An adaptive algorithm for detection of multiple - type , positively stained nuclei in IHC images with minimal prior information : Application to OLIG 2 staining for gliomas

In this study, we propose a method to detect and segment the oligodendrocytes and gliomas in OLIG2 immunoperoxidase stained tissue sections. In general, glioma cells and oligodendrocytes mostly differ in shape and size within the tissue slide. In Olig2 stained tissue images, gliomas are represented with irregularly shaped nuclei with varying sizes and brown shades. On the other hand, oligodendrocytes have more regular round nuclei shapes and smaller in size when compared to glioma cells found in oligodendroglioma, astrocytomas, or oligoastrocytomas. In our algorithm, the first task is to detect the OLIG2 positive cell regions within the image. The second task is to segment each cell nucleus and count the number of cell nuclei. However, the cell nuclei belonging to glioma cases have particularly irregular nuclei shapes and form cell clusters by touching or overlapping with each other. In addition to this clustered structure, the shading of the brown stain and the texture of the nuclei differ slightly within a tissue image. The final step of the algorithm is to classify glioma cells versus oligodendrocytes. Our proposed method starts with color segmentation to detect positively stained cells followed by the classification of single individual cells and cell clusters by Kmeans clustering. Detected cell clusters are segmented with the H-minima based watershed algorithm. The novel aspects of our work are: 1) the detection and segmentation of multiple-type, positively-stained nuclei by incorporating only minimal prior information; and 2) adaptively determining clustering parameters to adjust to the natural variation in staining as well as the underlying cellular structure while accommodating multiple cell types in the image. Performance of the algorithm to detect individual cells is evaluated by sensitivitiy and precision metrics. Promising results (90% sensitivity and 82% precision) were achieved with a preliminary dataset, including four tissue slides with ground truth markings by two pathologists. 1Description of Purpose The purpose of this study is to develop an algorithm that provides reliable and consistent cell nuclei detection and counting from OLIG2-stained tissues, especially for touching cell clusters with different cell sizes, shapes and texture. OLIG2 is a basic helix loop helix transcription factor required for the specification of motor neurons, oligodndrocytes, and is expressed in a subpopulation of progenitor cells in rodent brains (reviewed by Meijer et al [4]). Its transcriptional function is required for tumorigenesis in a genetically relevant mouse model of adult human gliomas that commonly show activation of EGF signaling and mutation of the tumor suppressor INK4a/ARF[5]. OLIG2 expression patterns localize to the cell nucleus of astrocytomas and oligodendrogliomas in adult [6] and pediatric patients[7], regardless of the tumor’s WHO grade. OLIG2 has emerged as a very useful diagnostic marker and has shown strong immunoreactivity even in rare glioblastoma variants, such as small cell glioblastoma and gliosarcoma [8]. In particular, several diagnostic challenges in brain tumor histopathology are resolved by OLIG2 immunohistochemiscal studies. For example, OLIG2 is positive in oligodendrogliomas and negative in central neurocytoma [9] and is predominantly negative in ependymomas [7]. A further diagnostic utility for OLIG2 is in the distinction between reactive microglia, which is CD68-positive and OLIG2-negative, from infiltrating astrocytoma, which is typically OLIG2-positive and CD68-negative. This distinction is particularly helpful when evaluating the pathology of patients that experienced a prior surgical resection for glioblastoma that have shown, upon follow-up MRI imaging, new areas with gadolinium enhancement. These new areas of gadolinium enhancement may represent true tumor progression or pseudo-progression (a treatment reaction). True tumor progression indicates treatment failure and requires treatment change, whereas pseudoprogression permits conservative management. The evaluation of these tissues is challenging because both reactive processes and recurrent tumor processes may show morphological overlap on the H&E stain. In particular, astrocytoma cell nuclei and microglial cell nuclei show significant morphological overlap. In this context, diffusely positive OLIG2 immunohistochemistry in glioma cells helps identify pathologies representing patients that have experienced a tumor recurrence. A challenge in pathological interpretation of such cases rests in the distinction between non-neoplastic oligodendrocytes, which also express OLIG2, and recurrent glioblastoma.