Spontaneous activity in peripheral sensory neurons has been associated with pain in people with chronic neuropathic and musculoskeletal conditions. It can also be observed in rodent models of chronic pain, where it has traditionally been studied by making electrophysiology recordings from nerve fibres emerging from the dorsal root ganglion. However, these experiments are technically challenging, low-throughput, and prone to bias, with small diameter fibres being most at risk of being under sampled. More recently, in vivo calcium imaging has been adopted to study sensory neuron activity in rodent models of pain. As this method becomes more widely used in the preclinical field, the need arises for a standardised approach to facilitate data analysis and allow comparisons between datasets. Here we provide a python-based package for the analysis of typical in vivo calcium imaging experiments in the pain field i.e. assessment of responses to evoked stimuli but also spontaneous activity, which we demonstrate on recordings performed in the fourth lumbar dorsal root ganglion of mice expressing the calcium sensor GCaMP6s in peripheral neurons. For the analysis of spontaneous activity, we expand on a previously published machine learning protocol and evaluate the robustness of its predictions in multiple pain models. We discuss challenges faced when using this approach (e.g. movement artefacts, prep stability) and their impacts on robustly and reproducibly identifying spontaneous activity. The code will be made available on GitHub, together with a Google Collaboratory Notebook to allow other researchers to test our analysis tool on their own data.