Abstract: Method and system are disclosed for determining sample composition from spectral data acquired by a charged particle microscopy system. Chemical elements in a sample are identified by processing the spectral data with a trained neural network (NN). If the identified chemical elements not matching with a known elemental composition of the sample, the trained NN is retrained with the spectral data and the known elemental composition of the sample. The retrained NN can then be used to identify chemical elements within other samples.

Abstract: The invention relates to a method of examining a sample using a charged particle microscope, comprising the steps of providing a charged particle beam, as well as a sample, and scanning said charged particle beam over said sample. A first detector is used for detecting emissions of a first type from the sample in response to the beam scanned over the sample. Using spectral information of detected emissions of the first type, a plurality of mutually different phases are assigned to said sample. An image representation of said sample is provided, wherein said image representation contains different color hues. The color hues are selected from a pre-selected range of consecutive color hues in such a way that the selected color hues comprise mutually corresponding intervals within said pre-selected range of consecutive color hues.

Abstract: The invention relates to a method of examining a sample using a charged particle microscope, comprising the steps of providing a charged particle beam, as well as a sample, and scanning said charged particle beam over at least part of said sample. A first detector is used for obtaining measured detector signals corresponding to emissions of a first type from the sample at a plurality of sample positions. According to the method, a set of data class elements is provided, wherein each data class element relates an expected detector signal to a corresponding sample information value. The measured detector signals are processed, and processing comprises comparing said measured detector signals to said set of data class elements; determining at least one probability that said measured detector signals belong to a certain one of said set of data class elements; and assigning at least one sample information value and said at least one probability to each of the plurality of sample positions.