Cluster analysis of signal-intensity time course in dynamic breast MRI: does unsupervised vector quantization help to evaluate small mammographic lesions?

We examined whether neural network clustering could support the characterization of diagnostically challenging breast lesions in dynamic magnetic resonance imaging (MRI). We examined 88 patients with 92 breast lesions (51 malignant, 41 benign). Lesions were detected by mammography and classified Breast Imaging and Reporting Data System (BIRADS) III (median diameter 14 mm). MRI was performed with a dynamic T1-weighted gradient echo sequence (one precontrast and five postcontrast series). Lesions with an initial contrast enhancement ≥50% were selected with semiautomatic segmentation. For conventional analysis, we calculated the mean initial signal increase and postinitial course of all voxels included in a lesion. Secondly, all voxels within the lesions were divided into four clusters using minimal-free-energy vector quantization (VQ). With conventional analysis, maximum accuracy in detecting breast cancer was 71%. With VQ, a maximum accuracy of 75% was observed. The slight improvement using VQ was mainly achieved by an increase of sensitivity, especially in invasive lobular carcinoma and ductal carcinoma in situ (DCIS). For lesion size, a high correlation between different observers was found (R2 = 0.98). VQ slightly improved the discrimination between malignant and benign indeterminate lesions (BIRADS III) in comparison with a standard evaluation method.

[1]  W. Kaiser,et al.  MR imaging of the breast: fast imaging sequences with and without Gd-DTPA. Preliminary observations. , 1989, Radiology.

[2]  Geoffrey C. Fox,et al.  A deterministic annealing approach to clustering , 1990, Pattern Recognit. Lett..

[3]  Geoffrey C. Fox,et al.  Vector quantization by deterministic annealing , 1992, IEEE Trans. Inf. Theory.

[4]  S E Harms,et al.  MR imaging of the breast with rotating delivery of excitation off resonance: clinical experience with pathologic correlation. , 1993, Radiology.

[5]  E Grabbe,et al.  [Signal characteristics of malignant and benign lesions in dynamic 2D-MRT of the breast]. , 1993, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[6]  Rolf Vosshenrich,et al.  Signalverhalten maligner und benigner Läsionen in der dynamischen 2D-MRT der Mamma , 1993 .

[7]  O Lucidarme,et al.  Nonpalpable breast tumors: diagnosis with contrast-enhanced subtraction dynamic MR imaging. , 1994, Radiology.

[8]  [Focal and diffuse lesions in dynamic MR-mammography of healthy probands]. , 1995, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[9]  A Horsman,et al.  Dynamic Contrast-Enhanced Magnetic Resonance Imaging of the Breast Combined with Pharmacokinetic Analysis of Gadolinium-DTPA Uptake in the Diagnosis of Local Recurrence of Early Stage Breast Carcinoma , 1995, Investigative radiology.

[10]  M D Schnall,et al.  Staging of suspected breast cancer: effect of MR imaging and MR-guided biopsy. , 1995, Radiology.

[11]  Fokale und diffuse Läsionen in der dynamischen MR-Mammographie gesunder Probandinnen , 1995 .

[12]  A. Schauer,et al.  [Ductal carcinoma in situ in dynamic MR-mammography at 1.5 T]. , 1996, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[13]  T. Wong,et al.  Automatic motion correction for breast MR imaging. , 1996, Radiology.

[14]  U. Fischer,et al.  Das duktale In-situ-Karzinom in der dynamischen MR-Mammographie bei 1,5 T , 1996 .

[15]  Dominik R. Dersch,et al.  Eigenschaften neuronaler Vektorquantisierer und ihre Anwendung in der Sprachverarbeitung , 1996 .

[16]  G Lutterbey,et al.  Healthy premenopausal breast parenchyma in dynamic contrast-enhanced MR imaging of the breast: normal contrast medium enhancement and cyclical-phase dependency. , 1997, Radiology.

[17]  [Morphology and contrast enhancement of ductal carcinoma in situ in dynamic 1.0 T MR mammography]. , 1997, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[18]  C. Claussen,et al.  Menstrual cycle and age: influence on parenchymal contrast medium enhancement in MR imaging of the breast. , 1997, Radiology.

[19]  A Heinig,et al.  Contrast-enhanced MRI of the breast: accuracy, value, controversies, solutions. , 1997, European journal of radiology.

[20]  M. Schnall,et al.  MR imaging of ductal carcinoma in situ. , 1997, Radiology.

[21]  M. Reiser,et al.  Morphologie und Anreicherungsverhalten des duktalen In-situ-Karzinoms in der dynamischen MR-Mammographie bei 1,0 T , 1997 .

[22]  K. Obermayer,et al.  PHASE TRANSITIONS IN STOCHASTIC SELF-ORGANIZING MAPS , 1997 .

[23]  Axel Wismüller,et al.  A Neural Network Approach to Functional MRI Pattern Analysis — Clustering of Time-Series by Hierarchical Vector Quantization , 1998 .

[24]  U. Fischer,et al.  MR imaging of mammographically detected clustered microcalcifications: is there any value? , 1998, Radiology.

[25]  J F Debatin,et al.  Improved diagnostic accuracy in dynamic contrast enhanced MRI of the breast by combined quantitative and qualitative analysis. , 1998, The British journal of radiology.

[26]  C. Kuhl,et al.  Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions? , 1999, Radiology.

[27]  M A Helvie,et al.  Linear motion correction in three dimensions applied to dynamic gadolinium enhanced breast imaging. , 1999, Medical physics.

[28]  C. Kuhl,et al.  MRI of breast tumors , 2000, European Radiology.

[29]  U. Fischer,et al.  International investigation of breast MRI: results of a multicentre study (11 sites) concerning diagnostic parameters for contrast-enhanced MRI based on 519 histopathologically correlated lesions , 2001, European Radiology.

[30]  E. Grabbe,et al.  Classification of hypervascularized lesions in CE MR imaging of the breast , 2002, European Radiology.

[31]  G Brix,et al.  Classification of signal-time curves from dynamic MR mammography by neural networks. , 2001, Magnetic resonance imaging.

[32]  M Thelen,et al.  Improved artificial neural networks in prediction of malignancy of lesions in contrast-enhanced MR-mammography. , 2003, Medical physics.

[33]  The emerging role of breast magnetic resonance imaging. , 2003, The Journal of the Oklahoma State Medical Association.

[34]  Peter Aspelin,et al.  Application of artificial neural networks to the analysis of dynamic MR imaging features of the breast , 2004, European Radiology.

[35]  Klaus Schulten,et al.  Self-organizing maps: stationary states, metastability and convergence rate , 1992, Biological Cybernetics.

[36]  Axel Wismüller,et al.  Cluster Analysis of Biomedical Image Time-Series , 2002, International Journal of Computer Vision.

[37]  O. Rieker,et al.  Selbstorganisierende neuronale Netze zur automatischen Detektion und Klassifikation von Kontrast(mittel)-verstärkten Läsionen in der dynamischen MR-Mammographie , 2005 .

[38]  H. Kauczor,et al.  [Self-organizing neural networks for automatic detection and classification of contrast (media) enhancement of lesions in dynamic MR-mammography]. , 2005, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[39]  C. Kuhl,et al.  Diagnostic usefulness of segmental and linear enhancement in dynamic breast MRI , 2005, European Radiology.