Open biopsy

Open biopsy

Mammographic Screening and Breast Cancer Management – Part 1

Published in Sandeep Reddy, Artificial Intelligence, 2020

Currently, most published literature in the field relies on supervised methodologies, those with a definitive ‘label’ (‘target’ or ‘y’). These are typically in the forms of human assessment of likelihood of malignancy from mammograms (BI-RADS score (American College of Radiology, 2013) – see Table 4.4) or other images, or pathological examination (macroscopic and microscopic). In practice, some cancerous tissue is required to confirm the presence of cancer with microscopy and staining and tissue processing techniques. This tissue may be obtained with various biopsy methods including a fine-needle aspiration, core biopsy, and open biopsy by a surgeon.

Multi-scale feature fusion for histopathological image categorisation in breast cancer

View Article

Journal Information

Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2023

Yingping Wang, Xing Deng, Haijian Shao, Yingtao Jiang

This paper utilises the BreakHis dataset (Spanhol et al. 2015), which is available on the Kaggle website and is typically used to investigate the classification of breast cancer using pathological images. The dataset contains 7909 pathological images of 82 patients, including 2480 benign samples and 5429 malignant ones. Surgical open biopsy and haematoxylin and eosin (HE) staining were used to collect samples from breast tissue biopsy specimens. After immunohistochemical evaluation, the final microscopic images of breast tumour tissue were obtained by using different magnifications. The distribution of the quantity of samples is shown in Table 2, and each sample is a colour image with a size of 700 × 460 pixels.

Explore chapters and articles related to this topic

Mammographic Screening and Breast Cancer Management – Part 1

Multi-scale feature fusion for histopathological image categorisation in breast cancer