Dual-path convolutional neural network using micro-FTIR imaging to predict breast cancer subtypes and biomarkers levels: estrogen receptor, progesterone receptor, HER2 and Ki67

The study focuses on using deep learning to predict breast cancer subtypes and biomarker levels. Four biomarkers – Estrogen Receptor (ER), Progesterone Receptor (PR), HER2, and Ki67 – are used to classify subtypes such as Luminal A, Luminal B, HER2 subtype, and Triple-Negative Breast Cancer (TNBC). The study introduces a novel 2D deep learning approach, CaReNet-V2, developed to classify breast cancer and predict biomarker levels. The technique is potentially helpful for breast cancer biopsies evaluation, standing out as a screening analysis technique and helping to prioritize patients.

Publication date: 24 Oct 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2310.15099