• BoneFinder

  • BoneFinder

  • Bone shape analyses made easy!

    Automatic – Accurate – Adaptive
    Fast
    GET STARTED
  • Bone shape analyses made easy!

    Automatic – Accurate – Adaptive
    Fast
    GET STARTED

ABOUT

BoneFinder is a fully automatic software tool to find and segment skeletal structures from 2D radiographs by placing a set of points along the bone contour or at key landmark positions. It was originally designed for the segmentation of the hip joint but is now also used for other skeletal structures such as the joints of the hand or the knee joint.

BoneFinder firstly identifies the rough position of the bone in the image and then outlines its contour. To do so, BoneFinder follows a machine-learning approach. This means that it has been trained on lots of examples in order to learn what to look for in an image, and it now uses this acquired knowledge to find and segment similar bones in new unseen images. BoneFinder has been found to lead to very robust and accurate point placements across skeletal application areas.

The resulting point positions can then be used for a range of shape analyses such as building statistical shape models or automatically deriving conventional geometric measurements.

Scientific background: BoneFinder uses a Hough Forest like approach to detect the structure of interest in the image, and then applies Random Forest Regression-Voting in the Constrained Local Model framework to locally refine all point positions. More details on these methods and relevant references can be found in our peer-reviewed journal publications of international standing in IEEE TMI and IEEE PAMI. Patent pending.

BoneFinder was written by Claudia Lindner, Tim Cootes and other members of the Centre for Imaging Sciences at The University of Manchester, UK. Funding for the development of BoneFinder has been received from the Medical Research Council UK and the Engineering and Physical Sciences Research Council UK.

PERFORMANCE RESULTS

We tested BoneFinder on a range of different skeletal structures using two-fold cross-validation experiments.
The results show that BoneFinder achieves state-of-the-art performance across application areas.

We tested the BoneFinder hip module on 839 AP pelvic radiographs (527 females, 312 males) using two-fold cross-validation experiments. The aim was to place 65 dense points along the front-view contour of the proximal femur. We report mean point-to-curve errors (where "mean" refers to averaging the errors over all points per image) as a percentage of the shaft width (based on a subset of calibrated images we estimated the latter to be 37mm).

BoneFinder outperforms alternative matching techniques significantly when starting searching from the mean shape at true pose. It achieves a mean point-to-curve error of less than 0.9mm for 99% of all images (i.e. for 99% of images the error is less than 0.9mm).

BoneFinder is fully automatic and highly accurate. Without any initialisation it achieves a mean point-to-curve error of less than 0.9mm for 99% of all images. This is equally good as a local search started from the mean shape at true pose (see left plot).

Examples of segmentation results of the fully automatic system (sorted by mean point-to-curve error percentiles): a) median – 0.4mm; b) 91.2% – 0.6mm, highest global search error (i.e. the global search had a success rate of 100%); c) 99.0% – 0.9mm; d) maximal overall error – 2.7mm.


Data shown on this page is copyright (c) 2013 IEEE (DOI: 10.1109/TMI.2013.2258030). Personal use of this material is permitted. However, permission to use this material for any other purposes must be obtained from the IEEE by sending a request to pubs-permissions@ieee.org.

We tested the BoneFinder hip module on 756 AP pelvic radiographs, aiming to place 81 dense points along the contour of the proximal femur including the trochanters. We report mean point-to-curve errors (where "mean" refers to averaging the errors over all points per image) as a percentage of the shaft width (based on an average of 37mm).

The fully automatic BoneFinder proximal femur module achieves a mean point-to-curve error of within 1.0mm for 99% of all images (i.e. for 99% of images the error is within 1.0mm).

Segmentation example showing the 95%ile of the fully automatic BoneFinder proximal femur module with a mean point-to-curve error of 0.7mm for this image. Red points define the femoral shaft width as reference length.


Data shown on this page is copyright (c) 2013 Springer-Verlag (DOI: 10.1007/978-3-642-40763-5_23).

We tested the BoneFinder knee module on 500 AP knee radiographs, aiming to place 87 dense points along the contours of the distal femur and the proximal tibia. We report mean point-to-curve errors (where "mean" refers to averaging the errors over all points per image) as a percentage of the tibial plateau width (based on an average of 75mm).

The fully automatic BoneFinder knee module achieves a mean point-to-curve error of less than 1.0mm for 99% of all images (i.e. for 99% of images the error is less than 1.0mm).

Segmentation example showing the 95%ile of the fully automatic BoneFinder knee module with a mean point-to-curve error of 0.7mm for this image. Red points define the tibial plateau width as reference length.


Data shown on this page is copyright (c) 2013 Springer-Verlag (DOI: 10.1007/978-3-642-40763-5_23).

We tested the BoneFinder hand joint module on 564 hand radiographs, aiming to annotate the joints of the hand (fingers and wrist) with 37 points. We report mean point-to-point errors (where "mean" refers to averaging the errors over all points per image) as a percentage of the wrist width (based on an average of 50mm).

The fully automatic BoneFinder hand joint module significantly outperforms alternative matching techniques, achieving a mean point-to-point error of within 1.1mm for 99% of all images (i.e. for 99% of images the error is within 1.1mm).

Segmentation example showing the 95%ile of the fully automatic BoneFinder hand joint module with a mean point-to-point error of 0.8mm for this image. Red points define the wrist width as reference length.


Data shown on this page is copyright (c) 2015 IEEE (DOI: 10.1109/TPAMI.2014.2382106). Personal use of this material is permitted. However, permission to use this material for any other purposes must be obtained from the IEEE by sending a request to pubs-permissions@ieee.org.

PUBLICATIONS

  • Open Access
    Fully Automatic System for Accurate Localisation and Analysis of Cephalometric Landmarks in Lateral Cephalograms
    C. Lindner, C.-W. Wang, C.-T. Huang, C.-H. Li, S.-W. Chang and T. Cootes. Scientific Reports, Vol. 6, Art. Nr. 33581, 2016. DOI: 10.1038/srep33581
  • Open Access
    Multi-point Regression Voting for Shape Model Matching
    P.A. Bromiley, C. Lindner, J. Thomson, M. Wrigley and T.F. Cootes. 20th International Conference On Medical Imaging Understanding and Analysis (MIUA) 2016, Loughborough, UK. Elsevier Procedia Computer Science, Vol. 90, pages 48-53, 2016. DOI: 10.1016/j.procs.2016.07.009
  • Open Access
    (abstract)
    Fully automated radiographic knee shape analysis of the OAI dataset: Is knee shape asymmetry an early indicator of unilateral knee OA?
    C. Lindner and T.F. Cootes. 9th International Workshop on Osteoarthritis Imaging (IWOAI) 2016, Oulu, Finland. Workshop Proceedings, page 57, 2016.
  • Open Access
    A benchmark for comparison of dental radiography analysis algorithms
    C.-W. Wang, C.-T. Huang, J.-H. Lee, C.-H. Li, S.-W. Chang, M.-J. Siao, T.-M. Lai, B. Ibragimov, T. Vrtovec, O. Ronneberger, P. Fischer, T.F. Cootes and C. Lindner. Medical Image Analysis, Vol. 31, pages 63-76, 2016. DOI: 10.1016/j.media.2016.02.004
  • Open Access
    (abstract)
    BoneFinder: Automated Bone Shape and Appearance Analysis in 2D Radiographs
    C. Lindner. Invited talk at Berliner Colloquium für Wissenschaftliche Visualisierung, Konrad-Zuse-Zentrum für Informationstechnik Berlin (ZIB), Berlin, Germany. November, 2015.
  • Open Access
    (accepted author version)
    Learning-based Shape Model Matching: Training Accurate Models with Minimal Manual Input
    C. Lindner, J. Thomson, The arcOGEN Consortium and T.F. Cootes. 18th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2015, Part III), Munich, Germany. Springer-Verlag Lecture Notes in Computer Science 9351, pages 580-587, 2015. DOI: 10.1007/978-3-319-24574-4_69
  • Open Access
    Investigation of Association between Hip Osteoarthritis Susceptibility Loci and Radiographic Proximal Femur Shape
    C. Lindner, S. Thiagarajah, J.M. Wilkinson, K. Panoutsopoulou, A.G. Day-Williams, The arcOGEN Consortium, T.F. Cootes and G.A. Wallis. Arthritis & Rheumatology, Vol. 67, No. 8, pages 2076-2084, 2015. DOI: 10.1002/art.39186
  • Open Access
    (accepted author version)
    Fully Automatic Cephalometric Evaluation using Random Forest Regression-Voting
    C. Lindner and T.F. Cootes. IEEE International Symposium on Biomedical Imaging (ISBI) 2015 – Grand Challenges in Dental X-ray Image Analysis – Automated Detection and Analysis for Diagnosis in Cephalometric X-ray Image, 2015.
  • Open Access
    (accepted author version)
    Robust and Accurate Shape Model Matching using Random Forest Regression-Voting
    C. Lindner, P.A. Bromiley, M.C. Ionita and T.F. Cootes. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 37, No. 9, pages 1862-1874, 2015. DOI: 10.1109/TPAMI.2014.2382106
  • Open Access
    (accepted author version)
    Fully Automatic Cephalometric Evaluation using Random Forest Regression-Voting
    S.A. Adeshina, C. Lindner and T.F. Cootes. 11th International Conference on Electronics, Computer and Computation (ICECCO 2014), Abuja, Nigeria. DOI: 10.1109/ICECCO.2014.6997559
  • Open Access
    (accepted author version)
    Automatic Extraction of Hand-Bone Shapes using Random Forest Regression-Voting in the Constrained Local Model Framework
    C. Lindner and T.F. Cootes. 1st International Symposium on Statistical Shape Models (SHAPE 2014), Delmont, Switzerland. Congress Handbook, page 17, 2014.
  • Open Access
    Increasing Shape Modelling Accuracy by Adjusting for Subject Positioning: An Application to the Analysis of Radiographic Proximal Femur Symmetry using Data from the Osteoarthritis Initiative
    C. Lindner, G.A. Wallis and T.F. Cootes. Bone, Vol. 61, pages 64-70, 2014. DOI: 10.1016/j.bone.2014.01.003
  • Open Access
    (accepted author version)
    Development of a Fully Automatic Shape Model Matching (FASMM) System to Derive Statistical Shape Models from Radiographs: Application to the Accurate Capture and Global Representation of Proximal Femur Shape
    C. Lindner, S. Thiagarajah, J.M. Wilkinson, arcOGEN Consortium, G.A. Wallis and T.F. Cootes. Osteoarthritis and Cartilage, Vol. 21, No. 10, pages 1537-1544, 2013. DOI: 10.1016/j.joca.2013.08.008
  • Open Access
    (accepted author version)
    Analysis of Proximal Femur Symmetry using Statistical Shape Models based on Data from the Osteoarthritis Initiative
    C. Lindner, G.A. Wallis and T.F. Cootes. Bone Research Society & British Orthopaedic Research Society Joint Annual Meeting 2013, Oxford. The Bone and Joint Journal.
  • Open Access
    (accepted author version)
    Accurate Bone Segmentation in 2D Radiographs using Fully Automatic Shape Model Matching based on Regression-Voting
    C. Lindner, S. Thiagarajah, J.M. Wilkinson, arcOGEN Consortium, G.A. Wallis and T.F. Cootes. 16th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2013, Part II), Nagoya, Japan. Springer-Verlag Lecture Notes in Computer Science 8150, pages 181-189, 2013. DOI: 10.1007/978-3-642-40763-5_23
  • Open Access
    (accepted author version)
    Fully Automatic Segmentation of the Proximal Femur using Random Forest Regression Voting
    C. Lindner, S. Thiagarajah, J.M. Wilkinson, arcOGEN Consortium, G.A. Wallis and T.F. Cootes. IEEE Transactions on Medical Imaging, Vol. 32, No. 8, pages 1462-1472, 2013. DOI: 10.1109/TMI.2013.2258030
  • Open Access
    (abstract)
    Fully Automatic System to Accurately Segment the Proximal Femur in Anteroposterior Pelvic Radiographs
    C. Lindner, S. Thiagarajah, J.M. Wilkinson, arcOGEN Consortium, G.A. Wallis and T.F. Cootes. UK Radiological Congress (UKRC) 2013, Liverpool, UK. Congress Handbook, page 56, 2013.
  • Open Access
    (abstract)
    Fully Automatic Proximal Femur Segmentation in Pelvic Radiographs using Random Forest Regression Voting
    C. Lindner. Invited talk at the Rank Prize Funds Symposium on Medical Imaging Meets Computer Vision, Grasmere, UK. March, 2013.
  • Open Access
    (accepted author version)
    Robust and Accurate Shape Model Fitting using Random Forest Regression Voting
    T.F. Cootes, M. Ionita, C. Lindner and P. Sauer. 12th European Conference on Computer Vision (ECCV 2012, Part VII), Florence, Italy. Springer-Verlag Lecture Notes in Computer Science 7578, pages 278-291, 2012. DOI: 10.1007/978-3-642-33786-4_21
  • Open Access
    (accepted author version)
    Accurate Fully Automatic Femur Segmentation in Pelvic Radiographs using Regression Voting
    C. Lindner, S. Thiagarajah, J.M. Wilkinson, arcOGEN Consortium, G.A. Wallis and T.F. Cootes. 15th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2012, Part III), Nice, France. Springer-Verlag Lecture Notes in Computer Science 7512, pages 353-360, 2012. DOI: 10.1007/978-3-642-33454-2_44

DOWNLOAD

BoneFinder


BoneFinder is freely available for non-commercial research purposes.

The software and manual can be downloaded from www.click2go.umip.com/i/software/Bonefinder.html.

Please contact us if you would be interested in the commercial use of the technology.

By default, the BoneFinder research licence comes with the proximal femur model1 as shown on the left. If you are interested in detecting and analysing other skeletal structures please do not hesitate to contact us. We are also working on increasing the number of standard models that will be provided together with the BoneFinder research licence.

Please cite the following paper when publishing anything resulting from the usage of BoneFinder:
C. Lindner, S. Thiagarajah, J.M. Wilkinson, The arcOGEN Consortium, G.A. Wallis and T.F. Cootes. "Fully Automatic Segmentation of the Proximal Femur using Random Forest Regression Voting", IEEE Transactions on Medical Imaging, Vol. 32, No. 8, pp. 1462-1472, 2013. DOI: 10.1109/TMI.2013.2258030

1Note that this a front-view proximal femur model that excludes the lesser and greater trochanters and approximates the superior lateral edge from an anterior perspective.

BoneFinder Markup-Tool


The BoneFinder Markup-Tool is a convenient tool to manually generate point placements. These could then be used for quantitative shape analyses and/or to train a fully automatic structure-specific BoneFinder model.

The tool includes guided-annotation which allows the display of the name, description and image of the next point to be placed.

The first version of the BoneFinder Markup-Tool will be available for download soon. Watch this space.

CONTACT

If you are interested in BoneFinder and its capabilities please do not hesitate to get in touch by emailing claudia.lindner@manchester.ac.uk
www.manchester.ac.uk/research/claudia.lindner