Imaging flow cytometry is an indispensable cell-analytic tool that provides multi-parametric measurements with high-dimensional feedback derived from single-cell images. It enables profound insights into cell signalling, co-localisation, cell-to-cell interaction and deoxyribonucleic acid studies. However, the traditional frame-based sensor adopted in imaging flow cytometry can inhibit its performance, bound by the triangle of imaging constraints: speed, resolution and sensitivity; increasing one parameter can lead to degradation in others. This trade-off correlation has fundamentally hindered the development and generalisation of imaging flow cytometry. In addition, the rich spatial information acquired through imaging flow cytometry has exceptional uses with machine learning models to automate cell analysis, gating and revealing rare cell events. Herein, we introduced a neuromorphic imaging cytometry approach to characterise cells with superior temporal resolution, data efficiency and fluorescence sensitivity. Taking advantage of data sparsity in neuromorphic vision, the proposed platform and curated dataset were combined with hybrid spiking neural network models to perform cell classification and enable in-depth analysis. To the best of our knowledge, our research enabled the first novelty of neuromorphic-enabled cytometry applications. This dissertation encompassed the entire research milestones including the initial conceptualisation of neuromorphic imaging cytometry with artificial particles; implementation of fundamental cytometric functions; object detection with biological cells; advanced machine learning cell analysis by a lightweight model on convoluted cell classes and morphologies. Combining the data sparsity in neuromorphic imaging with a lightweight hybrid spiking neural network model and operation platform, this paradigm can become a fundamental backbone for next-generation, machine learning-driven cytometry.