Digital pathology is now underpinning biomarker and drug discovery, clinical diagnosis via companion diagnostics and provides valuable tools for the current move towards providing personal stratified medicine. The potential digital pathology has for giving better insight into many disease mechanisms is due to firstly it's digital (and therefore linkable/indexable) nature, and secondly the sheer amount of data that can now be generated by imaging hardware and image analysis methodologies. The work carried out in this thesis was an examination of the current barriers to verifying and exploiting the large amounts of information that is available via the use of digital pathology. The large size of whole slide images generated in digital pathology means that lossy image compression becomes a necessity due to storage limitations. Lossy image compression however discards data in a non-linear transformation, e.g. JPEG image compression, and therefore changes the values of pixels in the images. As image analysis algorithms use these values to perform quantification and segmentation of regions within the image, the first piece of this thesis examines the affect image compression has on fundamental image calculations common in image analysis algorithms. Tissue imagery data is complex but the' visual pathological information provided is the basis of clinical diagnosis. These patterns are currently used to show the symptoms of disease. This thesis details a system that can make this information easily searchable using these patterns as the input criterion for the search. A system that may highlight tissue with specific malignant morphological properties could clearly provide a valuable resource in large clinical trials where manual analysis of specimens requires many work hours. The imaging information provided by digital pathology is not independent from other experimental and clinical data and the final piece of this thesis provides a framework for connecting clinico-pathological, genetic and imaging information in an attempt to provide a much more detailed and connected view of related diseases. This work highlights the considerations necessary in designing studies employing image analysis and underpins the future importance of digital pathology applications in modern tissue-based translational research.