SUMMARY Repair and regeneration of articular cartilage remains a clinical and scientific challenge. Reliable assessment tools for evaluating outcomes of cartilage repair are critical for both refinement of existing methods and development of new techniques. Histological analysis of biopsies is the gold standard for repair assessment; however, biopsies are invasive procedures and therefore limited in clinical evaluation and studies of the cartilage repair. The common medical imaging methods, such as x-ray radiography, computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound, can perform imaging non-destructively; however, their spatial resolutions are not sufficient to reveal the complex cell and matrix architecture of articular cartilage. Though some imaging techniques are non-destructive and can image tissue, such as arthroscopy, laser scanning confocal arthroscopy (LSCA) and optical coherence tomography (OCT), they are all performed as surgical procedures because of using thick endoscopy probes. The inability to perform clinical in vivo imaging on cartilage tissue with high spatial resolution remains a problem. To solve the problem, we propose to develop a nonlinear optical microscopy (NLOM) based endomicroscopy system for assessment of cartilage repair in vivo. In NLOM imaging of cartilage tissue, second harmonic generation (SHG) signal provides high- resolution information of fibril organization of collagen while two-photon excited fluorescence (TPEF) enables visualization of chondrocytes and elastin fibers. However, the current cartilage NLOM imaging devices all use tabletop systems that are too bulky to be used directly in clinical applications. Thus, our long-term goal is to translate this technology into a clinical imaging tool for assessment of articular cartilage repair and treatment at the cellular level. In this application, we will focus on three specific aims as follows. (1) We will determine the efficacy of using NLOM to evaluate morphological changes of articular cartilage. Using spontaneous OA guinea pigs as an articular cartilage pathology model, we will test if NLOM imaging can detect the quantitative differences among the early stages of OA cartilage tissues. (2) We will design and build a compact and high- speed NLOM imaging system with a thin rod objective as the imaging probe. A numerical simulation model will be developed to help optimize the system design. (3) With the developed endomicroscope, we will first evaluate its performance by performing a similar quantitative imaging study as described in Aim1 on excised cartilage tissues from guinea pigs with OA. We will then use the endomicroscope and tabletop system to perform a quantitative imaging study on a cartilage repair model to test if the endomicroscope can detect morphological differences between tissues in non-treated and microfracture treated defects. With the success of this study, we will be able to determine the usefulness and limitation for u...