Accurate post-mortem alignment for Focused Ion Beam and Scanning Electron Microscopy (FIB-SEM) tomography

Abstract

Drifts in the three directions (X, Y, Z) during the FIB-SEM slice-and-view tomography is an important issue in 3D-FIB experiments which may induce significant inaccuracies in the subsequent volume reconstruction and further quantification of morphological volume parameters of the sample microstructure. Cross-correlation is frequently applied directly to the cross-section image series for aligning FIB sliced images. This solution is hazardous and can be flawed as it has been easily demonstrated by a dedicated test experiment. As a result, a novel aligning procedure based on the quantification of the topography of the sample surface has been developed. This new approach will be compared to the common cross-correlation methods, as well as another approach consisting in using an artificial reference marker fabricated during the FIB procedure. All these methods will then be discussed in terms of accuracy and liability.

Okapi-EM: A napari plugin for processing and analyzing cryogenic serial focused ion beam/scanning electron microscopy images

“…While there have been efforts at improving image acquisition hardware and developing real-time correction (25)(26)(27) , these methods are limited by the need to ensure that the sample is not overexposed during data collection. Therefore, they do not provide fine alignment correction and require the use of subsequent alignment software (24) . Several software packages/plugins are available for this task, such as the closed-source and pay-for-service Amira by ThermoFisher (28) and open-source options such as ImageJ plugins Linear Stack Alignment with SIFT (29) and TrakEM2 (30) .…”

“…During this process, a number of issues can cause drift or misalignment in the image stack, such as the movement of the sample due to mechanical stress, small temperature variations and slight, iterative, stage movements going from milling to imaging positions that accumulate over time ( 23 ) . Inaccurate placement of the milling area by the operator or software may also lead to the observation of shearing during subsequent imaging ( 24 ) . These misalignments may further be amplified by factors including charging effects and instabilities caused by external disturbance ( 24 ) .…”

Okapi-EM – a napari plugin for processing and analysing cryogenic serial FIB/SEM images

“…During this process, a number of issues can cause drift or misalignment in the image stack, such as the movement of the sample due to mechanical stress, small temperature variations and slight, iterative, stage movements going from milling to imaging positions that accumulate over time (20) . Inaccurate placement of the milling area by the operator or software may also lead to the observation of shearing during subsequent imaging (21) . These misalignments may further be amplified by factors including charging effects and instabilities caused by external disturbance (21) .…”

“…Inaccurate placement of the milling area by the operator or software may also lead to the observation of shearing during subsequent imaging (21) . These misalignments may further be amplified by factors including charging effects and instabilities caused by external disturbance (21) . As a result, an important element in the alignment of the resultant SEM images is the shearing or skewing along the direction perpendicular to the ion beam (or other layer removal option) trajectory (20) , which is particularly significant along the slow scan direction.…”

“…While there have been efforts at improving image acquisition hardware and developing real-time correction (22)(23)(24) , these methods are limited by the need to ensure that the sample is not overexposed during data collection. Therefore, they do not provide fine alignment correction and require the use of subsequent alignment software (21) . Several software packages/plugins are available for this task, such as the closed-source and pay-for-service Amira by ThermoFisher (25) and open-source options such as ImageJ plugins Linear Stack Alignment with SIFT (26) and TrakEM2 (27) .…”