Supercomputer algorithms detected the spread of cancer to lymph nodes in women with breast cancer as well as or better than pathologists.

Digital imaging of tissue sample slides for pathology has become possible in recent years because of advances in slide scanning technology. Artificial intelligence, where computers learn to do tasks that normally require  human intelligence, has potential for making diagnoses. Using supercomputer algorithms to analyze digital pathology slide images could potentially improve the accuracy and efficiency of pathologists.

 
Researchers competed in an international challenge in 2016 to produce computer algorithms to detect the spread of breast cancer by analyzing tissue slides of sentinel lymph nodes, the lymph node closest to a tumor and the first place it would spread. The performance of the algorithms was compared against the performance of a panel of 11 pathologists participating in a simulation exercise.

Authors: Babak Ehteshami Bejnordi, M.S., Radboud University Medical Center, Nijmegen, the Netherlands and coauthors

Results:

  • Some computer algorithms were better at detecting cancer spread than pathologists in an exercise that mimicked routine pathology workflow.
  • Some algorithms were as good as an expert pathologist interpreting images without any time constraints.

Study Limitations: The test data on which algorithms and pathologists were evaluated are not comparable to the mix of cases pathologists encounter in clinical practice.

Study Conclusions: Supercomputer algorithms detected the spread of cancer to lymph nodes in women with breast cancer as well as or better than pathologists. Evaluation in a clinical setting is required to determine the benefit of using artificial intelligence in pathology to detect cancer requires.

The editorial, "Deep Learning Algorithms for Detection of Lymph Node Metastases From Breast Cancer," by Jeffrey Alan Golden, M.D., Brigham and Women's Hospital, Boston

The study, "Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes," by Tien Yin Wong, M.D., Ph.D., Singapore National Eye Center, Singapore, and coauthors.

Follow a link to this report here.