Raman spectroscopy is one of the most specific non-destructive optical techniques to identify the chemical composition of a sample. Further, there is great hope that in the future it may be a powerful biomedical imaging technique for in vitro or in vivo microscopy, providing molecular contrast without exogenous contrast agents.However, due to the small Raman cross-section, for many applications the acquisition is prohibitively slow. Techniques to solve this problem and to increase the Raman signal levels are coherent anti-Stokes Raman spectroscopy (CARS), surface enhanced Raman spectroscopy (SERS) and stimulated Raman spectroscopy (SRS). However, in many cases, they are currently not able to provide rapid, highly sensitive detection of an undistorted signal with a broad spectral coverage.The aim of the project is to investigate Fourier domain mode locked (FDML) lasers for the application to stimulated Raman detection. A variety of physical effects, unique to FDML lasers, enables strategies to substantially increase the Raman signal level. This can provide access to highly sensitive Raman spectroscopy and high speed Raman microscopy. The techniques to increase the detection sensitivity include concepts like single- and double-resonant enhancement cavities, high power fibre amplification, dynamic spectral zooming, advanced modulation schemes and parallel designs.The first part of the project addresses a comprehensive understanding of the underlying physical effects and how to increase the Raman signal by several orders of magnitude using these various strategies. The aim of the second part is to investigate, in how far these improved FDML based Raman systems can be applied to transient real time spectroscopy, analytical sensing, and Raman microscopy.