False-positive analysis of functional MRI during simulated deep brain stimulation: a phantom study.

Area of Science:

• Neuroimaging
• Medical Physics

Background:

• Functional MRI (fMRI) is crucial for evaluating deep brain stimulation (DBS) effects.
• Understanding and controlling false-positive signals in fMRI is essential for accurate interpretation of DBS-related brain activity.

Purpose of the Study:

• To investigate and quantify false-positive activations/deactivations in fMRI during simulated deep brain stimulation (DBS).
• To evaluate the impact of interscan waiting time (ISWT) on fMRI false-positive rates.

Main Methods:

• fMRI experiments were conducted on an agar-gel phantom with implanted DBS electrodes using a 1.5T scanner.
• Gradient-echo echo-planar imaging (GE-EPI) sequences were employed with varying ISWT (3500 msec and 5160 msec).
• Statistical Parametric Mapping (SPM2) software was used to analyze data and detect false-positive voxels across different P-value thresholds.

Main Results:

• The number of false-positive voxels did not significantly differ between experimental and control conditions for both short and long ISWT.
• False-positive counts increased with less stringent P-value thresholds (e.g., ~40 at P < 0.05 compared to zero at P < 0.0001).
• The observed pattern of increasing false positives with P-value was consistent across all tested conditions.

Conclusions:

• A statistical threshold of P < 0.001 or stricter is recommended to effectively control false-positive findings in fMRI studies with similar designs.
• A shorter ISWT of 3500 msec did not lead to an increase in false-positive fMRI reactions compared to a longer ISWT of 5160 msec.