Automated detection of the arterial input function using normalized cut clustering to determine cerebral perfusion by dynamic susceptibility contrast‐magnetic resonance imaging

Abstract

Purpose: To propose a new clustering method for the automatic detection of arterial input function (AIF) with high accuracy in dynamic susceptibility contrastmagnetic resonance imaging (DSC-MRI). Materials and Methods:A novel method for automatically determining the AIF was proposed to facilitate the analysis of MR perfusion, which relied on normalized cut (Ncut) clustering. Its performance was compared with those of two other previously reported clustering methods: k-means and fuzzy c-means (FCM) techniques, in terms of the detection accuracy and computational time. Both simulated perfusion data and data collected from 42 healthy human subjects were applied to investigate the feasibility of the proposed approach.Results: In the simulation study, the partial volume effect (PVE) level, peak value (PV), time to peak (TTP), full width at half maximum (FWHM), area under AIF curve (AUC), root mean square error (RMSE) between the estimated AIF and true AIF, and M value given by [PV/ (FWHMÂTTP)] were 45. 45, 4.2737, 29.92, 6.4563, 76.4836, 0.0519, and 0.0221 for Ncut-based AIF, 96.45, 3.8385, 31.74, 7.5133, 75.7364, 0.3295, and 0.0161 for FCM-based AIF, 91.18, 3.8990, 31.73, 7.4544, 76.0476, 0.3128, and 0.0165 for k-means-based AIF, 0, 4.4592, 29.51, 6.2016, 76.8669, 0, and Conclusion: Ncut clustering yield AIFs more in line with the expected AIF, and might be preferred to FCM and kmeans clustering methods sensitive to randomly selected initial centers.

Evaluation of an automated method for arterial input function detection for first-pass myocardial perfusion cardiovascular magnetic resonance

“…Peruzzo et al [ 12 ] method discards voxels that poorly fit the expected cerebral AIF characteristics and classifies the remaining voxels with agglomerative hierarchical clustering to select the AIF voxels. Yin et al [ 13 , 14 ] presented two studies, one using hierarchical clustering and another using a normalized cut clustering scheme to select the final cerebral AIF cluster.…”

Automated measurement of arterial input function in first-pass myocardial perfusion magnetic resonance images using independent component analysis

“…Next, a linear voting scheme is used to determine which two regions are most characteristic of the RV and LV cavities in the images. Features used for voting classification include: distance to the center, distance to other regions, size of each region, SI upslope, peak value (PV), time to peak (TTP), full width at half maximum (FWHM), and an M-value [9] which combines the previous three features as shown in Equation-1.…”

“…Other statistics of the AIF were also compared, including TTP, FWHM, PV, SI upslope, and M-value (see section 2.2.2). Accurate AIFs are normally characterized by high values of PV, upslope, M-value, and low values of TTP and FWHM [9]. The execution time of the automated algorithm was also measured on a dataset of 40 image series using an Intel Core i7-3770 3.4GHz CPU and compared against the time required to manually selected ROI for the AIF measurement.…”

“…Peruzzo et al's [7] method discards voxels poorly fitting expected AIF characteristics and then classifies the remaining voxels with agglomerative hierarchical clustering to select the AIF voxels. Yin et al [8,9] also presented two studies, one using the hierarchical clustering and another using a normalized cut clustering scheme to select the final AIF cluster.…”

Normalized Cut optimization based on color perception findings. A comparative study

“…It has been used in different applications [8,[13][14][15][16][17] and many variants have been presented [18][19][20][21][22][23]. In these works, the parameters required to compute the pixel-to-pixel similarity as well as the stopping condition were manually established for each and every image, a process that strongly diminishes the method's usability and makes the reproduction of their algorithms difficult.…”