Rail inclination is a well-known important track design parameter. It may have a measurable influence on the running dynamic behaviour of railway vehicles, as it affects equivalent conicity. Their effects are clearly visible when training Machine Learning (ML) algorithms for different purposes. This has been observed in on-going research regarding the detection of rail alignment using computer vision for in-service condition-monitoring. This paper briefly summarises the condition-monitoring research, and goes into detail regarding the effects of inclination and conicity explained from a vehicle dynamics viewpoint.