Boosting phase-contrast MRI performance in idiopathic normal pressure hydrocephalus diagnostics by means of machine learning approach.

OBJECTIVE: Phase-contrast MRI allows detailed measurements of various parameters of CSF motion. This examination is technically demanding and machine dependent. The literature on this topic is ambiguous. Machine learning (ML) approaches have already been successfully utilized in medical research, but none have yet been applied to enhance the results of CSF flowmetry. The aim of this study was to evaluate the possible contribution of ML algorithms in enhancing the utilization and results of MRI flowmetry in idiopathic normal pressure hydrocephalus (iNPH) diagnostics. METHODS: The study cohort consisted of 30 iNPH patients and 15 healthy controls examined on one MRI machine. All major phase-contrast parameters were inspected: peak positive, peak negative, and average velocities; peak amplitude; positive, negative, and average flow rates; and aqueductal area. The authors applied ML algorithms to 85 complex features calculated from a phase-contrast study. RESULTS: The most distinctive parameters with p < 0.005 were the peak negative velocity, peak amplitude, and negative flow. From the ML algorithms, the Adaptive Boosting classifier showed the highest specificity and best discrimination potential overall, with 80.4% ± 2.9% accuracy, 72.0% ± 5.6% sensitivity, 84.7% ± 3.8% specificity, and 0.812 ± 0.047 area under the receiver operating characteristic curve (AUC). The highest sensitivity was 85.7% ± 5.6%, reached by the Gaussian Naive Bayes model, and the best AUC was 0.854 ± 0.028 by the Extra Trees classifier. CONCLUSIONS: Feature extraction algorithms combined with ML approaches simplify the utilization of phase-contrast MRI. The highest-performing ML algorithm was Adaptive Boosting, which showed good calibration and discrimination on the testing data, with 80.4% accuracy, 72.0% sensitivity, 84.7% specificity, and 0.812 AUC. Phase-contrast MRI boosted by the ML approach can help to determine shunt-responsive iNPH patients.

Magnetic resonance markers of bilateral neuronal metabolic dysfunction in patients with unilateral internal carotid artery occlusion.

OBJECTIVES: To evaluate cerebral hemodynamic, metabolic and anatomic changes occurring in patients with unilateral occlusion of the internal carotid artery (ICA). MATERIALS AND METHODS: Twenty-two patients with unilateral occlusion of ICA and twenty age and sex matched healthy subjects were included in the study. Single voxel proton magnetic resonance spectroscopy (1H-MRS) of the centrum semiovale, semi-automated hippocampal volumetry in T1-weighted scans and transcranial Doppler examination (TCD) with calculation of Breath Holding Index (BHI) were performed in both groups. Metabolic, anatomic, and hemodynamic features were compared between the two groups. RESULTS: The N-acetylaspartate (NAA)/choline (Cho) ratio was significantly lower in both hemispheres of enrolled patients compared to controls (p = 0.005 for the side with occlusion, p = 0.04 for the side without occlusion). The hippocampus volume was significantly reduced bilaterally in patients compared to healthy subjects (p = 0.049). A statistically significant difference in BHI values was observed between the side with occlusion and without occlusion (p = 0.037) of the patients, as well as between BHI values of the side with occlusion and healthy volunteers (p = 0.014). DISCUSSION: Patients with unilateral ICA occlusion have reduced NAA/Cho ratio in the white matter of both hemispheres and have bilateral atrophy of hippocampus. The alteration of hemodynamics alone cannot explain these changes.

Structural volumetry in NPH diagnostics and treatment-future or dead end?

To assess automated volumetric analysis as a potential presurgical diagnostic tool or as a method to potentially shed light on normal pressure hydrocephalus (NPH) pathophysiology. MRI imaging according to our protocol was performed in 29 NPH patients, 45 non-NPH (but suspected) patients and 15 controls. Twenty patients underwent a second MRI 3 months after ventriculoperitoneal (VP) shunt surgery. All structures relevant to NPH diagnosis were automatically segmented using commercial software. The results were subsequently tested using ANOVA analysis. Significant differences in the volumes of the corpus callosum, left hippocampus, internal globus pallidus, grey and white matter and ventricular volumes were observed between NPH group and healthy controls. However, the differences between NPH and non-NPH groups were non-significant. Three months after, VP shunt insertion decreased ventricular volume was the only clearly significant result (p value 0.0001). Even though a detailed volumetric study shows several significant differences, volumetric analysis as a standalone method does not provide a simple diagnostic biomarker, nor does it shed a light on an unknown NPH aetiology.

Role of DESH, callosal angle and cingulate sulcus sign in prediction of gait responsiveness after shunting in iNPH patients.

Primary endpoint of this single-centre, prospective consecutive cohort study was to evaluate DESH score, CA, CSS and Evans index of suspected iNPH patients against the reference standard of lumbar infusion test (LIT) and external lumbar drainage (ELD) in prediction of gait response after VP shunt implantation in patients with idiopathic normal pressure hydrocephalus (iNPH). Patients were assigned to NPH and non-NPH groups based on LIT and ELD results. Age-matched controls were added for group comparison. 32 NPH, 46 non-NPH and 15 control subjects were enrolled in the study. There were significant differences in mean preoperative DESH scores of NPH, non-NPH and control groups (6.3 ± 2.3 ([±SD]) (range 2-10) vs 4.5 ± 2.4 (range 0-10) vs 1.0 ± 1.2 (range 0-4)). Differences in mean CA and Evans index were not significant between NPH and non-NPH groups. CSS showed 62.5% sensitivity, 60.87% specificity, 52.63% PPV and 70% NPV for differentiation of NPH and non-NPH groups. A CA of 68 degrees had 48.49% sensitivity, 76.09% specificity, 59.26% PPV 67.31% NPV and DESH score of 4 had 93.75% sensitivity, 41.30% specificity, 52.63% PPV and 90.48% NPV for differentiation between NPH and non-NPH groups. The groups of probable iNPH patients with gait impairment diagnosed by high DESH score or positive functional testing did not overlap and DESH score did not correlate with gait improvement after ELD. DESH score should not be used as a simple diagnostic or prognostic marker of iNPH and we could not confirm the benefit of measurement of callosal angle and cingulate sulcus sign.

Distinctive Patterns of Seizure-Related White Matter Alterations in Right and Left Temporal Lobe Epilepsy.

Background: We hypothesized that right and left temporal lobe epilepsy (RTLE and LTLE, respectively) have distinctive spatial patterns of white matter (WM) changes that can be differentiated and interpreted with the use of multiple diffusion parameters. We compared the global microstructure of fiber bundles with regard to WM alterations in both RTLE and LTLE, addressing some of the methodological issues of previous studies. Methods: Diffusion tensor imaging data from 17 patients with RTLE (age: 40.7 ± 10.4), 15 patients with LTLE (age: 37.3 ± 10.4), and 15 controls (age: 34.8 ± 11.2) were used in the study. WM integrity was quantified by fractional anisotropy (FA), mean diffusivity (MD), longitudinal diffusivity (LD), and radial diffusivity (RD). The diffusion parameters were compared between the groups in tracts representing the core of the fiber bundles. The volumes of hippocampi and amygdala were subsequently compared across the groups, while the data were adjusted for the effect of hippocampal sclerosis. Results: Significantly reduced FA and increased MD, LD, and RD were found bilaterally over widespread brain regions in RTLE. An increase in MD and RD values was observed in widespread WM fiber bundles ipsilaterally in LTLE, largely overlapping with regions where FA was lower, while no increase in LD was observed. We also found a difference between the LTLE and RTLE groups for the right hippocampal volume (with and without adjustment for HS), whereas no significant volume differences were found between patients and controls. Conclusions: It appears that patients with RTLE exhibit a more widespread pattern of WM alterations that extend far beyond the temporal lobe in both ipsilateral and contralateral hemisphere; furthermore, these changes seem to reflect more severe damage related to chronic degeneration. Conversely, more restrained changes in the LTLE may imply a pattern of less severe axonal damage, more restricted to ipsilateral hemisphere. Comprehensive finding of more prominent hippocampal atrophy in the RTLE raises an interesting issue of seizure-induced implications on gray matter and WM microstructure that may not necessarily mean a straightforward causal relationship. Further correlations of diffusion-derived metrics with neuropsychological and functional imaging measures may provide complementary information on underlying WM abnormalities with regard to functional hemispheric specialization.

Accuracy of VarioGuide Frameless Stereotactic System Against Frame-Based Stereotaxy: Prospective, Randomized, Single-Center Study.

INTRODUCTION: Frameless stereotactic brain biopsy systems are widely used today. VarioGuide (VG) is a relatively novel frameless system. Its accuracy was studied in a laboratory setting but has not yet been studied in the clinical setting. The purpose of this study was to determine its accuracy and diagnostic yield and to compare this with frame-based (FB) stereotaxy. MATERIAL AND METHODS: Overall, 53 patients (33 males and 20 females, 60 ± 15 years old) were enrolled into this prospective, randomized, single-center study. Twenty-six patients were randomized into the FB group and 27 patients into the VG group. Real trajectory was pointed on intraoperative magnetic resonance. The distance of the targets and angle deviation between the planned and real trajectories were computed. The overall discomfort of the patient was subjectively assessed by the visual analog scale score. RESULTS: The median lesion volume was 5 mL (interquartile range [IQR]: 2-16 mL) (FB) and 16 mL (IQR: 2-27 mL) (VG), P = 0.133. The mean distance of the targets was 2.7 ± 1.1 mm (FB) and 2.9 ± 1.3 mm (VG), P = 0.456. Mean angle deviation was 2.6 ± 1.3 deg (FB) and 3.5 ± 2.1 deg (VG), P = 0.074. Diagnostic yield was 93% (25/27) in VG and 96% (25/26) in FB, P = 1.000. Mean operating time was 47 ± 26 minutes (FB) and 59 ± 31 minutes (VG), P = 0.140. One minor bleeding was encountered in the VG group. Overall patient discomfort was significantly higher in the FB group (visual analog scale score 2.5 ± 2.1 vs. 1.2 ± 0.6, P = 0,004). CONCLUSIONS: The VG system proved to be comparable in terms of the trajectory accuracy, rate of complications and diagnostic yield compared with the "gold standard" represented by the traditional FB stereotaxy for patients undergoing brain biopsy. VG is also better accepted by patients.

Difference in white matter microstructure in differential diagnosis of normal pressure hydrocephalus and Alzheimer's disease.

OBJECTIVES: Alzheimer's disease (AD) and normal pressure hydrocephalus (NPH) are both associated with cognitive decline and ventriculomegaly. While promising approach in differentiating between the two diseases, only a few diffusion tensor imaging (DTI) studies compared directly NPH and AD patients. The current study compares global whitematter (WM) alterations in AD and NPH addressing some of the methodological issues of previous studies. PATIENTS AND METHODS: Diffusion tensor images were obtained from 17 patients with NPH, 14 with AD, and 17 healthy controls. White matter integrity was quantified by fractional anisotropy (FA), mean (MD), axial (λ1) and radial diffusivity (RD). The diffusion parameters were compared between the groups in 'skeletonised' tracts representing the core of the fibre bundles. RESULTS: Reduced FA was found in NPH patients throughout the corpus callosum, particularly in the splenium, along with increased RD. On the other hand, FA, MD and RD were higher in NPH in the cortico-fugal fibres arising from the frontal and parietal cortex. While no FA changes were detected in AD patients compared to controls, widespread increased RD was observed. When comparing NPH and AD patients, higher FA, MD and RD was observed in the corona radiata in the periventricular fibres arising from the frontal and parietal cortex in NPH patients. The ventricular volumes were correlated with diffusivity parameters in the tracts next to the ventricles in AD and NPH patients. CONCLUSION: Our analysis identified a pattern of WM diffusion alterations that can differentiate NPH patients from controls and AD patients.

One-Year Clinical and Computed Tomography Angiographic Outcomes After Bioresorbable Vascular Scaffold Implantation During Primary Percutaneous Coronary Intervention for ST-Segment-Elevation Myocardial Infarction: The PRAGUE-19 Study.

BACKGROUND: Bioresorbable vascular scaffolds (BVS) represent promising new technology, but data on their long-term outcomes in ST-segment-elevation myocardial infarction (STEMI) setting are missing. The aim was to analyze 1-year clinical and computed tomographic angiographic outcomes after BVS implantation in STEMI. METHODS AND RESULTS: PRAGUE-19 is a prospective multicenter single-arm study enrolling consecutive STEMI patients undergoing primary percutaneous coronary intervention (pPCI) with intention-to-implant BVS. A total of 343 STEMI patients were screened during 15 months enrollment period, and 70 patients (mean age 58.6±10.3 and 74% males) fulfilled entry criteria and BVS was successfully implanted in 96% of them. All patients were invited for clinical and computed tomographic angiographic control 1 year after BVS implantation. Restenosis was defined as ≥75% area stenosis within the scaffolded segment. Three events were potentially related to BVS: 1 in-stent restenosis (treated 7 months after pPCI with drug-eluting balloon), 1 stent thrombosis (treated 2 weeks after pPCI by balloon dilatation-this patient stopped all medications after pPCI), and 1 sudden death at home 9 months after pPCI. Four other patients had events definitely unrelated to BVS. Overall, 1-year mortality was 2.9%. Computed tomographic angiography after 1 year was performed in 59 patients. All BVS were widely patent, and binary restenosis rate was 2% (the only restenosis mentioned above). Mean in-scaffold minimal luminal area was 7.8±2.6 mm(2), area stenosis was 20.1±16.3%, minimal luminal diameter was 3.0±0.6 mm, and diameter stenosis was 12.8±11.1%. CONCLUSIONS: BVS implantation in STEMI is feasible and safe and offers excellent 1-year clinical and angiographic outcomes.

Recognition of anaplastic foci within low-grade gliomas using MR spectroscopy.

BACKGROUND: The recognition of anaplastic foci within low-grade gliomas is of extreme importance in patients under follow-up for Grade II gliomas. We present the algorithm of MR spectroscopy (MRS)-guided brain biopsy and its correlation with tumour histology. METHODS: Twenty-seven patients harbouring suspected Grade II/III glioma were examined on our 3T MR. 2D PRESS-CSI metabolite images of Choline/Creatine, Creatine/N-acetylaspartate and Choline/N-acetylaspartate were calculated and exported to the DICOM format. According to these maps, a stereobiopsy was performed at the point of maximum Choline/Creatine ratio prior to tumour resection. In the case of enhancing tumour, a subsequent biopsy was performed from the point of enhancement. Comparisons were made between the histology of the biopsied specimens and the resected tumours. RESULTS: Eleven tumours were diagnosed as high-grade and sixteen as low-grade lesions. The correlation between main spectroscopic ratios (Cho/Cr and Cho/NAA) was strongly positive at the points of maximum Cho/Cr. Similar results were obtained at the points of contrast enhancement. Comparison of histological parameters of biopsy samples at the points of maximum Cho/Cr and histological examination of the completely resected tumours gives a strong correlation of tumour grade, number of mitoses and Ki-67 expression. The diagnostic accuracy of MRS-guided biopsy was 84%. The absolute value of Cho/NAA was higher in high-grade compared to that of low-grade lesions. The value of Cho/NAA ratio of 0.9 using MRS produced a sensitivity and specificity of 78% in the differentiation between low-grade and high-grade lesions. Combining MRS with structural MR, the sensitivity increased to 86% and the specificity to 80%. CONCLUSIONS: Strong correlation was demonstrated between Cho/Cr and Ch/NAA ratios. Strong correlation was demonstrated between histological parameters of biopsy samples taken using Cho/Cr ratio and those from total tumour examination. Diagnostic accuracy of MRS-guided biopsy was 84%. Sensitivity and specificity of MRS combined with structural MR reaches 86% and 80%.

Functional magnetic resonance imaging of cerebral activation during spinal cord stimulation in failed back surgery syndrome patients.

Spinal cord stimulation (SCS) consisting of electrical stimulation of the dorsal spinal cord using epidural electrodes has been shown to relieve chronic neuropathic pain. To analyze the cerebral activation patterns related to SCS, and to evaluate the effects of SCS on the processing of acute experimental pain, we performed functional magnetic resonance imaging (fMRI) on eight patients suffering from failed back surgery syndrome who were also being treated with SCS for severe pain in their legs and lower back. Three types of stimulation were used, each lasting 36s: (i) SCS, (ii) heat pain (HP) applied to the leg affected by neuropathic pain, and (iii) simultaneous HP and SCS. During SCS, we found increased activation of the medial primary sensorimotor cortex somatotopically corresponding to the foot and/or perineal region, contralateral posterior insula, and the ipsilateral secondary somatosensory cortex (S2). Decreased activation was seen in the bilateral primary motor cortices and the ipsilateral primary somatosensory cortex corresponding to the shoulder, elbow and hand. Compared to separately presented HP and SCS, simultaneous HP and SCS showed statistically significant activation of the bilateral inferior temporal cortex and the ipsilateral cerebellar cortex. The activation of the primary motor cortex, insula and S2 during SCS may directly interfere with the processing of neuropathic pain. When SCS is associated with heat pain, the paralimbic association cortex and cerebellum show activation exceeding the sum of activations resulting from separate SCS and heat pain stimulation. The explanation of this could possibly rest with the continuous comparisons of simultaneous pain and somatosensory sensations occurring in a single dermatome.

Effects of spinal cord stimulation on the cortical somatosensory evoked potentials in failed back surgery syndrome patients.

OBJECTIVE: To evaluate the functional activation of the somatosensory cortical regions in neuropathic pain patients during therapeutic spinal cord stimulation (SCS). METHODS: In nine failed back surgery syndrome patients, the left tibial and the left sural nerves were stimulated in two sessions with intensities at motor and pain thresholds, respectively. The cortical somatosensory evoked potentials were analyzed using source dipole analysis based on 111 EEG signals. RESULTS: The short-latency components of the source located in the right primary somatosensory cortex (SI: 43, 54 and 65ms) after tibial nerve stimulation, the mid-latency SI component (87ms) after sural nerve stimulation, and the mid-latency components in the right (approximately 161ms) and left (approximately 168ms) secondary somatosensory cortices (SII) were smaller in the presence of SCS than in absence of SCS. The long-latency source component arising from the mid-cingulate cortex (approximately 313ms) was smaller for tibial and larger for sural nerve stimuli during SCS periods compared to periods without SCS. CONCLUSIONS: SCS attenuates the somatosensory processing in the SI and SII. In the mid-cingulate cortex, the effect of SCS depends on the type of stimulation and nerve fibers involved. SIGNIFICANCE: Results suggest that the effects of SCS on cortical somatosensory processing may contribute to a reduction of allodynia during SCS.

Source imaging of the cortical 10 Hz oscillations during cooling and warming in humans.

Primary cold and warm afferent fibers show a robust overshoot in their firing during periods of temperature change, which subsides during tonic thermal stimulation. Our objective was to analyze cortical activation, on a scale of hundreds of milliseconds, occurring during the process of dynamic cooling and warming, based on an evaluation of the amplitude changes seen in 10 Hz electroencephalographic oscillations. Eleven right-handed subjects were exposed to innocuous cold ramp stimuli (from 32 degrees C to 22 degrees C, 10 degrees C/s) and warm ramp stimuli (32 degrees C to 42 degrees C, 10 degrees C/s) on the thenar region of their right palm, using a contact thermode. EEG was recorded from 111 scalp sites, and the 10 Hz current source densities were modeled using low-resolution electromagnetic tomography. During cooling, the earliest amplitude decreases of 10 Hz oscillations were seen in the contralateral posterior insula and secondary somatosensory cortex (SII), and the premotor cortex (PMC). During warming, the earliest events were only observed in the PMC and occurred approximately 0.7 s later than during cooling. Linear regression analysis between 10 Hz current source densities and temperature variations revealed cooling-sensitive activation in the bilateral posterior insula, PMC and the anterior cingulate cortex. During warming, the amplitude of 10 Hz oscillations in the PMC and posterior insula correlated with stimulus temperature. Dynamic thermal stimulation activates, in addition to the posterior insula and parietal operculum, the lateral PMC. The activation of the anterior cingulate cortex during cooling may aid in the anticipation of the cold temperature end-point and provide continuous evaluation of the thermal stimulus.

Somatosensory-evoked potentials are influenced differently by isometric muscle contraction of stimulated and non-stimulated hand in humans.

Chronic pain is associated with motor dysfunctions, and stimulation of the motor cortex has been shown to alleviate chronic pain. Recently Nakata et al. [Pain 107 (2004) 91-98] showed differentiated patterns of neuromagnetic evoked fields following painful laser stimulation during phasic movements of stimulated and non-stimulated hand. Phasic movements and static contractions differ in their functional activation of the motor cortices. Therefore, we decided to analyze the effects of isometric contractions of intrinsic right and left hand muscles on cortical sources of somatic-evoked potentials related to a painful galvanic stimulation of the right middle finger. Using spatio-temporal source dipole analysis of 111 electroencephalographic signals in 10 right-handed men, source activities were evaluated in the left primary somatosensory cortex (S1), left (S2(L)) and right (S2(R)) secondary somatosensory cortex, anterior cingulate cortex (ACC) and posterior cingulate cortex (PCC). Ipsilateral hand muscle contraction was associated with a decrease of source activity in S1 and with subsequent increases in S2(L) and the PCC. Contralateral hand muscle contraction was accompanied by a decrease of source activity in bilateral S2 cortices followed by decreases in the S1 and anterior cingulate cortex. Results suggest early suppression of source activity in S1 during ipsilateral hand muscle contractions and in bilateral S2 during contralateral hand muscle contractions.

EEG source analysis and fMRI reveal two electrical sources in the fronto-parietal operculum during subepidermal finger stimulation.

Using functional magnetic resonance imaging (fMRI) and electroencephalographic (EEG) source dipole analysis in 10 normal subjects, two electrical source dipoles in the contralateral fronto-parietal operculum were identified during repetitive painful subepidermal stimulation of the right index finger. The anterior source dipole peaking at 79 +/- 8 ms (mean +/- SD) was located in the frontal operculum, and oriented tangentially toward the cortical surface. The posterior source dipole peaking at 118 +/- 12 ms was located in the upper bank of the Sylvian fissure corresponding to the second somatosensory cortex (S2). The orientations of the posterior source dipoles displayed large variability, but differed significantly (P < 0.05) from the orientations of the anterior source dipoles. Electrical sources and fMRI clusters were also observed in ipsilateral fronto-parietal operculum. However, due to low signal-to-noise ratio of ipsilateral EEG sources in individual recordings, separation of sources into anterior and posterior clusters was not performed. Combined fMRI and source dipole EEG analysis of individual data suggests the presence of two distinct electrical sources in the fronto-parietal operculum participating in processing of somatosensory stimuli. The anterior region of the fronto-parietal operculum shows earlier peak activation than the posterior region.

Desynchronization of cortical rhythms following cutaneous stimulation: effects of stimulus repetition and intensity, and of the size of corpus callosum.

OBJECTIVE: Event-related desynchronization (ERD) and synchronization (ERS) of the Rolandic electroencephalographic (EEG) rhythms following brief, innocuous electrocutaneous stimulation were studied with respect to stimulus intensity and repetition and the size of corpus callosum (CC). METHODS: EEG was recorded using 82 closely spaced electrodes in 13 right-handed subjects. The subjects received 650 brief electrical stimuli to the right index finger at irregular intervals (6-12 s) in 5 blocks. The intensities of the stimuli varied randomly at 20, 30, 50, 65 and 80% of pain threshold. RESULTS: Mu- and beta-ERD of 0.3-0.6 s latency over the contra- and ipsilateral S1/M1 area was observed in all subjects. Post-stimulus beta-ERS over the contra- and ipsilateral frontal cortices with a peak latency of 0.6-0.8 s was found in 9 subjects. Stimuli presented in the second half of the experiment were followed by a smaller ipsilateral mu-ERD and smaller contra- and ipsilateral beta-ERD than stimuli applied in the first block. Mu- and beta-ERD and beta-ERS distinguished weak (20%) from intermediate and strong stimuli (>35%) but not the intermediate from strong stimuli. The amplitude of ipsilateral beta-ERS correlated positively with the size of intermediate truncus of CC (r(9)=0.71, P<0.05). In contrast, ipsilateral ERD showed no significant correlations with the size of CC. CONCLUSIONS: Habituation of ipsilateral mu-ERD and bilateral beta-ERD and beta-ERS suggests that these cortical responses are parts of the orienting response, and fail to disentangle fine intensity gradations. Ipsilateral beta-ERS appears to be mediated by the transcallosal fiber system.

Source activity in the human secondary somatosensory cortex depends on the size of corpus callosum.

If corpus callosum (CC) mediates the activation of the secondary somatosensory area (SII) ipsilateral to the side of stimulation, then the peak latencies of the contra- and ipsilateral SII activity as well as the amplitude of the ipsilateral SII activity should correlate with the size of CC. Innocuous electrical stimuli of five different intensities were applied to the ventral surface of the right index finger in 15 right-handed men. EEG was recorded using 82 closely spaced electrodes. The size of CC and of seven callosal regions was measured from the mid-sagittal slice of a high-resolution anatomical MRI. The activation in the contralateral and ipsilateral SII was evaluated using spatio-temporal source analysis. At the strongest stimulus intensity, the size of the intermediate part of the callosal truncus correlated negatively with the interpeak latency of the sources in ipsi- and contralateral SII (r = -0.83, P < 0.01). Stepwise regression analysis showed that the large size of the intermediate truncus of CC was paralleled by a latency reduction of peak activity of the ipsilateral SII, whereas both contra- and ipsilateral peak latencies were positively correlated. The peak amplitude of the ipsilateral SII source correlated positively with the size of the intermediate truncus of CC, and with the peak amplitudes of sources in the primary somatosensory cortex (SI) and in the mesial frontal cortex. The results suggest that in right-handed neurologically normal men, the size of the intermediate callosal truncus contributes to the timing and amplitude of ipsilateral SII source activity.