AUTOMATED MEASUREMENT OF FOOT DEFORMITIES: FLATFOOT, HIGH ARCH, CALCANEAL FRACTURE
DOI:
https://doi.org/10.5566/ias.1980Keywords:
calcaneal fracture, computer aided diagnostic, extremely randomized trees, flatfoot, high arch, x-raysAbstract
Radiographic measurements of foot deformities are used to determine, among other things, such conditions as flatfoot, high arch, or calcaneal fracture. Those measurements are achieved by estimating four angles. Manual assessment of those angles is time-consuming not to mention inevitable errors of such approximation. To the best of the authors knowledge, currently there is no research focusing on finding those four angles. In this paper an algorithm for automatic assessment of those angles, based on extremely randomized trees, is being proposed. Moreover this diagnostic assisting system was intended to be as generic as possible and could be applied, to some degree, to other similar problems. To demonstrate usefulness of this method, correlations of automated measurements with manual ones against correlations of manual measurements with manual ones are being compared. The significance level for manual-manual measurements comparison is less than 0.001 in case of all four angles. The significance level for automated-manual measurements comparison is also less than 0.001 in all cases. The results show that the search for the aforementioned angles can be automated. Even with the use of a generic algorithm a high degree of precision can be achieved, allowing for a more efficient diagnosis.
References
Altman DG, Bland JM (1983). Measurement in medicine: The analysis of method comparison studies. J of the R Stat Soc ser D The Statistician 32:307–17.
Arif SMMRA, Knapp K, Slabaugh G (2018). Fully automatic cervical vertebrae segmentation framework for x-ray images. Comput Methods and Programs in Biomed 157:95 – 111. [doi: 10.1016/j.cmpb.2018.01.006].
Arslan G, Yirgin IK, Tasguzen A (2014). A measuremental approach to calcaneal fractures. Eur J of Trauma and Emerg Surg 40:593–599. [doi: 10.1007/s00068-013-0359-2].
Briscoe E, Feldman J (2011). Conceptual complexity and the bias/variance tradeoff. Cogn 118:2 – 16. [doi: 10.1016/j.cognition.2010.10.004].
Donner R, Menze BH, Bischof H, Langs G (2013). Global localization of 3d anatomical structures by prefiltered hough forests and discrete optimization. Med Image Anal 17:1304 – 1314. [doi: 10.1016/j.media.2013.02.004].
Geurts P, Ernst D, Wehenkel L (2006). Extremely randomized trees. Machine Learning 63:3–42. [doi: 10.1007/s10994-006-6226-1].
Hunter JD (2007). Matplotlib: A 2d graphics environment. Comput in Sci Eng 9:90–5. [doi: 10.1109/MCSE.2007.55].
Jones E, Oliphant T, Peterson P, et al. (2011). SciPy: Open source scientific tools for Python. http://www.scipy.org/. [Accessed 24 May 2018].
Kao EF, Lu CY, Wang CY, Yeh WC, Hsia PK (2018). Fully automated determination of arch angle on weight-bearing foot radiograph. Comput Methods and Programs in Biomed 154:79 – 88. [doi: 10.1016/j.cmpb.2017.11.009].
Khoshhal K, El Fouhil A, Al-Nakshabandi N, M Zamzam M, A Al-Boukai A, Zamzami M (2005). Böhler’s and gissane’s angles of the calcaneus in the saudi population. Saudi Med J 25:1967–70.
Lepetit V, Fua P (2006). Keypoint recognition using randomized trees. IEEE Trans on Pattern Anal and Mach Intell 28:1465–79. [doi: 10.1109/TPAMI.2006.188].
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011). Scikitlearn: Machine learning in Python. J of Mach Learn Res 12:2825–30.
Pinto A, Pereira S, Rasteiro D, Silva CA (2018). Hierarchical brain tumour segmentation using extremely randomized trees. Pattern Recognit 82:105–17. [doi: 10.1016/j.patcog.2018.05.006].
Scalzo F, Hamilton R, Asgari S, Kim S, Hu X (2012). Intracranial hypertension prediction using extremely randomized decision trees. Med Eng and Phys 34:1058–1065. [doi: 10.1016/j.medengphy.2011.11.010].
Soltaninejad M, Yang G, Lambrou T, Allinson N, Jones TL, Barrick TR, Howe FA, Ye X (2017). Automated brain tumour detection and segmentation using superpixelbased extremely randomized trees in flair mri. Int J of Comput Assist Radiol and Surg 12:183–203. [doi: 10.1007/s11548-016-1483-3].
Waldt S, Woertler K (2014). Measurements and Classifications in Musculoskeletal Radiology. New York, NY 10001, USA: Thieme Medical Publishers.
Wojciechowski W, Molka A, Tabor Z (2016). Automated measurement of parameters related to the deformities of lower limbs based on x-rays images. Comput in Biol and Med 70:1 – 11. [doi: 10.1016/j.compbiomed.2015.12.027].
Yang CH, Chou KT, Chung MB, Chuang KS, Huang TC (2015). Automatic detection of calcaneal-fifth metatarsal angle using radiograph: A computeraided diagnosis of flat foot for military new recruits in taiwan. PLOS ONE 10:1–10. [doi: 10.1371/journal.pone.0131387].
Yates B (2009). Merriman’s Assessment of the Lower Limb. New York: Churchill Livingstone.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2019 Image Analysis & Stereology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
