdomen was measured and compared between the two reconstructions.ResultsThere was a strong (r = 0.86) and significant correlation between R and scores S. Eighty-six percent of waveforms with R ≥ 15 were scored as acceptable for respiratory gating. On average, there were 1.2 bed positions per patient examination with R ≥ 15. Waveforms with high R and S were found to originate from bed positions corresponding to the thorax and abdomen: 90% of waveforms with R ≥ 15 had bed centres in the range 5.6 cm superior to 27 cm inferior from the dome of the liver. For regions where respiratory motion was expected to be minimal, R tended to be < 6 and S tended to be 0. The use of DDG significantly increased the SUVmax of focal lesions, by an average of 11% when considering lesions in bed positions with R ≥ 15.ConclusionsThe majority of waveforms with high R corresponded to the part of the patient where respiratory motion was expected. The waveforms were deemed suitable for respiratory gating when assessed visually, and when used were found to increase SUVmax in focal lesions.

show abstract

Section: Discussionsupporting

confidence: 83%

Section: Introductionmentioning

confidence: 99%

Evaluation of data-driven respiratory gating waveforms for clinical PET imaging

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Early versions of this method were developed and evaluated by Thielemans et al (5)(6)(7). The commercial version has been evaluated in phantom studies where it was found to reliably provide a respiratory waveform (8). The quality of the waveforms extracted by the algorithm have been assessed from clinical 18 F-FDG PET data which shows an example of suitable and unsuitable waveforms (9).…”

Section: Introductionmentioning

confidence: 99%

Data-Driven Respiratory Gating Outperforms Device-Based Gating for Clinical ¹⁸F-FDG PET/CT

et al. 2020

View full text Add to dashboard Cite

Methods 144 whole-body 18 F-FDG PET/CT examinations were acquired using a Discovery D690 or D710 PET/CT scanner (GE Healthcare), with a respiratory gating waveform recorded by an external, device-based respiratory gating system. In each examination, two of the bed positions covering the liver and lung bases were acquired with duration of 6 minutes. Quiescent period gating retaining ~50% of coincidences was then able to produce images with an effective duration of 3 minutes for these two bed positions, matching the other bed positions. For each exam, 4 reconstructions were performed and compared: data-driven gating (DDG-retro), external devicebased gating (RPM-gated), no gating but using only the first 3 minutes of data (Ungated matched), and no gating retaining all coincidences (Ungated full). Lesions in the images were quantified and image quality was scored by a radiologist, blinded to the method of data processing. Results The use of DDG-retro was found to increase SUV max and to decrease the threshold-defined lesion volume in comparison to each of the other reconstruction options. Compared to RPMgated, DDG-retro gave an average increase in SUV max of 0.66±0.1 g/mL (n=87,p<0.0005). Although results from the blinded image evaluation were most commonly equivalent, DDG-retro was preferred over RPM-gated in 13% of exams while the opposite occurred in just 2% of exams. This was a significant preference for DDG-retro (p=0.008,n=121). Liver lesions were identified in 23 exams. Considering this subset of data, DDG-retro was ranked superior to Ungated full in 6/23 (26%) of cases. Gated reconstruction using the external device failed in 16% of exams, while DDG-retro always provided a clinically acceptable image. Conclusion In this clinical evaluation, the data-driven respiratory gating technique provided superior performance as compared to the external device-based system. For the majority of exams the performance was equivalent, but data-driven respiratory gating had superior performance in 13% of exams, leading to a significant preference overall.

show abstract

“…Principal component analysis is often used for data compression. An example of principal component analysis in nuclear medicine is to reduce respiration artifacts in PET scans (23). In this case, the task requires identifying the data that are most suggestive of breathing and using this information during the reconstruction process to reduce breathing artifacts.…”

Section: Anns (Fig 4)mentioning

confidence: 99%

Machine Learning in Nuclear Medicine: Part 1—Introduction

Uribe¹,

Mathotaarachchi

Gaudet

et al. 2019

J Nucl Med

View full text Add to dashboard Cite

Learning Objectives: On successful completion of this activity, participants should be able to (1) provide an introduction to machine learning, neural networks, and deep learning; (2) discuss common machine learning algorithms with illustrative examples and figures; and (3) compare machine learning algorithms and provide guidance on selection for a given application. Financial Disclosure: Sandra E. Black received in-kind funding to her institution from GE Healthcare and Avid Pharmaceuticals. The authors of this article have indicated no other relevant relationships that could be perceived as a real or apparent conflict of interest. CME Credit: SNMMI is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to sponsor continuing education for physicians. SNMMI designates each JNM continuing education article for a maximum of 2.0 AMA PRA Category 1 Credits. Physicians should claim only credit commensurate with the extent of their participation in the activity. For CE credit, SAM, and other credit types, participants can access this activity through the SNMMI website (http://www.snmmilearningcenter.org) through April 2022. This article, the first in a 2-part series, provides an introduction to machine learning (ML) in a nuclear medicine context. This part addresses the history of ML and describes common algorithms, with illustrations of when they can be helpful in nuclear medicine. Part 2 focuses on current contributions of ML to our field, addresses future expectations and limitations, and provides a critical appraisal of what ML can and cannot do.

show abstract

Evaluation of principal component analysis-based data-driven respiratory gating for positron emission tomography

Cited by 30 publications

References 33 publications

Evaluation of data-driven respiratory gating waveforms for clinical PET imaging

Evaluation of data-driven respiratory gating waveforms for clinical PET imaging

Data-Driven Respiratory Gating Outperforms Device-Based Gating for Clinical ¹⁸F-FDG PET/CT

Machine Learning in Nuclear Medicine: Part 1—Introduction

Contact Info

Product

Resources

About

Evaluation of principal component analysis-based data-driven respiratory gating for positron emission tomography

Cited by 30 publications

References 33 publications

Evaluation of data-driven respiratory gating waveforms for clinical PET imaging

Evaluation of data-driven respiratory gating waveforms for clinical PET imaging

Data-Driven Respiratory Gating Outperforms Device-Based Gating for Clinical 18F-FDG PET/CT

Machine Learning in Nuclear Medicine: Part 1—Introduction

Contact Info

Product

Resources

About

Data-Driven Respiratory Gating Outperforms Device-Based Gating for Clinical ¹⁸F-FDG PET/CT