Abstract Many scientific efforts have been devoted to understanding the brain functions, and the relevance of its dynamics during development, aging, and in diseased conditions. Alterations of the brain functions can result from multifactorial processes and be reflected by various biomarkers. The ability to quantify these changes at multiple scales will improve our understanding of brain anatomical and functional architectures, and the relations between these networks in both normal and diseased conditions. By virtue of the rich optical absorption contrast, high spatial and temporal resolutions, and relatively deep penetration, photoacoustic tomography (PAT) is a promising imaging modality that can address the limitations of functional magnetic resonance imaging (fMRI) and functional optical imaging for humans. There are major challenges associated with the skull that need to be addressed before PAT can be adopted to functional human brain imaging. The human skull severely distorts the photoacoustic (PA) signals, giving rise to suboptimal images. The standard approach to mitigate the distortion is to use CT-scans of the skull in concert with an image reconstruction method that considers the heterogeneity of the speed of sound. In addition, conventional ultrasound arrays are rigid and therefore cannot conform to the curvature of the skull. This creates a major challenge for efficient tissue-transducer coupling. Here, we propose to develop an ultrasound image- based approach for skull phase correction, which would eliminate the need of CT scans as well as image co- registration. To address the tissue-transducer coupling issue, we propose to use a stretchable matrix array that can seamlessly conform to the skull?s non-developable surface. Optical fibers will be integrated with the array to provide contact light delivery. The specific tasks to be completed during this grant period are: First, develop, fabricate, and characterize a 256-element stretchable sparse matrix array, second, numerically verify our phase correction and imaging algorithms, third, assess the array and algorithms experimentally using head-mimicking phantoms. At the end of this project we would have confirmed the performance of the stretchable matrix array and the accuracy of the phase correction and imaging algorithms. The preliminary results obtained from this proof-of- concept project will provide the basis upon which we can expand the size of the array and the number of elements to 1024. Such an array will offer a wide field of view for probing human brain functions and will be evaluated on human subjects.