With the broadening applications of spatial-omics technologies to resolve complex biological questions, there is a critical need to provide tools for single cell in situ analysis that empower biologists to interrogate the data. Cell segmentation and cell phenotyping are the two most crucial steps in biological tissue image analysis and often the results need to be visually validated by a biologist. These tools are commonly released as a package in a programming language, making the validation and usage difficult for a biologist. We have addressed this issue by developing a series of tools in the open-source image analysis software QuPath. These include a graphical user interface (GUI) for an existing cell segmentation tool DeepCell Mesmer, manual labelling tools for training machine learning models for cell phenotyping, and tools allowing for easier visualisation of phenotyping results. This poster will discuss the implementation of the tools and provide an example by analysing the tumour immune microenvironment of formalin-fixed paraffin-embedded (FFPE) tissues of a non-small cell lung cancer (NSCLC) cohort containing 89 patients. The images were scanned using the multiplex ion beam imaging time of flight (MIBI-ToF) technique with 38 markers.