<p>Human epidermal growth factor receptor 2 (HER2), a transmembrane tyrosine kinase receptor encoded by the <italic>ERBB2</italic> gene on chromosome 17q12, is a predictive and prognostic biomarker in invasive breast cancer (BC). Approximately 20% of BC are HER2-positive as a result of <italic>ERBB2</italic> gene amplification and overexpression of the HER2 protein. Quantification of <italic>HER2</italic> is performed routinely on all invasive BCs, to assist in clinical decision making for prognosis and treatment for <italic>HER2</italic>-positive BC patients by manually counting gene signals. We propose an automated system to quantify the <italic>HER2</italic> gene status from chromogenic <italic>in situ</italic> hybridization (CISH) whole slide images (WSI) in invasive BC. The proposed method selects untruncated and nonoverlapped singular nuclei from the cancer regions using color unmixing and machine learning techniques. Then, <italic>HER2</italic> and chromosome enumeration probe 17 (CEP17) signals are detected based on the RGB intensity and counted per nucleus. Finally, the <italic>HER2</italic>-to-CEP17 signal ratio is calculated to determine the <italic>HER2</italic> amplification status following the ASCO/CAP 2018 guidelines. The proposed method reduced the labor and time for the quantification. In the experiment, the correlation coefficient between the proposed automatic CISH quantification method and pathologist manual enumeration was 0.98. The <italic>p</italic>-values larger than 0.05 from the one-sided paired <italic>t</italic>-test ensured that the proposed method yields statistically indifferent results to the reference method. The method was established on WSI scanned by two different scanners. Through the experiments, the capability of the proposed system has been demonstrated.</p>
Whole slide imaging (WSI) scanner scans pathological specimens to produce digital slides to use in pathology practice, research and computational pathology which enables monitor-based diagnosis and image analysis. However, the scanned image is sometimes insufficient in quality such as focusing-error and noise. Therefore, a quality evaluation method is obligatory for practical use of WSI system. In previous work, referenceless quality evaluation technique was proposed for this purpose but some artefacts (i.e. tissue-fold, air-bubble) in slide would also be detected as false positives, while they are useless. In this paper, we proposed a method for the practical system to assess WSI quality with eliminating false detection due to the artefacts. Firstly, support vector machine (SVM) was utilized for detecting ROIs with artefacts and then the image quality was evaluated excluding detected ROIs. Through the experiments, the effectiveness of proposed system has been demonstrated.