Diagnostic assessment of brain tumours and non-neoplastic brain disorders in vivo using proton nuclear magnetic resonance spectroscopy and artificial neural networks

Poptani, Harish; Kaartinen, Jouni; Gupta, Rakesh K.; Niemitz, Matthias; Hiltunen, Yrjö; Kauppinen, Risto A.

doi:10.1007/s004320050284

Cited by 69 publications

(43 citation statements)

References 29 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Potential roles have been identified in pre-surgical diagnosis of tumour type and grade (4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19), monitoring of treatment response (20), and evaluation of tumour recurrence (21). However, optimised MRS analysis is required to enable the widespread clinical use of the technique.…”

mentioning

confidence: 99%

“…However, optimised MRS analysis is required to enable the widespread clinical use of the technique. Pattern classification of MRS data alone or in combination with MRI features has proven successful for accurately classifying brain tumours (1,(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17). Although brain tumours are the most common solid tumours in children and a major cause of childhood mortality, they are rarer than in adults, and only small numbers have been reported in the few studies using MRS to discriminate paediatric brain tumours (15)(16)(17)(18)(19).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Identification and characterisation of childhood cerebellar tumours by in vivo proton MRS

et al. 2008

View full text Add to dashboard Cite

1 H MRS has great potential for the clinical investigation of childhood brain tumours, but the low incidence in, and difficulties of performing trials on, children have hampered progress in this area. Most studies have used a long-TE, thus limiting the metabolite information obtained, and multivariate analysis has been largely unexplored. Thirty-five children with untreated cerebellar tumours (18 medulloblastomas, 12 pilocytic astrocytomas and five ependymomas) were investigated using a single-voxel short-TE PRESS sequence on a 1.5 T scanner. Spectra were analysed using LCModel TM to yield metabolite profiles, and key metabolite assignments were verified by comparison with high-resolution magic-angle-spinning NMR of representative tumour biopsy samples. In addition to univariate metabolite comparisons, the use of multivariate classifiers was investigated. Principal component analysis was used for dimension reduction, and linear discriminant analysis was used for variable selection and classification. A bootstrap cross-validation method suitable for estimating the true performance of classifiers in small datasets was used. The discriminant function coefficients were stable and showed that medulloblastomas were characterised by high taurine, phosphocholine and glutamate and low glutamine, astrocytomas were distinguished by low creatine and high N-acetylaspartate, and ependymomas were differentiated by high myo-inositol and glycerophosphocholine. The same metabolite features were seen in NMR spectra of ex vivo samples. Successful classification was achieved for glial-cell (astrocytoma þ ependymoma) versus non-glial-cell (medulloblastoma) tumours, with a bootstrap 0.632 þ error, e B.632þ , of 5.3%. For astrocytoma vs medulloblastoma and astrocytoma vs medulloblastoma vs ependymoma classification, the e B.632þ was 6.9% and 7.1%, respectively. The study showed that 1 H MRS detects key differences in the metabolite profiles for the main types of childhood cerebellar tumours and that discriminant analysis of metabolite profiles is a promising tool for classification. The findings warrant confirmation by larger multi-centre studies.

show abstract

mentioning

confidence: 99%

mentioning

confidence: 99%

Identification and characterisation of childhood cerebellar tumours by in vivo proton MRS

et al. 2008

View full text Add to dashboard Cite

show abstract

“…The potential diagnostic and prognostic capabilities of 1 H-MRS in brain tumours have received consider-able attention (Tedeschi et al, 1997;Nelson et al, 1999;Poptani et al, 1999;Preul et al, 2000). It would be reasonable to expect that morphological and clinical responses are preceded by a change in the tumour biochemistry.…”

mentioning

confidence: 99%

Monitoring temozolomide treatment of low-grade glioma with proton magnetic resonance spectroscopy

et al. 2004

View full text Add to dashboard Cite

Assessment of low-grade glioma treatment response remains as much of a challenge as the treatment itself. Proton magnetic resonance spectroscopy ( 1 H-MRS) and imaging were incorporated into a study of patients receiving temozolomide therapy for lowgrade glioma in order to evaluate and monitor tumour metabolite and volume changes during treatment. Patients (n ¼ 12) received oral temozolomide (200 mg m À2 day À1 ) over 5 days on a 28-day cycle for 12 cycles. Response assessment included baseline and three-monthly magnetic resonance imaging studies (pretreatment, 3, 6, 9 and 12 months) assessing the tumour size. Short (TE (echo time) ¼ 20 ms) and long (TE ¼ 135 ms) echo time single voxel spectroscopy was performed in parallel to determine metabolite profiles. The mean tumour volume change at the end of treatment was À33% (s.d. ¼ 20). The dominant metabolite in long echo time spectra was choline. At 12 months, a significant reduction in the mean choline signal was observed compared with the pretreatment (P ¼ 0.035) and 3-month scan (P ¼ 0.021). The reduction in the tumour choline/water signal paralleled tumour volume change and may reflect the therapeutic effect of temozolomide.

show abstract

“…Introduction Recent work has suggested that magnetic resonance spectroscopy (MRS) may be useful in evaluating cancerous lesions [1]. MRS data have been obtained from a variety of body organs including the brain [2,3]. In a brain tumour resection, many (> 12) biopsy samples may be taken.…”

Section: Related Contentmentioning

confidence: 99%

Micro-coils for MR spectroscopy by deep silicon etching

Syms¹,

Ahmad²,

Young³

et al. 2005

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

A process for batch fabrication of low-cost microcoils for magnetic resonance spectroscopy is demonstrated. Conductors are fabricated on oxidised silicon substrates by electroplating metals inside a deep photoresist mould, and passivated using an epoxy-based resist. Through-wafer deep reactive ion etching is used to define sample volumes and stencil cuts around each die, and dies are separated by snap-out. Single-coil and multiple-coil sensors are constructed by stacking parts on baseplates fabricated on the same wafer. Single-turn coils have a Q-factor of ≈ 15 at 63.6 MHz, and Helmholtz coils a Q-factor of ≈ 13. Magnetic resonance imaging and spectroscopy are performed, and a SNR of 900 is achieved.

show abstract

Diagnostic assessment of brain tumours and non-neoplastic brain disorders in vivo using proton nuclear magnetic resonance spectroscopy and artificial neural networks

Abstract: The present data show the clinical utility of non-invasive 1H MRS by automated ANN analysis in the diagnosis of tumour and non-tumour cerebral disorders.

Cited by 69 publications

References 29 publications

Identification and characterisation of childhood cerebellar tumours by in vivo proton MRS

Identification and characterisation of childhood cerebellar tumours by in vivo proton MRS

Monitoring temozolomide treatment of low-grade glioma with proton magnetic resonance spectroscopy

Micro-coils for MR spectroscopy by deep silicon etching

Contact Info

Product

Resources

About