Intellispace portal 9

LV functional measurements, including end‐diastolic volume, end‐systolic volume, ejection fraction, and systolic volume, were computed in the 3 different datasets. These measurements were performed using a commercially available semi‐automatic software package, IntelliSpacePortal 9.0 (Philips Healthcare, Best, The Netherlands). Bland–Altman analysis³⁰ and paired 2‐tailed Student t test were used to compare the LV functional parameters of FB^rad with BH^cart.

A senior radiologist (T. L.) evaluated the MR images. A second senior radiologist (K. L.) validated the measurements in a subgroup. DTI raw data post-processing and complete MRI analysis were performed using IntelliSpace Portal (IntelliSpace Portal 9.0, Philips Healthcare, Amsterdam, The Netherlands).
In the DTI sequence, the sciatic nerve was examined using six freehand drawn ROIs in six adjacent slices of color-coded fractional anisotropy (FA) images in correlation with the anatomical information of the b = 0 and 2D T2 TSE images. The mean of these six FA values was then determined to obtain the final FA value of each subject. Fiber tracking of the nerve was performed to depict the examined part of the thigh.
To determine the average intramuscular fat fractions and T₂ times, respectively, subtotal ROIs were drawn freehand into each part of the quadriceps femoris muscle (vastus lateralis, intermedius, medialis, rectus femoris) and the short and long heads of the biceps femoris muscle on the most proximal slice in each of the PDFF and T₂ maps. The ROIs were drawn within 2 mm of the muscle boundaries. The differing area sizes (A_i) of the individual ROIs (ROI_i with individual fat fractions (FF_i)) were taken into account using the formula FF_mean_over_ROIs = sum (A_i × FF_i)/sum (A_i), where the sum is the summation over all ROIs.

Images were processed on a dedicated workstation (IntelliSpace Portal 9.0, Philips) to compute multiplanar reconstructions (MPR), maximum intensity projections (MIP), and volume rendering (VR) images. The CAD-RADS assessment categories and modifiers [10 (link)], quantitative and qualitative images analyses were performed by two radiologists with 4 (reader 1) and 7 years of experience (reader 2) in CCTA, CTA, and 3D vascular images interpretation, blinded each other and to clinical data.

Details of the DECT scan technique are provided in Supplementary Material 1. Polyenergetic images (PEIs) were generated using an iterative reconstruction algorithm (iDose 4, level 3; Philips Healthcare) to represent conventional CT image sets and then transferred to the picture archiving and communication system. Furthermore, 40-keV virtual monoenergetic images (VMIs) [22 (link)–24 (link)] and other spectral-based imaging datasets, including material-decomposition maps of iodine density (ID) and effective atomic numbers (Z_eff) maps, were generated using the postprocessing workstation (IntelliSpace Portal 9.0, Philips Healthcare).

Two blind-ended radiologists having MRI experience of more than 10 years independently calculated the ADC value in the breast lesions after placing a round or elliptical region of interest (ROI) in the IntelliSpace Portal 9.0 operating system console of Philips. Three uniform sizes ROIs were placed in the ADC map image, and then, the mean ADC value was obtained for analysis.

TKV was estimated through multiplanar reformation and measured using the IntelliSpace Portal 9.0 workstation (Philips, Amsterdam, The Netherlands). The maximal longitudinal length (L) of the kidney was determined on T2-weighted tilted coronal slices parallel to the long renal axis (Figure 2D). The maximal width (W) was determined perpendicular to the L in the same plane in which the L was localized (Figure 2C). Finally, the maximal depth (D) was determined perpendicular to the L in a thick sagittal slice (Figure 2D) [18 (link)]. TKV was estimated as follows: $$\mathrm{Estimated}\mathrm{TKV}\left(\mathrm{mL}\right)=\frac{\mathsf{\pi }}{6}\times \mathrm{L}\times \mathrm{W}\times \mathrm{D}$$
Nonenhanced MRI and CT images at the L3 level were analyzed to determine TAM, VAT, and SAT areas processed on a compatible computer by using open-source software (ImageJ version 1.51; National Institutes of Health, Bethesda, MD, USA). Contours were obtained using a manual tracing method (Figure 3). A radiologist (C.-H.W.) with 10 years of experience in abdominal imaging processed the images based on the method described previously [20 (link)]. The TAM, VAT, and SAT indices were calculated as follows: TAM, VAT, and SAT area/(BH [m])². Sarcopenia was defined as a TAM index of <52.4 cm²/m² in men and <38.5 cm²/m² in women, according to Prado’s protocol [21 (link)].

For both the test and the retest datasets, the PET and low-dose CT images were processed independently. Imaging reading was performed using dedicated software for PET/CT imaging (Philips IntelliSpace Portal 9.0, Eindhoven, The Netherlands).
In PET, tumor lesion size was measured in the axial plane using a fixed PET windowing upper level (UL) of 10 used for stretching of the greyscale. Both long and short axis were measured; lesion area was calculated according to the simple formula for round and oval lesions: A = π × half long axis × half short axis. Within this area, the pixel with the highest standardized uptake value is designated the SUVmax (injected dose/kg body weight). Thus SUVmax was measured in all tumor lesions. The small size of most of the lesions did not allow for measurement of other meaningful SUVs such as SUVpeak that need lesions of at least 1 cm³. Low dose CT was used to check for appropriateness of the lesion area measured in PET if possible (i.e., with the exception of some bone metastases not visible on CT).