The morphology of anisotropic 3D-printed hydroxyapatite scaffolds.

Three-dimensional (3D) scaffolds with tailored pores ranging from the nanometer to millimeter scale can support the reconstruction of centimeter-sized osseous defects. Three-dimensional-printing processes permit the voxel-wise fabrication of scaffolds. The present study rests upon 3D-printing with nano-porous hydroxyapatite granulates. The cylindrical design refers to a hollow bone with higher density at the periphery. The millimeter-wide central channel follows the symmetry axis and connects the perpendicularly arranged micro-pores. Synchrotron radiation-based micro computed tomography has served for the non-destructive characterization of the scaffolds. The 3D data treatment is essential, since, for example, the two-dimensional distance maps overestimate the mean distances to the material by 33-50% with respect to the 3D analysis. The scaffolds contain 70% micrometer-wide pores that are interconnected. Using virtual spheres, which might be related to the cells migrating along the pores, the central channel remains accessible through the micro-pores for spheres with a diameter of up to (350+/-35)mum. Registering the tomograms with their 3D-printing matrices has yielded the almost isotropic shrinking of (27+/-2)% owing to the sintering process. This registration also allows comparing the design and tomographic data in a quantitative manner to extract the quality of the fabricated scaffolds. Histological analysis of the scaffolds seeded with osteogenic-stimulated progenitor cells has confirmed the suitability of the 3D-printed scaffolds for potential clinical applications.

Assessment of 18F PET signals for automatic target volume definition in radiotherapy treatment planning.

INTRODUCTION: Positron emission tomography (PET) alone or in combination with computer tomography (PET/CT) is increasingly used in target volume assessment. A standardized way of converting PET signals into target volumes is not available at present. MATERIALS AND METHODS: Assuming a uniform signal emission from a tumour and surrounding normal tissues, a model-based method was developed to determine a relative threshold level (Th(rel)) for gross tumour volume delineation. Two phantoms consisting of cylindrical and spherical sources of diameter ranging from 4.5 to 43 mm in a tank and (18)F activities ranging from 0.001 to 0.15 MBq/ml for tank and sources, respectively, were used for PET/CT imaging. A Th(rel) was calculated that best corresponded to the physical diameter of the cylindrical sources. Software (SW) was generated to automatically delineate volumes based on this threshold. The SW was validated for in vitro and in vivo PET signals. RESULTS: The Th(rel) best representing the source diameter was 41+/-2.5% (95% confidence level) of the background-subtracted signal. The mean deviation for sources of diameter > or =12.5 mm was < or =1.5 mm. The Th(rel) was constant for diameters > or =12.5 mm. For source diameters <12.5 mm, the 41% level over-estimated the real source diameter by a factor depending on the diameter. In an in vitro set-up the SW was capable of segmenting solitary PET volumes to within 1.4 mm (1SD). For non-homogeneous signals in a clinical set-up minimal manual intervention is presently required to separate target from non-target signals. The SW may slightly underestimate target volumes when compared with CT-based volumes, but works well as a first approximation. The volume can be manually adapted to give the ultimate target volume. CONCLUSIONS: SW-based automatic delineation of the volume of (18)F activity is feasible and highly reproducible. Volumes can be subsequently modified by the clinician if necessary. This approach will increase the efficiency of the planning process.

Automated functional image-guided radiation treatment planning for rectal cancer.

PURPOSE: Computer tomography-based (CT-based) tumor-volume definition is time consuming and is subject to clinical interpretation. CT is not accessible for standardized algorithms for the purpose of treatment-volume planning. We have evaluated the accuracy of target-volume definition based on the positron emission tomography (PET) data from an integrated PET/CT system with 2-[(18)F]fluoro-2-deoxy-D-glucose (FDG) for standardized target-volume delineation. MATERIALS AND METHODS: Eleven patients with rectal cancer who were undergoing preoperative radiation therapy (RT) were studied. A standardized region-growing algorithm was tested to replace the CT-derived gross tumor volume by the PET-derived gross tumor volume (PET-GTV) or the biologic target volume (BTV). A software tool was developed to automatically delineate the appropriate tumor volume as defined by the FDG signal, the PET-GTV, and the planning target volume (PTV). The PET-derived volumes were compared with the target volumes from CT. RESULTS: The BTV defined for appropriate GTV assessment was set at a single peak threshold of 40% of the signal of interest. Immediate treatment volume definition based on the choice of a single-tumor volume-derived PET-voxel resulted in a tumor volume that strongly correlated with the CT-derived GTV (r(2) = 0.84; p < 0.01) and the volume as assessed on subsequent anatomic-pathologic analysis (r(2) = 0.77; p < 0.01). In providing sufficient extension margins from the CT-derived GTV and the PET-derived GTV, to PTV, respectively, the correlation of the CT-derived and PET-derived PTV was sufficiently accurate for PTV definition for external-beam therapy (r(2) = 0.96; p < 0.01). CONCLUSION: Automated segmentation of the PET signal from rectal cancer may allow immediate and sufficiently accurate definition of a preliminary working PTV for preoperative RT. If required, correction for anatomic precision and geometric resolution may be applied in a second step. Computed PET-based target-volume definition could be useful for the definition of standardized simultaneous internal-boost volumes for intensity-modulated radiation therapy (IMRT) based on biologic target volumes.

Non-destructive three-dimensional evaluation of a polymer sponge by micro-tomography using synchrotron radiation.

X-ray micro-tomography, a non-destructive technique is used to uncover the complex 3-D micro-architecture of a degradable polymer sponge designed for bone augmentation. The measurements performed at HASYLAB at DESY are based on a synchrotron radiation source resulting in a spatial resolution of about 5.4 microm. In the present communication we report the quantitative analysis of the porosity and of the pore architecture. First, we elucidate that synchrotron radiation at the photon energy of 9 keV has an appropriate cross section for this low-weight material. Modifications in sponge micro-architecture during measurement are not detected. Second, the treatment of the data, an amount of 2.5 Gbyte to generate binary data is described. We compare the 3-D with the 2-D analysis in a quantitative manner. The obtained values for the mean distance to material within the sponge calculated from 2-D and 3-D data of the whole tomogram differ significantly: 12.5 microm for 3-D and 17.6 microm for 2-D analysis. If the pores exhibit a spherical shape as frequently found, the derived mean pore diameter, however, is overestimated only by 6% in the 2-D image analysis with respect to the 3-D evaluation. This approach can be applied to different porous biomaterials and composites even in a hydrated state close to physiological conditions, where any surface preparation artifact is avoided.

Evaluation of complex joint motion with computer-based analysis of fluoroscopic sequences.

RATIONALE AND OBJECTIVES: To develop a computer-assisted analysis of complex joint motion based on standard fluoroscopic sequences. MATERIALS AND METHODS: Fluoroscopic sequences of 10 normal shoulders and 20 patients after total shoulder arthroplasty were recorded during abduction and adduction. The analysis of the shoulder motion was based on automated tracking of selected components (models) of the shoulder joint. After processing the digitized images with an edge-detection procedure, visible edges were defined as base models. These models were tracked frame by frame. Several automated postprocessing evaluation procedures were developed and tested. RESULTS: The amounts of rotation and translation of the glenohumeral joint can be quantified and related. The mean translation parallel to the glenoid in the normal shoulder joints was 2.8 mm and 1.9 mm in joints after shoulder arthroplasty. The mean translation perpendicular to the glenoid was 0.9 mm and 0.8 mm, respectively. In complex motion patterns (ie, combined glenohumeral and thoracoscapular motion), electronic stabilization of an object (ie, stabilization of scapula) and, thus, separate analysis of each component. CONCLUSION: Computer assisted analysis allows automated evaluation of complex joint motion based on standard fluoroscopic sequences.