Leveraging Machine Learning for Advanced Biomedical Imaging: Insights from Speckle Pattern Analysis

In this chapter we present the usage of back scattered laser light for biomedical imaging and sensing. The back scattered laser light forms a self-interference pattern called secondary speckle pattern. By applying various machine learning approaches to process the spatial and the temporal variations in the distribution of those patterns, various biomedical modalities are extracted. Specifically, we give several applicable examples: we analyze the temporal fluctuations of the speckle patterns collected from the sclera to perform remote sensing of blood oxygen saturation. Then, we use the spatial distribution of the speckle patterns to perform biometric authentication and identification. Finally, we use the spatial–temporal variations of those patterns collected from different regions of the brain in order to detect stimulation of different cortexes associated with human senses. We end the chapter by addressing the challenges and the future perspectives related to the proposed novel sensing methodology and then conclude the chapter.