Abstract
Alzheimer's disease is the most frequent type of dementia for elderly patients. Due to aging populations, the occurrence of this disease will increase in the next years. Early diagnosis is crucial to be able to develop more powerful treatments. Brain perfusion changes can be a marker for Alzheimer's disease. In this article, we study the use of SPECT perfusion imaging for the diagnosis of Alzheimer's disease differentiating between images from healthy subjects and images from Alzheimer's disease patients. Our classification approach is based on a linear programming formulation similar to the 1-norm support vector machines. In contrast with other linear hyperplane-based methods that perform simultaneous feature selection and classification, our proposed formulation incorporates proximity information about the features and generates a classifier that does not just select the most relevant voxels but the most relevant “areas” for classification resulting in more robust classifiers that are better suitable for interpretation. This approach is compared with the classical Fisher linear discriminant (FLD) classifier as well as with statistical parametric mapping (SPM). We tested our method on data from four European institutions. Our method achieved sensitivity of 84.4% at 90.9% specificity, this is considerable better the human experts. Our method also outperformed the FLD and SPM techniques. We conclude that our approach has the potential to be a useful help for clinicians.
Similar content being viewed by others
References
McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984) Mental and clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer's disease. Neurology 34(7):939–944
Folstein MF, Folstein SE, McHugh PR (1975) “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12(3):189–198
Gosche KM, Mortimer JA, Smith CD, Markesbery WR, Snowdon DA (2002) Hippocampal volume as an index of Alzheimer neuropathology: findings from the Nun Study. Neurol 58(10):1476–1482
Goethals I, van de Wiele C, Slosman D, Dierckx R (2002) Brain SPET perfusion in early Alzheimer's disease: where to look? Eur J Nucl Med 29(8):975–978
Claus JJ, van Harskamp F, Breteler MM, Krenning EP, de Koning I, van der Cammen TJ, Hofman A, Hasan D (1994) The diagnostic value of SPECT with Tc 99m HMPAO in Alzheimer's disease: a population-based study. Neurol 44(3):454–461
Messa C, Perani D, Lucignani G, Zenorini A, Zito F, Rizzo G, Grassi F, Del Sole A, Franceschi M, Gilardi MC, et al (1994) High-resolution technetium-99m-HMPAO SPECT in patients with probable Alzheimer's disease: comparison with fluorine-18-FDG PET. J Nucl Med 35(2):210–216
Jobst KA, Smith AD, Barker CS, Wear A, King EM, Smith A, Anslow PA, Molyneux AJ, Shepstone BJ, Soper N, et al (1992) Association of atrophy of the medial temporal lobe with reduced blood flow in the posterior parietotemporal cortex in patients with a clinical and pathological diagnosis of Alzheimer's disease. J Neurol Neurosurg Psychiatry 55(3):190–194
Talbot PR, Lloyd JJ, Snowden JS, Neary D, Testa HJ (1998) A clinical role for 99mTc-HMPAO SPECT in the investigation of dementia? J Neurol Neurosurg Psychiatry 64(3):306–313
Van Gool WA, Walstra GJ, Teunisse S, Van der Zant FM, Weinstein HC, Van Royen EA (1995) Diagnosing Alzheimer's disease in elderly, mildly demented patients: the impact of routine single photon emission computed tomography. J Neurol 242(6):401–405
McMurdo ME, Grant DJ, Kennedy NS, Gilchrist J, Findlay D, McLennan JM (1994) The value of HMPAO SPECT scanning in the diagnosis of early Alzheimer's disease in patients attending a memory clinic. Nucl Med Commun 15(6):405–409
Kogure D, Matsuda H, Ohnishi T, Asada T, Uno M, Kunihiro T, Nakano S, Takasaki M (2000) Longitudinal evaluation of early Alzheimer's disease using brain perfusion SPECT. J Nucl Med 41(7):1155–1162
Minoshima S, Giordani B, Berent S, Frey KA, Foster NL, Kuhl DE (1997) Metabolic reduction in the posterior cingulate cortex in very early Alzheimer's disease. Ann Neurol 42(1):85–94
Ishii K, Sasaki M, Yamaji S, Sakamoto S, Kitagaki H, Mori E (1997) Demonstration of decreased posterior cingulate perfusion in mild Alzheimer's disease by means of H215O positron emission tomography. Eur J Nucl Med 24(6):670–673
Ibanez V, Pietrini P, Alexander GE, Furey ML, Teichberg D, Rajapakse JC, Rapoport SI, Schapiro MB, Horwitz B (1998) Regional glucose metabolic abnormalities are not the result of atrophy in Alzheimer's disease. Neurol 50(6):1585–1593
Braak H, Braak E (1997) Diagnostic criteria for neuropathologic assessment of Alzheimer's disease. Neurobiol Aging 18(4):S85–S88
Bobinski M, de Leon MJ, Convit A, De Santi S, Wegiel J, Tarshish CY, Saint Louis LA, Wisniewski HM (1999) MRI of entorhinal cortex in mild Alzheimer's disease. Lancet 353(9146):38–40
Rodriguez G, Vitali P, Calvini P, Bordoni C, Girtler N, Taddei G, Mariani G, Nobili F (2000) Hippocampal perfusion in mild Alzheimer's disease. Psychiatry Res 100(2):65–74
Dawson MR, Dobbs A, Hooper HR, McEwan AJ, Triscott J, Cooney J (1994) Artificial neural networks that use single-photon emission tomography to identify patients with probable Alzheimer's disease. Eur J Nucl Med 21(12):1303–1311
Hamilton D, O'Mahony D, Coffey J, Murphy J, O'Hare N, Freyne P, Walsh B, Coakley D (1997) Classification of mild Alzheimer's disease by artificial neural network analysis of SPET data. Nucl Med Commun 18(9):805–810
Frackowiak RSJ, Friston KJ, Frith CD, Dolan R (1997) Human brain function. Academic Press
Ashburner J, Friston K, Holmes A, Poline J-B (1999) Statistical parametric mapping, SPM'99. The Welcome Department of Cognitive Neurol, Institute of Neurol, University College London. Freely available at: http://www.fil.ion.ucl.ac.uk/spm
Stoeckel J, Malandain G, Migneco O, Koulibaly PM, Robert P, Ayache N, Darcourt J (2001) Classification of SPECT images of normal subjects versus images of Alzheimer's disease patients. In: Niessen WJ, Viergever MA (eds) Proceedings of the 4th international conference on medical image computing and computer-assisted intervention (MICCAI'01), Utrecht, The Netherlands. Lecture notes in computer science, vol 2208, pp 666–674
Brown LG (1992) A survey of image registration techniques. ACM Comput Surveys 24(4):325–376
van den Elsen PA, Pol EJD, Viergever MA (1993) Medical image matching—a review with classification. IEEE Eng Med Biol 12(4):26–39
Maintz JBA, Viergever MA (1998) A survey of medical image registration. Med Image Anal 2(1):1–37
Hill DL, Batchelor PG, Holden M, Hawkes DJ (2001) Medical image registration. Phys Med Biol 46(3):R1–R45
Roche A, Malandain G, Pennec X, Ayache N (1998) The correlation ratio as a new similarity metric for multimodal image registration. In: Wells WM, Colchester ACF, Delp S (eds) Medical image computing and computer-assisted intervention (MICCAI'98), Boston, USA. Lecture notes in computer science, vol 1496, pp 1115–1124
Press WH, Teukolsky SA, Vetterling WT, Flannery BP (1997) Numerical recipes. The art of scientific computing, 2nd edn. Cambridge University Press
Prima S, Ourselin S, Ayache N (2002) Computation of the mid-sagittal plane in 3D brain images. IEEE Trans Med Image 21(2):122–138
Yonekura Y, Nishizawa S, Mukai T, Fujita T, Fukuyama H, Ishikawa M, Kikuchi H, Konishi J, Andersen AR, Lassen NA (1988) SPECT with [99mTc]-d, l-hexamethyl-propylene amine oxime (HM-PAO) compared with regional cerebral blood flow measured by PET: effects of linearization. J Cereb Blood Flow Metab 8(6):S82–S89
Schmidt K (2002) Against: can ROI methodology/normalised tissue activities be used instead of absolute blood flow measurements in the brain? Eur J Nucl Med 29(7):953–956
Vapnik VN (1995) The nature of statistical learning theory. Springer, New York
Bradley PS, Mangasarian OL (1998) Feature selection via concave minimization and support vector machines. In: Shavlik J (ed) Machine learning proceedings of the fifteenth international conference(ICML '98), San Francisco, CA. Morgan Kaufmann, pp 82–90. Available at: ftp://ftp.cs.wisc.edu/math-prog/tech-reports/98-03.ps
Mangasarian OL (1999) Arbitrary-norm separating plane. Oper Res Lett 24:15–23. Available at: ftp://ftp.cs.wisc.edu/math-prog/tech-reports/97-07r.ps
Evgeniou T, Pontil M, Poggio T (2000) Regularization networks and support vector machines. Adv Comput Math 13:1–50
Smola A, Bartlett PL, Schölkopf B, Schuurmann J (eds) (2000) Advances in large margin classifiers. MIT Press, Cambridge, MA
Soonawala D, Amin T, Ebmeier KP, Steele JD, Dougall NJ, Best J, Migneco O, Nobili F, Scheidhauer K (2002) Statistical parametric mapping of (99m)Tc-HMPAO-SPECT images for the diagnosis of Alzheimer's disease: normalizing to cerebellar tracer uptake. Neuroimage 17(3):1193–1202
Mika S, Rätsch G, Müller K-R (2000) A mathematical programming approach to the kernel fisher algorithm. NIPS, pp 591–597
ILOG CPLEX Division, 889 Alder Avenue, Incline Village, Nevada. CPLEX Optimizer, 2004. http://www.cplex.com/+
Grova C, Janin P, Biraben A, Buvat I, Benali H, Bernard AM, Scarabin JM, Gibaud B (2001) Validation of MRI/SPECT similarity based registration methods using realistic simulations of normal and pathological data. In: Niessen WJ, Viergever MA (eds) Medical image computing and computer-assisted intervention (MICCAI'01). Lecture notes in computer science, vol 2208, pp 257–282
Grova C (2002) Simulations réalistes de donées de tomographie monophotonique (TEMP) pour l'évaluation de methodes de recalage TEMP/IRM utilisant des mesures de statistiques de similarité: application dans le contexte de la fusion de données en épilepsie. PhD thesis, Université de Rennes I
Author information
Authors and Affiliations
Corresponding author
Additional information
Glenn Fung received a B.S. degree in pure mathematics from Universidad Lisandro Alvarado in Barquisimeto, Venezuela, then earned an M.S. in applied mathematics from Universidad Simon Bolivar, Caracas, Venezuela, where later he worked as an assistant professor for 2 years. Later, he earned an M.S. degree and a Ph.D. degree in computer sciences from the University of Wisconsin-Madison. His main interests are optimization approaches to machine learning and data mining, with emphasis in support vector machines. In the summer of 2003, he joined the computer aided diagnosis group at Siemens, Medical Solutions in Malvern, PA, where he has been applying machine learning techniques to solve challenging problems that arise in the medical domain. His recent papers are available at www.cs.wisc.edu/gfung.
Jonathan Stoeckel received a B.E. degree from Xi’an Jiao Tong University, Xi'an, China, in 1993 and an M.E. degree from Shanghai Jiao Tong University, Shanghai, China, in 1996. From 1997 to 1998, he did research work in the Data Mining Group at the School of Computing and Information Technology, Griffith University, Brisbane, Australia. He is currently a Ph.D. student at the Department of Computer Science, Dartmouth College, USA. His research interests include data mining, multimedia, database and software engineering.
Rights and permissions
About this article
Cite this article
Fung, G., Stoeckel, J. SVM feature selection for classification of SPECT images of Alzheimer's disease using spatial information. Knowl Inf Syst 11, 243–258 (2007). https://doi.org/10.1007/s10115-006-0043-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10115-006-0043-5