The feasibility of K i parametric imaging in clinical dynamic 18 F‐FDG total‐body PET using a simulated‐data‐driven machine learning algorithm

Despite K parametric images providing high sensitivity and specificity in clinical diagnosis and therapeutic evaluation, their clinical application is limited by prolonged scan durations. Previous research has employed machine learning algorithms to generate reliable K images from F-FDG PET scans wi...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Medical physics (Lancaster) Ročník 52; číslo 10; s. e70053
Hlavní autori: Gu, Wenjian, Liu, Weiping, Yang, Wentong, Wang, Yihan, Liu, Ze, Zhu, Zhanshi, Li, Yanxiao, Xu, Tianyi, Shi, Hongcheng, Wang, Kuanquan, Luo, Gongning, Chen, Xiaojun, Zhou, Yun
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States 01.10.2025
Predmet:
Algorithms
Feasibility Studies
Fluorodeoxyglucose F18
Humans
Image Processing, Computer-Assisted - methods
Machine Learning
Positron-Emission Tomography - methods
Whole Body Imaging - methods
total‐body PET
Patlak plot
machine learning
parametric image
18F‐FDG
ISSN:0094-2405, 2473-4209
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Despite K parametric images providing high sensitivity and specificity in clinical diagnosis and therapeutic evaluation, their clinical application is limited by prolonged scan durations. Previous research has employed machine learning algorithms to generate reliable K images from F-FDG PET scans with a duration of 10 min, but these algorithms require extensive real-world data for effective training. This study explored the feasibility of a simulated-data-driven approach for clinical K parametric imaging with reduced scan duration. A cluster analysis was conducted using K-Means on 60-minute dynamic total-body PET data from 25 subjects, followed by the construction of a noise-free simulated dataset using the Patlak equation. Subsequently, noise at varying levels was incorporated into the noise-free simulated dataset, resulting in four distinct training sets. Each dataset contains 1 200 000 simulated K values as labels, along with corresponding tissue-to-blood concentration ratios categorized by noise level (free, low, middle, and high). Based on the same simulation methodology, we also generated four corresponding test datasets at different noise levels (each with identical size to the training dataset). Ultimately, XGBoost models, each trained with data at different noise levels, were employed to predict K values from short-duration (50-60 min) PET scans. The accuracy of the K values generated from short-duration scans by our proposed method was evaluated using both simulated data (the four test datasets at varying noise levels) and real-world data (dynamic total-body PET data from 25 subjects). The K images generated by the conventional Patlak method with a t* of 20 min post-injection were chosen as the gold standard. Evaluation based on both simulated and real-world data indicates that training algorithms with a noise-inclusive simulated dataset significantly improve the average accuracy of K values. In evaluations with real-world data, K images from 50-60-minute dynamic PET scans generated using our proposed approach demonstrated superior performance compared to the conventional Patlak method, achieving a Pearson's correlation coefficient of 0.94 (vs. 0.42 for Patlak), a lower normalized mean square error of 0.11 (vs. 5.33), and a higher peak signal-to-noise ratio of 64.32 (vs. 47.87). The simulated-data-driven approach can generate reliable K images from clinical F-FDG dynamic total-body PET scans, thereby reducing the costs associated with collecting real-world data.
ISSN:0094-2405
2473-4209
DOI:10.1002/mp.70053