Deep learning-based framework for automatic cranial defect reconstruction and implant modeling.

BACKGROUND AND OBJECTIVE: This article presents a robust, fast, and fully automatic method for personalized cranial defect reconstruction and implant modeling. METHODS: We propose a two-step deep learning-based method using a modified U-Net architecture to perform the defect reconstruction, and a dedicated iterative procedure to improve the implant geometry, followed by an automatic generation of models ready for 3-D printing. We propose a cross-case augmentation based on imperfect image registration combining cases from different datasets. Additional ablation studies compare different augmentation strategies and other state-of-the-art methods. RESULTS: We evaluate the method on three datasets introduced during the AutoImplant 2021 challenge, organized jointly with the MICCAI conference. We perform the quantitative evaluation using the Dice and boundary Dice coefficients, and the Hausdorff distance. The Dice coefficient, boundary Dice coefficient, and the 95th percentile of Hausdorff distance averaged across all test sets, are 0.91, 0.94, and 1.53 mm respectively. We perform an additional qualitative evaluation by 3-D printing and visualization in mixed reality to confirm the implant's usefulness. CONCLUSION: The article proposes a complete pipeline that enables one to create the cranial implant model ready for 3-D printing. The described method is a greatly extended version of the method that scored 1st place in all AutoImplant 2021 challenge tasks. We freely release the source code, which together with the open datasets, makes the results fully reproducible. The automatic reconstruction of cranial defects may enable manufacturing personalized implants in a significantly shorter time, possibly allowing one to perform the 3-D printing process directly during a given intervention. Moreover, we show the usability of the defect reconstruction in a mixed reality that may further reduce the surgery time.

Methods of bronchial tree reconstruction and camera distortion corrections for virtual endoscopic environments.

The use of three-dimensional visualization of anatomical structures in diagnostics and medical training is growing. The main components of virtual respiratory tract environments include reconstruction and simulation algorithms as well as correction methods of endoscope camera distortions in the case of virtually-enhanced navigation systems. Reconstruction methods rely usually on initial computer tomography (CT) image segmentation to trace contours of the tracheobronchial tree, which in turn are used in the visualization process. The main segmentation methods, including relatively simple approaches such as adaptive region-growing algorithms and more complex methods, e.g. hybrid algorithms based on region growing and mathematical morphology methods, are described in this paper. The errors and difficulties in the process of tracheobronchial tree reconstruction depend on the occurrence of distortions during CT image acquisition. They are usually related to the inability to exactly fulfil the sampling theorem's conditions. Other forms of distortions and noise such as additive white Gaussian noise, may also appear. The impact of these distortions on the segmentation and reconstruction may be diminished through the application of appropriately selected image prefiltering, which is also demonstrated in this paper. Methods of surface rendering (ray-casting, ray-tracing techniques) and volume rendering will be shown, with special focus on aspects of hardware and software implementations. Finally, methods of camera distortions correction and simulation are presented. The mathematical camera models, the scope of their applications and types of distortions were have also been indicated.