Abstract
A computer-aided classification system was developed for the assessment of the severity of hip osteoarthritis (OA) . Sixty-four radiographic images of normal and osteoarthritic hips were digitized and enhanced. Employing the Kellgren and Lawrence scale, the hips were grouped by three experienced orthopaedists into three OA-severity categories: Normal, Mild/Moderate and Severe. Utilizing custom-developed software, 64 ROIs corresponding to the radiographic Hip Joint Spaces were manually segmented and novel textural features were generated. These features were used in the design of a two-level classification scheme for characterizing hips as normal or osteoarthritic (1st level) and as of Mild/Moderate or Severe OA (2nd level). At each classification level, an ensemble of three classifiers was implemented. The proposed classification scheme discriminated correctly all normal hips from osteoarthritic hips (100% accuracy), while the discrimination accuracy between Mild/Moderate and Severe osteoarthritic hips was 95.7%. The proposed system could be used as a diagnosis decision-supporting tool.
Similar content being viewed by others
Abbreviations
- OA:
-
Osteoarthritis
- HJS:
-
Hip joint space
- KL:
-
Kellgren and Lawrence
- ROI:
-
Region of interest
- CLAHE:
-
Contrast limited adaptive histogram equalization
- GEN_Image:
-
Gabor energy image
- GLRLM:
-
Grey level run length matrix
- GEMRL:
-
Gabor energy measure run length
- PNN:
-
Probabilistic neural network
- k-NN:
-
k-Nearest–neighbour
- MV:
-
Majority vote
- CV:
-
Coefficient of variation
References
Aigner T, McKenna L (2002) Molecular pathology and pathobiology of osteoarthritic cartilage. CMLS Cell Mol Life Sci 59:5–18
Altman R, Alarcón G, Appelrouth D, et al (1991) The American college of rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis Rheum 34:505–514
Altman RD, Fries JF, Bloch DA, et al (1987) Radiographic assessment of progression in osteoarthritis. Arthritis Rheum 30:1214–1225
Atlamazoglou V, Yova D, Kavantzas N, Loukas S (2001) Texture analysis of fluorescence microscopic images of colonic tissue sections. Med Biol Eng Comput 39:145–151
Barandela R, Sánchez JS, Valdovinos RM (2003) New applications of ensembles of classifiers. Pattern Anal Appl 6:245–256
Bocchi L, Coppini G, De Dominicis R, Valli G (1997) Tissue characterization from X-ray images. Med Eng Phys 19:336–342
Boniatis I, Costaridou L, Cavouras D, Panagiotopoulos E, Panayiotakis G (2006) Quantitative assessment of hip osteoarthritis based on image texture analysis. Br J Radiol 79:232–238
Buckwalter A, Mankin HJ (1997) Instructional course lectures, the American Academy of Orthopaedic Surgeons — Articular Cartilage. Part II: degeneration and osteoarthrosis, repair, regeneration, and transplantation. J Bone Joint Surg Am 79:612–632
Campbell M, Machin D (1996) Medicals statistics, 2nd edn. Wiley Ltd., Chichester
Christodoulou CI, Pattichis CS, Kyriacou E, Pattichis MS, Pantziaris M, Nikolaides A (2005) Texture and morphological analysis of ultrasound images of the carotid plaque for the assessment of stroke. In: Costaridou L (eds) Medical image analysis methods. CRC Press, Taylor and Francis Group, Boca Raton, London, New York, Singapore, pp 87–135
Conrozier T, Tron AM, Balblanc JC, et al (1993) Measurement of the hip joint space using computerized image analysis. Rev Rhum Engl Ed 60:105–111
Daugman JG (1985) Uncertainty relation for resolution in space, spatial frequency, and orientation optimized by two-dimensional visual cortical filters. J Opt Soc Am 2:1160–1169
Efstathopoulos EP, Costaridou L, Kocsis O, Panayiotakis G (2001) A protocol-based evaluation of medical image digitizers. Br J Radiol 74:841–846
Galloway MM (1975) Texture analysis using gray level run lengths. Comput Graph Image Process 4:172–179
Gordon CL, Wu C, Peterfy CG, Duryea J, Klifa C, Genant HK (2001) Automated measurement of radiographic hip joint space width. Med Phys 28:267–277
Grigorescu SE, Petkov N, Kruizinga P (2002) Comparison of textural features based on Gabor filters. IEEE Trans Image Process 11:1160–1167
Ingvarsson T, Hägglund G, Lindberg H, Lohmander LS (2000) Assessment of primary hip osteoarthritis: comparison of radiographic methods using colon radiographs. Ann Rheum Dis 59:650–653
Jain AK, Duin RPW, Jianchang M (2000) Statistical pattern recognition: a review. IEEE Trans Pattern Anal 22:4–37
Kellgren JH, Lawrence JS (1957) Radiological assessment of osteoarthrosis. Ann Rheum Dis 16:494–501
Lilliefors HW (1967) On the Kolmogorov–Smirnov-test for normality with mean and variance unknown. J Am Stat Assoc 62:399–402
Lumiscan 75, system specifications. Lumisys Inc. 1998; http://www.lumisys.com/support/manuals.html.
Martel–Pelletier J, Pelletier J-P (2003) Osteoarthritis: recent developments. Curr Opin Rheumatol 15:613–615
Ory PA (2003) Radiography in the assessment of musculoskeletal conditions. Best Pract Res Clin Rheumatol 17:495–512
Pizer SM, Amburn EOP, Austin JD, Cromartie R, Geselowitz A, Greer T (1987) Adaptive histogram equalization and its variations. CVGIP (Comput Vis Graph Image Process) 39:355–368
Radin EL, Rose RM (1986) Role of subchondral bone in the initiation and progression of cartilage damage. Clin Orthop 213:34–40
Recht MP, Goodwin DW, Winalski CS, White LM (2005) MRI of articular cartilage: revisiting current status and future directions. Am J Roentgenol 185:899–914
Sakellaropoulos P, Costaridou L, Panayiotakis G (1999) An image visualisation tool in mammography. Med Inform Internet Med 24:53–73
Sakellaropoulos P, Costaridou L, Panayiotakis G (2000) Using component technologies for web—based wavelet enhanced mammographic image visualization. Med Inform Internet Med 25:171–181
Specht DF (1990) Probabilistic neural networks. Neural Netw 3:109–118
Spector TD, Cooper C (1993) Radiographic assessment of osteoarthritis in population studies: whither Kellgren and Lawrence? Osteoarthr Cartil 1:203–206
Sun Y, Günther KP, Brenner H (1997) Reliability of radiographic grading of osteoarthritis of the hip and knee. Scand J Rheumatol 26:155–165
Theodoridis S, Koutroumbas K (2003) Pattern Recognition, 2nd edn. Elsevier Academic Press, Amsterdam, Boston, Heidelberg
Tourassi GD (1999) Journey toward computer-aided diagnosis: role of image texture analysis. Radiology 213:317–320
Tuceryan M, Jain AK (1998) Texture analysis. In: Chen CH, Pau LF, Wang PSP (eds) The handbook of pattern recognition and computer vision, 2nd edn. World Scientific Publishing Co., Singapore, pp 207–248
van Belle G, Fisher LD, Heagerty PJ, Lumley T (2004) Biostatistics. A methodology for the health sciences, 2nd edn. Wiley-Interscience, NJ
Acknowledgements
The first author was supported by a grant by the State Scholarship Foundation (SSF), Greece. The authors thank the staff of the Departments of Orthopaedics and Radiology for their contribution to this work.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 Generation of Gabor textural features
A two-dimensional (2-D) Gabor filter G(x,y) can be considered as a sinusoidal plane wave of certain spatial frequency and orientation, modulated by a 2-D Gaussian envelope [12].
For the needs of this study, four (4) filter orientations were used: θ° = 0°, 45°, 90°, and 135°. For each orientation θ, a pair of filters with an anti-symmetric phase relationship was used [16]: G f,θ,0°(x,y) and G f,θ,−90°(x,y).
Each HJS image, corresponding to the determined Region Of Interest (ROI), was convolved with the, G f,θ,0°(x,y) as well as with the G f,θ,−90°(x,y) filter, according to Eqs. 6 and 7, respectively:
where I(i,j) is the input HJS-ROI image, G f,θ,0°(x,y) and G f,θ,−90°(x,y) are the z × z Gabor filters, while GFIMf,θ,0°(i,j) and GFIMf, θ, −90°(i,j) represent the filtered images corresponding to the G f,θ,0°(x,y) and Gf,θ,−90°(x,y) filters, respectively.
Based on the filtered images GFIMf,θ,0°(i,j) and GFIMf,θ,−90°(i,j), an image labelled as Gabor Energy Image (GEN_Image) was produced, according to Eq. 8:
Each point of the GEN_Image represents a measurement that is characterized as Gabor Energy [16].
Four (4) GEN_Images, corresponding to the filter orientations of θ = 0°, 45°, 90°, and 135°, were produced. From each GEN_Image, the following statistics were calculated as textural features and were used by the classification algorithms: mean value, variance, skewness, kurtosis, range and standard deviation.
Following multiple trials regarding the filter specifications, the best classification scores were achieved for textural features that were extracted from images that had been convolved with z × z Gabor filters, z = 5.
1.2 Generation of Gabor energy measure run length textural features
In the present study, new features are proposed based on the combination of GLRLM features and Gabor textural features.
These new features, labelled as Gabor Energy Measure Run Length (GEMRL) features, were extracted from each one of four Gabor Energy Images according to the following approach.
The Gabor Energy values of a Gabor Energy Image were transformed into the region 0–15 by means of a linear transformation providing a grey-level image of 16 discrete grey tones. Denoting this image as GEN_Image_θ_16 (where θ: 0, 45, 90 and 135° represents the orientation of the Gabor filter applied on the image), the new features were generated employing the Eqs. 9–13:
where, j represents the length of the run for the grey tone i, G and R are the numbers of grey tones and run-lengths in the GEN_Image_θ_16, respectively, PN is the number of pixels in the GEN_Image_θ_16, while the r d , g d , P are defined in the Eqs. 14–16:
where, q d (i,j) represents each element of the GLRLM computed along the angular direction d (d: 0, 45, 90 and 135°).
From each GEN_Image_θ_16, four GLRLM were calculated for the angular directions d of 0, 45, 90 and 135°. For each one of the GEMRL features, described by Eqs. 9–13, four values were extracted (one value from each GLRLM), as proposed by Galloway [14]. The mean of these four values was used as the final feature value [14].
Rights and permissions
About this article
Cite this article
Boniatis, I., Costaridou, L., Cavouras, D. et al. Osteoarthritis severity of the hip by computer-aided grading of radiographic images. Med Bio Eng Comput 44, 793–803 (2006). https://doi.org/10.1007/s11517-006-0096-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-006-0096-3