Logo TU Ilmenau


Prof. Dr. Jens Haueisen


Telefon +49 3677 69-2860

E-Mail senden


Validierung von bioelektrischen und biomagnetischen Mess- und Auswerteverfahren mittels physikalischer Phantomen


Source reconstruction is a widely used method to estimate the location, orientation, and strength of bioelectrical sources from surface potential measurements or from magnetic field measurements. It is extensively used in the localization of neuronal activity in the brain, but it is also used to localize electrophysiological activity in the heart and in other biomedical research areas. Source reconstruction comprises the computation of bioelectric and biomagnetic fields due to given sources (the forward problem) and the source parameter estimation based on given measurements (the inverse problem).

Inverse bioelectric and biomagnetic problems are computationally complex and don’t have unique solutions. Consequently, validation should be an inherent part of development and application of data analysis procedures. Validation approaches in general include simulations, phantom measurements, in vitro and in vivo measurements.

Physical phantom measurements provide a unique means to assess to performance of source localization techniques. Unlike simulations, they take into account the real world influences, such as for example environmental noise or 3D positioning errors, thus giving an error estimate of the entire source reconstruction procedure. Unlike in vivo measurements, no physiological uncertainties exist and the ground truth in terms of source position, strength, orientation, and extent is known.

Figure 1: Front view of the physical torso phantom

Figure 2: The torso phantom with a bulk of anisotropic skeins (red lines) and a current dipole (blue, below the bulk of anisotropic skeins). At the surface of the phantom the electrodes (blue) are displayed. The 195 magnetic sensors (red) above the phantom are arranged in triplets (vectorial field measurement, Argos 200 System). POS1 and POS2 indicate the set-ups for the more lateral and more inferior dipole position, respectively. The artificial current dipole is rotated with the help of the mechanical construction indicated in blue.

We provide open data for a dipole localization benchmark problem and an extended source reconstruction problem (the latter was accepted as TEAM Workshop benchmark 31).

Besides the validation of source reconstruction schemes, phantoms have also been used to experimentally investigate the influence of anisotropic conductivity onto magnetic fields and electric potentials (Liehr & Haueisen 2008; Sengül et al. 2008). Figure 1 shows a torso phantom in which a bulk of anisotropic material was introduced (Figure 2). The magnetic field and the electric potential produced by a single electric current dipole were measured with and without the anisotropic bulk in the torso.  Figure 3 shows the normalized signal strengths and demonstrates the strong influence of anisotropic volume conduction.

Phantoms can also be used for experimental validation of theoretical concepts, such as the vortex current concept in Liehr et al. 2005 and Dutz et al. 2006.

Figure 3: Normalized signal strengths over the angle between anisotropy orientation and dipole direction. Results for POS2 of the dipole (see Figure 2 above) are displayed. For the different experiments, the signal strength was reduced between 17% and 43% for different dipole positions when comparing the parallel alignment of dipole orientation and anisotropy direction with the orthogonal alignment.

We provide open data for a dipole localization benchmark problem and an extended source reconstruction problem (the latter was accepted as TEAM Workshop benchmark 31).


Project partners were not deposited.

Publikationen & Patente

Nach oben