Lower jawbone data generation for deep learning tools under MeVisLab

In this contribution, the preparation of data for training deep learning networks that are used to segment the lower jawbone in computed tomography (CT) images is proposed. To train a neural network, we had initially only ten CT datasets of the head-neck region with a diverse number of image slices from the clinical routine of a maxillofacial surgery department. In these cases, facial surgeons segmented the lower jawbone in each image slice to generate the ground truth for the segmentation task. Since the number of present images was deemed insufficient to train a deep neural network efficiently, the data was augmented with geometric transformations and added noise. Flipping, rotating and scaling images as well as the addition of various noise types (uniform, Gaussian and salt-and-pepper) were connected within a global macro module under MeVisLab. Our macro module can prepare the data for general deep learning data in an automatic and flexible way. Augmentation methods for segmentation tasks can easily be incorporated.

[1]  Christopher Nimsky,et al.  Template-Cut: A Pattern-Based Segmentation Paradigm , 2012, Scientific Reports.

[2]  Christopher Nimsky,et al.  Robust Detection and Segmentation for Diagnosis of Vertebral Diseases Using Routine MR Images , 2014, Comput. Graph. Forum.

[3]  T. Kapur,et al.  Fiber Tractography Based on Diffusion Tensor Imaging Compared with High-angular-resolution Diffusion Imaging with Compressed Sensing: Initial Experience , 2022 .

[4]  Andrew Zisserman,et al.  Return of the Devil in the Details: Delving Deep into Convolutional Nets , 2014, BMVC.

[5]  Christopher Nimsky,et al.  Cube-Cut: Vertebral Body Segmentation in MRI-Data through Cubic-Shaped Divergences , 2014, PloS one.

[6]  Dieter Schmalstieg,et al.  Computer-Aided Planning of Cranial 3D Implants , 2016 .

[7]  James V. Miller,et al.  GBM Volumetry using the 3D Slicer Medical Image Computing Platform , 2013, Scientific Reports.

[8]  Jan Egger,et al.  Development of an open source software module for enhanced visualization during MR-guided interstitial gynecologic brachytherapy , 2014, SpringerPlus.

[9]  Christopher Nimsky,et al.  Boundary estimation of fiber bundles derived from diffusion tensor images , 2010, International Journal of Computer Assisted Radiology and Surgery.

[10]  Dieter Schmalstieg,et al.  Interactive Volumetry Of Liver Ablation Zones , 2015, Scientific Reports.

[11]  Bülent Sankur,et al.  Survey over image thresholding techniques and quantitative performance evaluation , 2004, J. Electronic Imaging.

[12]  L. Joshua Leon,et al.  Watershed-Based Segmentation and Region Merging , 2000, Comput. Vis. Image Underst..

[13]  Christopher Nimsky,et al.  Pituitary Adenoma Segmentation , 2011, ArXiv.

[14]  Bernd Freisleben,et al.  A Fast Vessel Centerline Extraction Algorithm for Catheter Simulation , 2007, Twentieth IEEE International Symposium on Computer-Based Medical Systems (CBMS'07).

[15]  Patrice Y. Simard,et al.  Best practices for convolutional neural networks applied to visual document analysis , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[16]  Jan Egger,et al.  Image-guided therapy system for interstitial gynecologic brachytherapy in a multimodality operating suite , 2013, SpringerPlus.

[17]  Bernd Freisleben,et al.  Detection and visualization of endoleaks in CT data for monitoring of thoracic and abdominal aortic aneurysm stents , 2016, SPIE Medical Imaging.

[18]  Christopher Nimsky,et al.  Interactive-cut: Real-time feedback segmentation for translational research , 2014, Comput. Medical Imaging Graph..

[19]  Dieter Schmalstieg,et al.  Computer-aided planning and reconstruction of cranial 3D implants. , 2016, Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference.

[20]  Bernd Freisleben,et al.  Aorta Segmentation for Stent Simulation , 2011, ArXiv.

[21]  Ravi Kiran Sarvadevabhatla,et al.  An Introduction to Deep Convolutional Neural Nets for Computer Vision , 2017, Deep Learning for Medical Image Analysis.

[22]  J. Egger,et al.  Fast self-collision detection and simulation of bifurcated stents to treat abdominal aortic aneurysms (AAA) , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[23]  Christopher Nimsky,et al.  A Fast and Robust Graph-Based Approach for Boundary Estimation of Fiber Bundles Relying on Fractional Anisotropy Maps , 2010, 2010 20th International Conference on Pattern Recognition.

[24]  H. Handels.  Computergestützte Diagnostik und Therapie , 2009 .

[25]  Antje Baer,et al.  Handbook Of Medical Image Processing And Analysis , 2016 .

[26]  Christopher Nimsky,et al.  Integration of the OpenIGTLink Network Protocol for image‐guided therapy with the medical platform MeVisLab , 2012, The international journal of medical robotics + computer assisted surgery : MRCAS.

[27]  Heung-Il Suk,et al.  An Introduction to Neural Networks and Deep Learning , 2017, Deep Learning for Medical Image Analysis.

[28]  Christopher Nimsky,et al.  Pituitary Adenoma Volumetry with 3D Slicer , 2012, PloS one.

[29]  Christopher Nimsky,et al.  A Flexible Semi-Automatic Approach for Glioblastoma multiforme Segmentation , 2011, ArXiv.

[30]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[31]  Jan Egger,et al.  Aortic volume as an indicator of disease progression in patients with untreated infrarenal abdominal aneurysm. , 2012, European journal of radiology.

[32]  Christopher Nimsky,et al.  A Medical Software System for Volumetric Analysis of Cerebral Pathologies in Magnetic Resonance Imaging (MRI) Data , 2012, Journal of Medical Systems.

[33]  Jan Egger,et al.  PCG-Cut: Graph Driven Segmentation of the Prostate Central Gland , 2013, PloS one.

[34]  Bernd Freisleben,et al.  Graph-based Tracking Method for Aortic Thrombus Segmentation , 2009 .

[35]  Dieter Schmalstieg,et al.  HTC Vive MeVisLab integration via OpenVR for medical applications , 2017, PloS one.

[36]  Jan Egger,et al.  Refinement-Cut: User-Guided Segmentation Algorithm for Translational Science , 2014, Scientific Reports.

[37]  Christopher Nimsky,et al.  Nugget-Cut: A Segmentation Scheme for Spherically- and Elliptically-Shaped 3D Objects , 2010, DAGM-Symposium.

[38]  Dieter Schmalstieg,et al.  Exploit 18F-FDG enhanced urinary bladder in PET data for deep learning ground truth generation in CT scans , 2018, Medical Imaging.

[39]  Dieter Schmalstieg,et al.  Lower jawbone data generation for deep learning tools under MeVisLab , 2018, Medical Imaging.

[40]  Christopher Nimsky,et al.  Segmentation of Vertebral Bodies in MR Images , 2012, VMV.

[41]  Bernd Freisleben,et al.  PREOPERATIVE MEASUREMENT OF ANEURYSMS AND STENOSIS AND STENTSIMULATION FOR ENDOVASCULAR TREATMENT , 2007, 2007 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[42]  Christopher Nimsky,et al.  Modeling and visualization techniques for virtual stenting of aneurysms and stenoses , 2012, Comput. Medical Imaging Graph..

[43]  Bernd Freisleben,et al.  Simulation of bifurcated stent grafts to treat abdominal aortic aneurysms (AAA) , 2007, SPIE Medical Imaging.

[44]  Christopher Nimsky,et al.  Manual Refinement System for Graph-Based Segmentation Results in the Medical Domain , 2012, Journal of Medical Systems.