Deep Learning-Based Reconstruction of Interventional Tools from Four X-Ray Projections for Tomographic Interventional Guidance

Image guidance for minimally invasive interventions is usually performed by acquiring fluoroscopic images using a C-arm system. However, the projective data provide only limited information about the spatial structure and position of interventional tools such as stents, guide wires or coils. In this work we propose a deep learning-based pipeline for real-time tomographic (four-dimensional) interventional guidance at acceptable dose levels. In the first step, interventional tools are extracted from four cone-beam CT projections using a deep convolutional neural network (CNN). These projections are then reconstructed and fed into a second CNN, which maps this highly undersampled reconstruction to a segmentation of the interventional tools. Our pipeline is capable of reconstructing interventional tools from only four x-ray projections without the need for a patient prior with very high accuracy. Therefore, the proposed approach is capable of overcoming the drawbacks of today's interventional guidance and could enable the development of new minimally invasive radiological interventions by providing full spatiotemporal information about the interventional tools.

[1]  Loïc Le Folgoc,et al.  Attention U-Net: Learning Where to Look for the Pancreas , 2018, ArXiv.

[2]  Robert A. Kruger,et al.  Computerized Fluoroscopy in Real Time for Noninvasive Visualization of the Cardiovascular System , 1979 .

[3]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[4]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[5]  Emmanuel J. Candès,et al.  Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information , 2004, IEEE Transactions on Information Theory.

[6]  Barbara Flach,et al.  Deformable 3D–2D registration for CT and its application to low dose tomographic fluoroscopy , 2014, Physics in medicine and biology.

[7]  A R Cowen,et al.  The design and imaging characteristics of dynamic, solid-state, flat-panel x-ray image detectors for digital fluoroscopy and fluorography. , 2008, Clinical radiology.

[8]  A. Rényi,et al.  On projections of probability distributions , 1952 .

[9]  A. Coon,et al.  Cerebral aneurysm treatment: modern neurovascular techniques , 2016, Stroke and vascular neurology.

[10]  Michael Knaup,et al.  Deep learning-aided CBCT image reconstruction of interventional material from four x-ray projections , 2020 .

[11]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[12]  C. Walsh,et al.  CT-Guided Interventions: Current Practice and Future Directions , 2014 .

[13]  Peter Sandercock,et al.  For Personal Use. Only Reproduce with Permission from the Lancet Publishing Group , 2022 .

[14]  Donald Ervin Knuth,et al.  The Art of Computer Programming , 1968 .

[15]  M Kachelriess,et al.  Constrained reconstructions for 4D intervention guidance. , 2013, Physics in medicine and biology.

[16]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[17]  Peter Gritzmann,et al.  Discrete Tomography: Determination of Finite Sets by X-Rays , 1995, Universität Trier, Mathematik/Informatik, Forschungsbericht.

[18]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[19]  Jonathan Tompson,et al.  Efficient object localization using Convolutional Networks , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Yoshua Bengio,et al.  Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation , 2014, EMNLP.

[21]  L. Feldkamp,et al.  Practical cone-beam algorithm , 1984 .

[22]  B. Wood,et al.  CT fluoroscopy shielding: decreases in scattered radiation for the patient and operator. , 2006, Journal of vascular and interventional radiology : JVIR.

[23]  Marc Kachelrieß,et al.  Learned digital subtraction angiography (Deep DSA): method and application to lower extremities , 2019 .

[24]  Wolfhard Semmler,et al.  Real-time X-ray-based 4D image guidance of minimally invasive interventions , 2013, European Radiology.

[25]  S. Frick,et al.  Compressed Sensing , 2014, Computer Vision, A Reference Guide.

[26]  Gabriele Bianchi,et al.  Reconstructing plane sets from projections , 1990, Discret. Comput. Geom..

[27]  M. P. Judkins,et al.  Transluminal Treatment of Arteriosclerotic Obstruction: Description of a New Technic and a Preliminary Report of Its Application , 1964, Circulation.

[28]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[29]  E. Bücheler,et al.  [Catheter embolisation of renal tumours (author's transl)]. , 1976, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[30]  B. Welford Note on a Method for Calculating Corrected Sums of Squares and Products , 1962 .

[31]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[32]  J R Hesselink,et al.  Digital subtraction angiography. , 1984, Neurologic clinics.

[33]  R. Owen Embolization of musculoskeletal bone tumors. , 2010, Seminars in interventional radiology.

[34]  R. Spetzler,et al.  The Barrow Ruptured Aneurysm Trial: 6-year results. , 2015, Journal of neurosurgery.

[35]  Giampiero Beluffi Interventional radiology procedures in biopsy and drainage , 2013, La radiologia medica.

[36]  M. P. Judkins,et al.  Transluminal Treatment of Arteriosclerotic Obstruction , 1989 .

[37]  A. Reddy,et al.  C-arm fluoroscopic cone beam CT for guidance of minimally invasive spine interventions. , 2010, Pain physician.

[38]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[39]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[40]  D. Pelz,et al.  Digital subtraction angiography: current clinical applications. , 1985, Stroke.

[41]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.