Beam-F3 error analysis and validation of Neurally-F3 equations using realistic head models.

Area of Science:

• Neuroscience
• Medical Imaging
• Biomedical Engineering

Background:

• Accurate F3 point localization is crucial for non-invasive brain stimulation techniques like repetitive transcranial magnetic stimulation (rTMS).
• The widely used Beam-F3 heuristic lacks systematic accuracy assessment across diverse cranial anatomies.
• Existing methods may introduce significant errors in targeting the dorsolateral prefrontal cortex.

Purpose of the Study:

• To evaluate the accuracy of the Beam-F3 heuristic for F3 point localization using realistic head models.
• To develop an improved, sex-specific method for accurate F3 point localization.
• To enhance the reliability of targeting in rTMS.

Main Methods:

• Evaluated Beam-F3 accuracy against the 10-20 International System standard using 250 diverse realistic head models.
• Developed sex-specific multiple linear regression equations (Neurally-F3) based on scalp-distance measurements.
• Validated Neurally-F3 on an independent dataset of 128 head models.

Main Results:

• Beam-F3 exhibited significant mean errors (5.59-6.09 mm) in both female and male models.
• The Adjusted Beam-F3 method performed worse, with mean errors exceeding 8 mm.
• Neurally-F3 achieved sub-millimeter accuracy (0.19-0.23 mm) in validation, with R² > 0.98 during cross-validation.

Conclusions:

• The Beam-F3 method demonstrates significant inaccuracies for F3 point localization in realistic head models.
• Neurally-F3 offers a precise and practical solution for scalp-based F3 localization, overcoming Beam-F3's limitations.
• Neurally-F3 has the potential to significantly improve targeting accuracy and reliability in rTMS applications.