Abstract Datasets and codebase used to create the electron micrographs in Figures 2 and 3 of the paper "Shot noise-mitigated secondary electron imaging with ion count-aided microscopy" are presented. Each dataset consists of two time series of measured secondary electron detector and beam scan voltages. In the paper, these voltages were used to create SE images using the codebase. Methods The dataset was collected on 2 channels of a 4-channel, 16-bit, 100 MHz A/D converter. The signals are sampled at 10 nanoseconds (i.e. 100 million samples per second). A total of ~2.5 seconds of data were collected in each dataset, so the dimension of each dataset is approximately 250 million X 2 . The A/D converter automatically scaled raw values to voltages. The datasets are saved as matlab .m files. The beam current used for all acquisition was ~0.1 picoammeters, and the pixel dwell time for each dataset was 2 microseconds. The corresponding images have a size of 512 X 512 pixels. Description of the data and file structure Datasets used to create the images in Figures 2 and 3 of the manuscript "Shot noise-mitigated secondary electron imaging with ion count-aided microscopy" are presented here. The figure 2 dataset files are titled "Figure2_data1.m" through "Figure2_data4.m", and figure 3 dataset files are titled "Figure3_data1.m" through "Figure3_data4.m" . All data was acquired at Rutgers University. Both images have a resolution of 512 X 512 pixels. Each dataset consists of two synchronized time-series: the voltage signal from the Everhart-Thornley secondary electron (SE) detector and the voltage signal from the horizontal (slow) beam scan coil of a Zeiss Nanolab helium-ion microscope. The signals are sampled at 10 nanoseconds (i.e. 100 million samples per second). The dataset was collected on 2 channels of a 4-channel, 16-bit, 100 MHz A/D converter. A total of ~2.5 seconds of data were collected in each dataset, so the dimension of each dataset is approximately 250 million X 2 . The first channel contains the scan voltage data, and the second channel contains SE detector voltage data. The scan coil signal contains a series of sawtooths between ~ -4.5 and 4.5 V. Each complete sawtooth corresponds to the ion beam scanning across one horizontal row in the sample. An image frame consists of 512 such row scans. The detector voltage signal contains several pulses with voltages between 0 and ~1.8 V. Each pulse corresponds to SE detection events from an incident ion. The pulses have a FWHM of about 160 nanoseconds. The beam current used for all acquisition was ~0.1 picoammeters, and the pixel dwell time was 2 microseconds. After each image frame, the scan coil voltage remains at -4.5 V for a few microseconds, before restarting scanning. Each dataset contains 4 complete image frames, and there are 4 datasets for each image. Therefore, a total of 16 image frames are used to create the images in Figures 2 and 3. The start and stop of data collection was not synchronized to the start or stop of a frame. Therefore, each dataset also has partial frames near the start and end. These partial frames are ignored, and only complete frames are used to form the images in Figures 2 and 3. The total data used to create each image was broken up into 4 datasets because of the limited memory of the A/D converter. Each horizontal scan is broken up into 512 steps, corresponding to pixels on one line. The readout noise on the beam scan signal makes these steps hard to see. The noise can be reduced and the steps visualized by averaging over many line scan voltages (i.e. many sawtooths). Code/Software for processing datasets The Github codebase for processing the datasets is available here.