Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
This paper introduces a novel deep learning approach combining multiscale convolutions with skip connections and independent supervision to effectively localize candidate Kikuchi regions in RHEED images, overcoming the challenge of obscuring diffraction streaks. The method achieves a Kikuchi intersection-over-union (IoU) of $0.6290 \pm 0.0137$ on laboratory test images and $0.5014 \pm 0.0336$ on a clear-feature subset of public chalcogenide images, demonstrating its efficacy in distinguishing Kikuchi features from intense diffraction patterns. This work provides a robust foundation for automated analysis of RHEED data, enabling precise measurement of line positions, intersections, and local intensity.
A novel RHEED image analysis technique disentangles diffraction streaks from Kikuchi lines, achieving $0.63$ IoU and enabling precise geometric analysis previously hindered by intense scattering.
Kikuchi lines and bands in reflection high-energy electron diffraction (RHEED) carry information on crystal geometry and electron scattering, but are often obscured by intense diffraction streaks. We combine multiscale convolution with spatial detail from skip connections and independent supervision that allows overlapping regions to localize streaks and candidate Kikuchi regions separately. Using manual polygon annotations made before model-assisted editing, the Kikuchi intersection-over-union (IoU) on 29 laboratory test images from separate growth batches was $0.6290 \pm 0.0137$. We then performed supervised adaptation to public chalcogenide images and stratified images by Kikuchi visibility using the median ratings of three observers who rated the images separately, two of whom rated with predictions hidden. With all 244 training images, IoU for the clear-feature group of 10 test images was $0.5014 \pm 0.0336$, whereas ambiguous images gave lower values. With 100 training images, clear-group IoU was $0.5171 \pm 0.0073$; expanding the training set also reduced responses on images without target features. Reported uncertainties are sample standard deviations across four laboratory runs or three adaptation runs. Public-image adaptation used reference masks of mixed provenance, including prediction-derived drafts. The method converts visual cues into inspectable spatial regions, providing a basis for analysis of line positions, intersections, and local intensity.