Patient-Specific Semi-Supervised Learning for Postoperative Brain Tumor Segmentation

In contrast to preoperative brain tumor segmentation, the problem of postoperative brain tumor segmentation has been rarely approached so far. We present a fully-automatic segmentation method using multimodal magnetic resonance image data and patient-specific semi-supervised learning. The idea behind our semi-supervised approach is to effectively fuse information from both pre- and postoperative image data of the same patient to improve segmentation of the postoperative image. We pose image segmentation as a classification problem and solve it by adopting a semi-supervised decision forest. The method is evaluated on a cohort of 10 high-grade glioma patients, with segmentation performance and computation time comparable or superior to a state-of-the-art brain tumor segmentation method. Moreover, our results confirm that the inclusion of preoperative MR images lead to a better performance regarding postoperative brain tumor segmentation.

Technical complications and failures of zirconia-based prostheses supported by implants followed up to 7 years: a case series.

PURPOSE To evaluate technical complications and failures of zirconia-based fixed prostheses supported by implants. MATERIALS AND METHODS Consecutive patients received zirconia-based single crowns (SCs) and fixed dental prostheses (FDPs) on implants in a private clinical setting between 2005 and 2010. One dentist performed all surgical and prosthetic procedures, and one master technician performed and coordinated all laboratory procedures. One-piece computer-aided design/ computer-assisted manufacture technology was used to fabricate abutments and frameworks, which were directly connected at the implant level, where possible. All patients were involved in a recall maintenance program and were finally reviewed in 2012. Data on framework fractures, chipping of veneering ceramics, and other technical complications were recorded. The primary endpoint was failure of the prostheses, ie, the need for a complete remake. A life table analysis was calculated. RESULTS A total of 289 implants supported 193 zirconia-based prostheses (120 SCs and 73 FDPs) in 127 patients (51 men, 76 women; average age: 62.5 ± 13.4 years) who were reviewed in 2012. Twenty-five (13%) prostheses were cemented on 44 zirconia abutments and 168 (87%) prostheses were screw-retained directly at the implant level. Fracture of 3 frameworks (1 SC, 2 FDPs) was recorded, and significant chipping resulted in the remake of 3 prostheses (1 SC, 2 FDPs). The 7-year cumulative survival rate was 96.4% ± 1.99%. Minor complications comprised 5 loose screws (these were retightened), small chips associated with 3 prostheses (these were polished), and dislodgement of 3 prostheses (these were recemented). Overall, 176 prostheses remained free of technical problems. CONCLUSIONS Zirconia-based prostheses screwed directly to implants are clinically successful in the short and medium term.

A Longitudinal Study of Mental Health in Emerging Adults

Emerging adulthood is a time of instability. This longitudinal study investigated the relationship between mental health and need satisfaction among emerging adults over a period of five years and focused on gender-specific differences. Two possible causal models were examined: (1) the mental health model, which predicts that incongruence is due to the presence of impaired mental health at an earlier point in time; (2) the consistency model, which predicts that impaired mental health is due to a higher level of incongruence reported at an earlier point in time. Emerging adults (N = 1,017) aged 18–24 completed computer-assisted telephone interviews in 2003 (T1), 2005 (T2), and 2008 (T3). The results indicate that better mental health at T1 predicts a lower level of incongruence two years later (T2), when prior level of incongruence is controlled for. The same cross-lagged effect is shown for T3. However, the cross-lagged paths from incongruence to mental health are marginally associated when prior mental health is controlled for. No gender differences were found in the cross-lagged model. The results support the mental health model and show that incongruence does not have a long-lasting negative effect on mental health. The results highlight the importance of identifying emerging adults with poor mental health early to provide support regarding need satisfaction.

Preliminary Study on Finite Element Simulation for Optimizing Acetabulum Reorientation after Periacetabular Osteotomy

Periacetabular osteotomy (PAO) is an effective approach for surgical treatment of hip dysplasia. The aim of PAO is to increase acetabular coverage of the femoral head and to reduce contact pressures by reorienting the acetabulum fragment after PAO. The success of PAO significantly depends on the surgeon’s experience. Previously, we have developed a computer-assisted planning and navigation system for PAO, which allows for not only quantifying the 3D hip morphology for a computer-assisted diagnosis of hip dysplasia but also a virtual PAO surgical planning and simulation. In this paper, based on this previously developed PAO planning and navigation system, we developed a 3D finite element (FE) model to investigate the optimal acetabulum reorientation after PAO. Our experimental results showed that an optimal position of the acetabulum can be achieved that maximizes contact area and at the same time minimizes peak contact pressure in pelvic and femoral cartilages. In conclusion, our computer-assisted planning and navigation system with FE modeling can be a promising tool to determine the optimal PAO planning strategy.

FACTS: Fully Automatic CT Segmentation of a Hip Joint

Extraction of surface models of a hip joint from CT data is a pre-requisite step for computer assisted diagnosis and planning (CADP) of periacetabular osteotomy (PAO). Most of existing CADP systems are based on manual segmentation, which is time-consuming and hard to achieve reproducible results. In this paper, we present a Fully Automatic CT Segmentation (FACTS) approach to simultaneously extract both pelvic and femoral models. Our approach works by combining fast random forest (RF) regression based landmark detection, multi-atlas based segmentation, with articulated statistical shape model (aSSM) based fitting. The two fundamental contributions of our approach are: (1) an improved fast Gaussian transform (IFGT) is used within the RF regression framework for a fast and accurate landmark detection, which then allows for a fully automatic initialization of the multi-atlas based segmentation; and (2) aSSM based fitting is used to preserve hip joint structure and to avoid penetration between the pelvic and femoral models. Taking manual segmentation as the ground truth, we evaluated the present approach on 30 hip CT images (60 hips) with a 6-fold cross validation. When the present approach was compared to manual segmentation, a mean segmentation accuracy of 0.40, 0.36, and 0.36 mm was found for the pelvis, the left proximal femur, and the right proximal femur, respectively. When the models derived from both segmentations were used to compute the PAO diagnosis parameters, a difference of 2.0 ± 1.5°, 2.1 ± 1.6°, and 3.5 ± 2.3% were found for anteversion, inclination, and acetabular coverage, respectively. The achieved accuracy is regarded as clinically accurate enough for our target applications.

A complete-pelvis segmentation framework for image-free total hip arthroplasty (THA): methodology and clinical study

Background Complete-pelvis segmentation in antero-posterior pelvic radiographs is required to create a patient-specific three-dimensional pelvis model for surgical planning and postoperative assessment in image-free navigation of total hip arthroplasty. Methods A fast and robust framework for accurately segmenting the complete pelvis is presented, consisting of two consecutive modules. In the first module, a three-stage method was developed to delineate the left hemipelvis based on statistical appearance and shape models. To handle complex pelvic structures, anatomy-specific information processing techniques were employed. As the input to the second module, the delineated left hemi-pelvis was then reflected about an estimated symmetry line of the radiograph to initialize the right hemi-pelvis segmentation. The right hemi-pelvis was segmented by the same three-stage method, Results Two experiments conducted on respectively 143 and 40 AP radiographs demonstrated a mean segmentation accuracy of 1.61±0.68 mm. A clinical study to investigate the postoperative assessment of acetabular cup orientations based on the proposed framework revealed an average accuracy of 1.2°±0.9° and 1.6°±1.4° for anteversion and inclination, respectively. Delineation of each radiograph costs less than one minute. Conclusions Despite further validation needed, the preliminary results implied the underlying clinical applicability of the proposed framework for image-free THA.