Neural network-aided unsupervised input function estimation for dual-time-window PET Patlak analysis.
[PURPOSE] This study aims to develop and validate a dual-time-window (DTW) Patlak plot method that eliminates the need for invasive blood sampling and reduces scan duration.
APA
Shao W, Zhang Y, et al. (2025). Neural network-aided unsupervised input function estimation for dual-time-window PET Patlak analysis.. EJNMMI physics, 12(1), 100. https://doi.org/10.1186/s40658-025-00804-w
MLA
Shao W, et al.. "Neural network-aided unsupervised input function estimation for dual-time-window PET Patlak analysis.." EJNMMI physics, vol. 12, no. 1, 2025, pp. 100.
PMID
41359071
Abstract
[PURPOSE] This study aims to develop and validate a dual-time-window (DTW) Patlak plot method that eliminates the need for invasive blood sampling and reduces scan duration. We seek to improve the accuracy of the net influx constant ([Formula: see text]) estimation, addressing the inaccuracies inherent in traditional DTW and single-time-window methods, which often introduce bias and hinder comparability across different cohorts.
[METHOD] We developed an unsupervised, multi-branch neural network (NN) to assist in estimating missing data intervals within the DTW protocol, thereby facilitating accurate Patlak analysis. The model fits the mapping from time to the time-activity curve (TAC), generating multiple pseudo input functions (IFs). A correlation coefficient is then computed between each pseudo IF and the voxel-level measured data, extracting statistical information guided by the kinetic process. These correlation scores were used to construct a weighted statistic, serving as the final IF (NNIF). Our approach was validated using both simulation and clinical data, including [Formula: see text]-FDG PET scans from 67 lung cancer subjects. Additionally, we compared the performance of our method with other simplified quantification techniques to demonstrate its efficacy in achieving high-quality parametric imaging and reliable quantitative analysis within abbreviated scanning protocols.
[RESULT] Our proposed method achieved high accuracy in the estimation of IF, with a maximum mean absolute deviation (MAD) of 0.04 in a real patient study. The regressed [Formula: see text] derived from different DTW scan protocols exhibited good consistency. In simulation studies , the best relative absolute error (RAE) was 0.0302. In real patient study, the optimal average peak signal-to-noise ratio (PSNR) of parametric imaging reached 97.40 dB, while the best average R-squared ([Formula: see text]) in ROI-based quantitative analysis reached 0.991.
[CONCLUSIONS] We demonstrate the feasibility of using a weighted statistic, constructed from a multi-branch neural network, to accurately estimate the complete IF. This approach enables the generation of high-quality parametric images with shortened scan protocols, effectively reducing scanning time while ensuring accurate Patlak analysis.
[METHOD] We developed an unsupervised, multi-branch neural network (NN) to assist in estimating missing data intervals within the DTW protocol, thereby facilitating accurate Patlak analysis. The model fits the mapping from time to the time-activity curve (TAC), generating multiple pseudo input functions (IFs). A correlation coefficient is then computed between each pseudo IF and the voxel-level measured data, extracting statistical information guided by the kinetic process. These correlation scores were used to construct a weighted statistic, serving as the final IF (NNIF). Our approach was validated using both simulation and clinical data, including [Formula: see text]-FDG PET scans from 67 lung cancer subjects. Additionally, we compared the performance of our method with other simplified quantification techniques to demonstrate its efficacy in achieving high-quality parametric imaging and reliable quantitative analysis within abbreviated scanning protocols.
[RESULT] Our proposed method achieved high accuracy in the estimation of IF, with a maximum mean absolute deviation (MAD) of 0.04 in a real patient study. The regressed [Formula: see text] derived from different DTW scan protocols exhibited good consistency. In simulation studies , the best relative absolute error (RAE) was 0.0302. In real patient study, the optimal average peak signal-to-noise ratio (PSNR) of parametric imaging reached 97.40 dB, while the best average R-squared ([Formula: see text]) in ROI-based quantitative analysis reached 0.991.
[CONCLUSIONS] We demonstrate the feasibility of using a weighted statistic, constructed from a multi-branch neural network, to accurately estimate the complete IF. This approach enables the generation of high-quality parametric images with shortened scan protocols, effectively reducing scanning time while ensuring accurate Patlak analysis.
같은 제1저자의 인용 많은 논문 (5)
- Accurate diagnosis of non-small cell and small cell lung cancer by using machine learning models trained with physical science features extracted from pathological images.
- Clinical trials of bispecific antibody therapy for colorectal cancer: advanced and next steps.
- A systematic review and meta-analysis of exposure-response analysis of osimertinib in patients with non-small-cell lung cancer.
- Targeted lipidomics meets transcriptomics: how cinobufagin rewires fatty acid, sphingolipid, and glycerophospholipid metabolism to combat hepatoma cell growth.
- Association between diffusion tensor imaging analysis along the perivascular space (DTI-ALPS)-based glial-lymphatic dysfunction and cognitive impairment in non-small cell lung cancer.