No abstract

A number of recent publications have made use of the incremental output of stochastic parsers to derive measures of high utility for psycholinguistic modeling, following the work of Hale (2001;. In this paper, we present novel methods for calculating separate lexical and syntactic surprisal measures from a single incremental parser using a lexicalized PCFG. We also present an approximation to entropy measures that would otherwise be intractable to calculate for a grammar of that size. Empirical results demonstrate the utility of our methods in predicting human reading times.

Background: Operation planning in liver surgery depends on the precise understanding of the 3-dimensional (D) relation of the tumor to the intrahepatic vascular trees. To our knowledge, the impact of anatomical 3-D reconstructions on precision in operation planning has not yet been studied. Hypothesis: Three-dimensional reconstruction leads to an improvement of the ability to localize the tumor and an increased precision in operation planning in liver surgery. Design: We developed a new interactive computerbased quantitative 3-D operation planning system for liver surgery, which is being introduced to the clinical routine. To evaluate whether 3-D reconstruction leads to improved operation planning, we conducted a clinical trial. The data sets of 7 virtual patients were presented to a total of 81 surgeons in different levels of training. The tumors had to be assigned to a liver segment and subsequently drawn together with the operation proposal into a given liver model. The precision of the assignment to a liver segment according to Couinaud classification and the operation proposal were measured quantitatively for each surgeon and stratified concerning 2-D and different types of 3-D presentations. Results: The ability of correct tumor assignment to a liver segment was significantly correlated to the level of training (PϽ.05). Compared with 2-D computed tomography scans, 3-D reconstruction leads to a significant increase of precision in tumor localization by 37%. The target area of the resection proposal was improved by up to 31%. Conclusion: Three-dimensional reconstruction leads to a significant improvement of tumor localization ability and to an increased precision of operation planning in liver surgery.

Purpose: To develop a head and neck normal structures auto-contouring tool that could be used to automatically detect the errors in auto-contours from a clinically-validated auto-contouring tool. Methods: An auto-contouring tool based on convolutional neural networks (CNN) was developed for 16 normal structures of the head and neck and tested to identify the contour errors from a clinically-validated multi-atlas-based auto-contouring system (MACS). The CT scans and clinical contours from 3495 patients were semi-automatically curated and used to train and validate the CNN-based auto-contouring tool. The final accuracy of the tool was evaluated by calculating the Sørensen-Dice similarity coefficients (DSC) and Hausdorff distances between the automatically generated contours and physician-drawn contours on 174 internal and 24 external CT scans. Lastly, the CNN-based tool was evaluated on 60 patients’ CT scans to investigate the possibility to detect contouring failures. The contouring failures on these patients were classified as either minor or major errors. The criteria to detect contouring errors were determined by analyzing the DSC between the CNN- and MACS-based contours under two independent scenarios: 1. contours with minor error are clinically acceptable and 2. contours with minor errors are clinically unacceptable. Results: The average DSC and Hausdorff distance of our CNN-based tool were 98.4%/1.23cm for brain, 89.1%/0.42cm for eyes, 86.8%/1.28cm for mandible, 86.4%/0.88cm for brainstem, 83.4%/0.71cm for spinal cord, 82.7%/1.37cm for parotids, 80.7%/1.08cm for esophagus, 71.7%/0.39cm for lenses, 68.6%/0.72 for optic nerves, 66.4%/0.46cm for cochleas, and 40.7%/0.96cm for optic chiasm. With the error detection tool, the proportions of the clinically unacceptable MACS contours that were correctly detected were 0.99/0.80 on average except for the optic chiasm, when contours with minor errors are clinically acceptable/unacceptable respectively. The proportions of the clinically acceptable MACS contours that were correctly detected were 0.81/0.60 on average except for the optic chiasm, when contours with minor errors are clinically acceptable/unacceptable respectively. Conclusion: Our CNN-based auto-contouring tool performed well on both the publically-available and the internal datasets. Furthermore, our results show that CNN-based algorithms are able to identify ill-defined contours from a clinically-validated and used multi-atlas-based auto-contouring tool. Therefore, our CNN-based tool can effectively perform automatic verification of MACS contours.

These predicted high-risk CTVs provided close agreement to the ground-truth compared with current interobserver variability. The predicted contours could be implemented clinically, with only minor or no changes.