Novel computer-aided systems for interpreting immu- nohistochemistry (IHC) results in breast cancer based on deep learning algorithms: A systematic review

  • Sasan Salehi Nezamabadi Cancer Biology Research Center, Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran,
  • Haniyeh Rafiepoor Cancer Biology Research Center, Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran,
  • Mohammad Amin Barati School of Mechanical Engineering, University of Tehran, Tehran, Iran,
  • Elham Angouraj Taghavi Student Research Committee, School of Medicine, Shahroud University of Medical Sciences, Iran,
  • Golnar Khorsand School of Medicine, Tehran University of Medical Sciences, Tehran, Iran
  • Parsa Mirzayi School of Medicine, Tehran University of Medical Sciences, Tehran, Iran
  • Ali Taheri School of Mechanical Engineering, University of Tehran, Tehran, Iran,
  • Behzad Amanpour-Gharaei Cancer Biology Research Center, Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran,
  • Saman Asadi Islamic Azad university science and research branch, Tehran, Iran
  • Seyed-Ali Sadegh-Zadeh Department of Computing, School of Digital, Technologies and Arts, Staffordshire University, Stoke-on-Trent, ST2 4DE, UK,
  • Saeid Amanpour Cancer Biology Research Center, Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran,
Keywords: breast cancer, deep learning, computer-aided systems, IHC.

Abstract

Breast cancer is a prevalent disease worldwide and the accurate diagnosis and prog-nosis of breast cancer are essential for the development of effective treatment plans.Pathology remains the gold standard for diagnosis and prognosis but with limita-tions such as time-consuming manual scoring and some error-prone results. Re-cently, deep learning techniques, especially convolutional neural networks (CNN),have been proposed for the interpretation of immunohistochemistry (IHC) resultsin breast cancer. The objective of this systematic review is to critically assess the ex-isting literature on computer-aided systems for the interpretation of IHC results inbreast cancer based on deep learning algorithms. We included studies with modelsthat use novel approaches such as deep learning for quantitative measurements ofimmunohistochemically stained Ki-67, ER, PR, and HER2 images. We systematical-ly searched PubMed, Scopus, and web of science up to September 2022. 15 studies(seven HER2, seven Ki67, and one ER/PR scoring studies) met our inclusion criteria.Various AI-based assays have been developed for different applications in breast pa-thology, including diagnostic and prognostic applications, as well as predictive valuesand responses to treatment. These algorithms have shown promise in improving theaccuracy of breast cancer diagnosis and prognosis. It is essential to consider the dif-ferences in training and inter-observer variability while designing tools, and there isan urgent need to integrate the detection and analysis of various biomarkers at thesame place and time to facilitate the formation of patients’ reports and treatment.

Published
2025-05-25
Section
Articles