Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quantitative evaluation of unsupervised clustering algorithms for dynamic total-body PET image analysis

Oona Rainio, Maria K. Jaakkola, Riku Klén

TL;DR

This study addresses the lack of systematic evaluation of unsupervised clustering algorithms for dynamic total-body PET TAC analysis. It benchmarks 15 clustering methods on TACs from five organs across 30 patients using 5000 TACs per image, enabling quantitative comparison. The top performers—GMM, FCM, and ICA with MBK—achieve median accuracies around $0.89$, $0.83$, and $0.81$, respectively, with rapid processing times, informing automatic segmentation and analysis workflows. The results motivate further integration of these methods into whole-body PET pipelines and encourage exploration of location-aware and supervised alternatives for improved TAC clustering.

Abstract

Background. Recently, dynamic total-body positron emission tomography (PET) imaging has become possible due to new scanner devices. While clustering algorithms have been proposed for PET analysis already earlier, there is still little research systematically evaluating these algorithms for processing of dynamic total-body PET images. Materials and methods. Here, we compare the performance of 15 unsupervised clustering methods, including K-means either by itself or after principal component analysis (PCA) or independent component analysis (ICA), Gaussian mixture model (GMM), fuzzy c-means (FCM), agglomerative clustering, spectral clustering, and several newer clustering algorithms, for classifying time activity curves (TACs) in dynamic PET images. We use dynamic total-body $^{15}$O-water PET images collected from 30 patients with suspected or confirmed coronary artery disease. To evaluate the clustering algorithms in a quantitative way, we use them to classify 5000 TACs from each image based on whether the curve is taken from brain, right heart ventricle, right kidney, lower right lung lobe, or urinary bladder. Results. According to our results, the best methods are GMM, FCM, and ICA combined with mini batch K-means, which classified the TACs with a median accuracies of 89\%, 83\%, and 81\%, respectively, in a processing time of half a second or less on average for each image. Conclusion. GMM, FCM, and ICA with mini batch K-means show promise for dynamic total-body PET analysis.

Quantitative evaluation of unsupervised clustering algorithms for dynamic total-body PET image analysis

TL;DR

This study addresses the lack of systematic evaluation of unsupervised clustering algorithms for dynamic total-body PET TAC analysis. It benchmarks 15 clustering methods on TACs from five organs across 30 patients using 5000 TACs per image, enabling quantitative comparison. The top performers—GMM, FCM, and ICA with MBK—achieve median accuracies around , , and , respectively, with rapid processing times, informing automatic segmentation and analysis workflows. The results motivate further integration of these methods into whole-body PET pipelines and encourage exploration of location-aware and supervised alternatives for improved TAC clustering.

Abstract

Background. Recently, dynamic total-body positron emission tomography (PET) imaging has become possible due to new scanner devices. While clustering algorithms have been proposed for PET analysis already earlier, there is still little research systematically evaluating these algorithms for processing of dynamic total-body PET images. Materials and methods. Here, we compare the performance of 15 unsupervised clustering methods, including K-means either by itself or after principal component analysis (PCA) or independent component analysis (ICA), Gaussian mixture model (GMM), fuzzy c-means (FCM), agglomerative clustering, spectral clustering, and several newer clustering algorithms, for classifying time activity curves (TACs) in dynamic PET images. We use dynamic total-body O-water PET images collected from 30 patients with suspected or confirmed coronary artery disease. To evaluate the clustering algorithms in a quantitative way, we use them to classify 5000 TACs from each image based on whether the curve is taken from brain, right heart ventricle, right kidney, lower right lung lobe, or urinary bladder. Results. According to our results, the best methods are GMM, FCM, and ICA combined with mini batch K-means, which classified the TACs with a median accuracies of 89\%, 83\%, and 81\%, respectively, in a processing time of half a second or less on average for each image. Conclusion. GMM, FCM, and ICA with mini batch K-means show promise for dynamic total-body PET analysis.

Paper Structure

This paper contains 10 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Materials and methods
  3. Software requirements
  4. Data
  5. Clustering algorithms
  6. Evaluation of clusters
  7. Statistical testing
  8. Results
  9. Discussion
  10. Conclusion

Figures (2)

  • Figure 1: The same coronal slice in five different time frames of a dynamic PET image of one patient. The number of the time frame and the number of seconds (s) passed since the starting time is specified in the subcaptions.
  • Figure 2: 1000 time activity curves chosen from five different organs from a dynamic PET image of one patient. The $x$-axis varies from 0 to 23 based on the 24 time frames while the $y$-axis has the numeric elements of the TACs. The TACs themselves are in blue and their mean value curve is denoted in black.