Studies of the criteria for determining optimal location of medial patellofemoral ligament attachment sites.

Identifying appropriate attachment sites is important in the planning of medial patellofemoral ligament (MPFL) reconstruction. Two criteria are advanced to describe normal MPFL function, namely isometric criterion and desired pattern criterion. Subsequently, computational methods have applied these criteria to determine optimal attachment sites. So far, there is no study that compares the outcomes of these two criteria. For five subjects' 3D models of the patella and femur, three patellar sites and many femoral sites were identified as pairs of candidate attachment sites. For each patellar site, the criteria were applied to identify the matching femoral sites that satisfy them. The matching femoral site with the smallest length change was identified as the optimal femoral site. The desired pattern criterion finds fewer matching sites compared to the isometric criterion. In contrast, the isometric criterion can always find matching sites. The optimal femoral sites obtained vary significantly across different subjects. For most subjects, the optimal sites obtained using the isometric criterion are closer to known anatomical sites than those obtained using the desired pattern criterion. This study reaffirms that MPFL reconstruction is subject specific. The isometric criterion may be more reliable than the desired pattern criterion for determining optimal attachment sites. Graphical Abstract. Highlight of the paper. The location of the patella site significantly affects the location of the optimal femoral site. The isometric criterion option 1, with length at 0° regarded as MPFL's natural length, may be more reliable than other criteria or options for the planning of MPFL surgery because the optimal sites that it finds are closest to known anatomical sites.ᅟ.

Flip-avoiding interpolating surface registration for skull reconstruction.

Skull reconstruction is an important and challenging task in craniofacial surgery planning, forensic investigation and anthropological studies. Existing methods typically reconstruct approximating surfaces that regard corresponding points on the target skull as soft constraints, thus incurring non-zero error even for non-defective parts and high overall reconstruction error. This paper proposes a novel geometric reconstruction method that non-rigidly registers an interpolating reference surface that regards corresponding target points as hard constraints, thus achieving low reconstruction error. To overcome the shortcoming of interpolating a surface, a flip-avoiding method is used to detect and exclude conflicting hard constraints that would otherwise cause surface patches to flip and self-intersect. Comprehensive test results show that our method is more accurate and robust than existing skull reconstruction methods. By incorporating symmetry constraints, it can produce more symmetric and normal results than other methods in reconstructing defective skulls with a large number of defects. It is robust against severe outliers such as radiation artifacts in computed tomography due to dental implants. In addition, test results also show that our method outperforms thin-plate spline for model resampling, which enables the active shape model to yield more accurate reconstruction results. As the reconstruction accuracy of defective parts varies with the use of different reference models, we also study the implication of reference model selection for skull reconstruction.

Intravenous Thrombolysis for Acute Ischemic Stroke due to Cervical Internal Carotid Artery Occlusion.

BACKGROUND: Internal carotid artery (ICA) occlusions are poorly responsive to intravenous thrombolysis with tissue plasminogen activator (IV-tPA) in acute ischemic stroke (AIS). Most study populations have combined intracranial and extracranial ICA occlusions for analysis; few have studied purely cervical ICA occlusions. We evaluated AIS patients with acute cervical ICA occlusion treated with IV-tPA to identify predictors of outcomes. METHODS: We studied 550 consecutive patients with AIS who received IV-tPA and identified 100 with pure acute cervical ICA occlusion. We evaluated the associations of vascular risk factors, National Institutes of Health Stroke Scale (NIHSS) score, and leptomeningeal collateral vessel status via 3 different grading systems, with functional recovery at 90 days, mortality, recanalization of the primary occlusion, and symptomatic intracranial hemorrhage (SICH). Modified Rankin Scale score 0-1 was defined as an excellent outcome. RESULTS: The 100 patients had mean age of 67.8 (range 32-96) and median NIHSS score of 19 (range 4-33). Excellent outcomes were observed in 27% of the patients, SICH in 8%, and mortality in 21%. Up to 54% of the patients achieved recanalization at 24 hours. On ordinal regression, good collaterals showed a significant shift in favorable outcomes by Maas, Tan, or ASPECTS collateral grading systems. On multivariate analysis, good collaterals also showed reduced mortality (OR .721, 95% CI .588-.888, P = .002) and a trend to less SICH (OR .81, 95% CI .65-1.007, P = .058). Interestingly, faster treatment was also associated with favorable functional recovery (OR 1.028 per minute, 95% CI 1.010-1.047, P = .001). CONCLUSIONS: Improved outcomes are seen in patients with early acute cervical ICA occlusion and better collateral circulation. This could be a valuable biomarker for decision making.

Liver tumour segmentation using contrast-enhanced multi-detector CT data: performance benchmarking of three semiautomated methods.

OBJECTIVE: Automatic tumour segmentation and volumetry is useful in cancer staging and treatment outcome assessment. This paper presents a performance benchmarking study on liver tumour segmentation for three semiautomatic algorithms: 2D region growing with knowledge-based constraints (A1), 2D voxel classification with propagational learning (A2) and Bayesian rule-based 3D region growing (A3). METHODS: CT data from 30 patients were studied, and 47 liver tumours were isolated and manually segmented by experts to obtain the reference standard. Four datasets with ten tumours were used for algorithm training and the remaining 37 tumours for testing. Three evaluation metrics, relative absolute volume difference (RAVD), volumetric overlap error (VOE) and average symmetric surface distance (ASSD), were computed based on computerised and reference segmentations. RESULTS: A1, A2 and A3 obtained mean/median RAVD scores of 17.93/10.53%, 17.92/9.61% and 34.74/28.75%, mean/median VOEs of 30.47/26.79%, 25.70/22.64% and 39.95/38.54%, and mean/median ASSDs of 2.05/1.41 mm, 1.57/1.15 mm and 4.12/3.41 mm, respectively. For each metric, we obtained significantly lower values of A1 and A2 than A3 (P < 0.01), suggesting that A1 and A2 outperformed A3. CONCLUSIONS: Compared with the reference standard, the overall performance of A1 and A2 is promising. Further development and validation is necessary before reliable tumour segmentation and volumetry can be widely used clinically.

A cognitive virtual microscopic framework for knowledge-based exploration of large microscopic images in breast cancer histopathology.

Histopathological examination is a powerful method for prognosis of major diseases such as breast cancer. Analysis of medical images largely remains the work of human experts. Current virtual microscope systems are mainly an emulation of real microscopes with annotation and some image analysis capabilities. However, the lack of effective knowledge management prevents such systems from being computer-aided prognosis platforms. The cognitive virtual microscopic framework, through an extended modeling and use of medical knowledge, has the capacity to analyse histopathological images and to perform grading of breast cancer, providing pathologists with a robust and traceable second opinion.

Modeling torsion of blood vessels in surgical simulation and planning.

This paper proposes a hybrid approach for modeling torsion of blood vessels that undergo deformation and joining. The proposed model takes 3D mesh of the blood vessel as input. It first fits a generalized cylinder to extract the blood vessel's medial axis. Then, it uses rotation minimizing frame as a reference to model and measure the torsion of blood vessel after deformation. In general, the proposed approach can incorporate any kind of deformation algorithms. In our experiments, differential geometry method is used as an example. The test results show that our algorithm can correctly and effectively evaluate the amount of torsion caused by blood vessel deformation. In addition, it can also determine the configuration of the blood vessel with minimum torsion.

Predictive surgical simulation of aorta reconstruction in cardiac surgery.

This paper proposes a method for performing predictive simulation of complex cardiac surgery. It computes complex surgical results given a small amount of user inputs. In this way, the surgeon can easily explore various surgical options without having to go through all the detailed steps of the surgical procedure. Test results, using aorta reconstruction as an application example, show that the proposed method can generate realistic simulation results given different kinds of user inputs, thus demonstrating the feasibility of the approach.

Predictive simulation of bidirectional Glenn shunt using a hybrid blood vessel model.

This paper proposes a method for performing predictive simulation of cardiac surgery. It applies a hybrid approach to model the deformation of blood vessels. The hybrid blood vessel model consists of a reference Cosserat rod and a surface mesh. The reference Cosserat rod models the blood vessel's global bending, stretching, twisting and shearing in a physically correct manner, and the surface mesh models the surface details of the blood vessel. In this way, the deformation of blood vessels can be computed efficiently and accurately. Our predictive simulation system can produce complex surgical results given a small amount of user inputs. It allows the surgeon to easily explore various surgical options and evaluate them. Tests of the system using bidirectional Glenn shunt (BDG) as an application example show that the results produc by the system are similar to real surgical results.

Automatic breast cancer grading of histopathological images.

Breast cancer grading of histopathological images is the standard clinical practice for the diagnosis and prognosis of breast cancer development. In a large hospital, a pathologist typically handles 100 grading cases per day, each consisting of about 2000 image frames. It is, therefore, a very tedious and time-consuming task. This paper proposes a method for automatic computer grading to assist pathologists by providing second opinions and reducing their workload. It combines the three criteria in the Nottingham scoring system using a multi-resolution approach. To our best knowledge, there is no existing work that provide complete grading according to the Nottingham criteria.

Extraction of rules from artificial neural networks for nonlinear regression.

Neural networks (NNs) have been successfully applied to solve a variety of application problems including classification and function approximation. They are especially useful as function approximators because they do not require prior knowledge of the input data distribution and they have been shown to be universal approximators. In many applications, it is desirable to extract knowledge that can explain how Me problems are solved by the networks. Most existing approaches have focused on extracting symbolic rules for classification. Few methods have been devised to extract rules from trained NNs for regression. This article presents an approach for extracting rules from trained NNs for regression. Each rule in the extracted rule set corresponds to a subregion of the input space and a linear function involving the relevant input attributes of the data approximates the network output for all data samples in this subregion. Extensive experimental results on 32 benchmark data sets demonstrate the effectiveness of the proposed approach in generating accurate regression rules.

Explanation of the "virtual input" phenomenon.

We write this letter to comment on the "virtual input" phenomenon reported by Thaler (Neural Networks, 8(1) (1995) 55-65). The author attributed the phenomenon to the network's ability to perform pattern classification and completion, and reported that pruning probability affects the number of virtual inputs observed. Our independent study of Thaler's results, however, reveals a simpler explanation of the "virtual input" phenomenon.