Joakim Crafoord

A Versatile Functional–Anatomic Image Fusion Method for Volume Data Sets

We describe and validate a volumetric three-dimensional registration method, and compare it to our previously validated two-dimensional/three-dimensional method. CT/MRI and SPECT data from 14 patients were interactively fused using a polynomial warping technique. Registration accuracy was confirmed visually and by a nonsignificant F value from multivariate analysis of the transformed landmarks, a significant difference of the squared sum of intensity differences between the transformed/untransformed and the reference volume both at the 0.05 (p > 0.05) confidence level and an average 31% improvement of the correlation coefficient and cross correlation. For the two-dimensional/three-dimensional method, ROI center-to-center distance ranged from 1.42 to 11.32 mm (for liver) with an average of 6.13 mm ± 3.09 mm. The average ROI overlap was 92.51% with a 95% confidence interval of 90.20–96.88%. The new method is superior because it operates on the true three-dimensional volume. Both methods give good registration results, take 10 to 30 min, and require anatomic knowledge.

Computed Tomography vs. Digital Radiography Assessment for Detection of Osteolysis in Asymptomatic Patients With Uncemented Cups

Digital radiographic images in the anterior-posterior and lateral view have been gold standard for evaluation of peri-acetabular osteolysis for patients with an uncemented hip replacement. We compared digital radiographic images and computer tomography in detection of peri-acetabular osteolysis and devised a classification system based on computer tomography. Digital radiographs were compared with computer tomography on 206 hips, with a mean follow up 10 years after surgery. The patients had no clinical signs of osteolysis and none were planned for revision surgery. On digital radiographs, 192 cases had no osteolysis and only 14 cases had osteolysis. When using computer tomography there were 184 cases showing small or large osteolysis and only 22 patients had no osteolysis. A classification system for peri-acetabular osteolysis is proposed based on computer tomography that is easy to use on standard follow up evaluation.

A whole body atlas for segmentation and delineation of organs for radiation therapy planning

A semi-automatic procedure for delineation of organs, to be used as the basis of a whole body atlas database for radiation therapy planning was developed. The Visible Human Male Computed Tomography ...

Deviation between navigated and final 3-dimensional implant position in mini-invasive unicompartmental knee arthroplasty: A pilot study in 13 patients

Background and purpose Unicompartmental knee arthroplasty (UKA) is an established method of treating isolated gonartrosis. Modern techniques such as computer-assisted surgery (CAS) and minimally invasive surgery (MIS) are attractive complementary methods to UKA. However, the positioning of the components remains a concern. Thus, we performed a prospective study to assess whether there was deviation between the navigated implant position and the final implant position. Patients and methods We performed UKA with MIS and CAS in 13 patients. By comparing intraoperative navigation data with postoperative computed tomography (CT) measurements, we calculated the deviation between the computer-assisted implant position and the final 3-D implant position of the femoral and tibial components. Results The computer-assisted placement of the femoral and tibial component showed adequate position and consistent results regarding flexion-extension and varus-valgus. However, regarding rotation there was a large variation and 6 of 10 patients were outside the target range for both the femoral component and the tibial component. Interpretation Difficulties in assessing anatomical landmarks with the CAS in combination with MIS might be a reason for the poor rotational alignment of the components.

Risk Factors for Periacetabular Osteolysis and Wear in Asymptomatic Patients with Uncemented Total Hip Arthroplasties

Osteolysis is a silent disease leading to aseptic loosening. This has not been studied in a cohort of asymptomatic patients. The aim of this study was to detect factors that might be associated with the development of periacetabular osteolysis and wear around an uncemented cup. We assessed 206 patients with an uncemented cup, measuring wear and periacetabular osteolysis using computed tomography with a median follow-up of 10 years after surgery (range 7–14 years). EQ5D, pain from the hip, and satisfaction were assessed. The association between periacetabular osteolysis and wear, age, gender, activity, BMI, cup type, cup age, positioning of the cup, and surface coating was investigated with a proportional odds model. Wear and male gender were associated with an increased risk for periacetabular osteolysis. There was no association with periacetabular osteolysis for time from operation, patient age, UCLA Activity Score, liner thickness at time of operation, BMI, cup positioning, and type of implant. A thin liner at time of operation is correlated to increased wear. Linear wear rate was 0.18 mm/year and 46 of 206 patients had large periacetabular osteolysis. Asymptomatic patients with these implants should be followed up on a regular basis with a sensitive method such as CT in order to detect complications early.

Eliminated risk of iodine contrast cancellation with multibin spectral CT

This note compares the extent of contrast cancellation induced by iodinated contrast agents in energy integrating and photon counting multibin CT images. The contrast between a hypodense target and soft tissue is modeled for the two systems for a range of iodine concentrations and tube voltages. In energy integrating systems, we show that the contrast vanishes for low concentrations of iodine whereas the same effect is not seen in multibin systems. We conclude that it is the ability of multibin systems to apply weighting schemes post-acquisition that allows the operator to eliminate the risk of contrast cancellation between iodinated targets and the background.

Clinical Applications from Head to Toe Using a Semiautomatic 3D Inter/Intramodality Fusion Technique

Automated image registration techniques, particularly between modalities which clearly display anatomy and especially within the head, have become commonplace. Meanwhile advances in molecular imaging and the need to perform registration in other areas of body has driven the development of nonrigid and semiautomatic 3D volume fusion methods. This paper will focus on this latter class of 3D volume registration methods for a variety of clinical applications.

A Whole Body Atlas Based Segmentation for Delineation of Organs in Radiation Therapy Planning

Optimal radiation therapy (RT) can be achieved when accurate knowledge about the exact location of the target volume to be treated with respect to all organs at risk is available. A whole body atlas (WBA) can be utilized to convert the anatomy of a “standard man” into individual patients by applying warping on anatomical images and the anatomy of the atlas can be adjusted to an individual patient 2,3. The purpose of this work is to propose a semi-automatic segmentation procedure that utilizes polynomial warping together with active contour models, which could be used with WBA to delineate different organs in RT planning [4].

Comparison of three methods for registration of abdominal/pelvic volume data sets from functional-anatomic scans

Comparison of Three Image Methods for Registration of Abdominal/Pelvic Volume Data Sets from Functional-Anatomic Scans

Evaluation of polynomial image deformation for matching of 3D-abdominal MR-images using anatomical landmarks and for atlas construction

The aim of this study is to compare and evaluate the potential usability of linear and non-linear (polynomial) 3D-warping for constructing an atlas by matching abdominal MR-images from a number of different individuals using manually picked anatomical landmarks. The significance of this study lies in the fact that it illustrates the potential to use polynomial matching at a local or organ level. This is a necessary requirement for constructing an atlas and for fine intra-patient image matching and fusion. Finally 3D-image warping using anatomical landmark for inter-patient intra-modality image co-registration and fusion was found to be a very powerful and robust method. Additionally it can be used for intra-patient inter-modality image matching.