Pathologists who used Ibex’s Galen Breast for cancer diagnosis obtained very high accuracy in a multi-site clinical study.
Ibex Medical Analytics has shared outstanding outcomes for its Galen Breast solution in diagnosing multiple cancer types and announced further expansion of its AI portfolio for breast cancer diagnosis. The CE-Marked AI solution is generally available as Ibex partners with laboratories, hospitals and health systems across Europe to deploy Galen Breast into their routine workflow, supporting improved quality and efficiency in the diagnosis of breast biopsies.
Breast cancer is the most common malignant disease in women worldwide, with over two million new cases each year. As such, accurate and timely diagnosis is key to guiding treatment decisions by oncologists and improving patient survival rates.
Galen Breast demonstrated excellent outcomes in a blinded, multi-site clinical study at Institut Curie in France and Maccabi Healthcare Services in Israel. The study evaluated the performance of pathologists who used Ibex AI for diagnosing breast biopsies and compared them to pathologists who used only a microscope across multiple types of breast cancer including invasive and in-situ carcinoma as well as rare subtypes, such as metaplastic, mucinous, and other types of breast cancer.
The study results showed very high accuracy and utility of Galen Breast across multiple scanning and staining platforms, and established its potential for improving the quality of diagnosis, compared to using a microscope alone. The full study results will be presented by Judith Sandbank, MD, Director of the Pathology Institute at Maccabi Healthcare Services and one of the principal investigators in the study, at the European Congress of Pathology which takes place in Basel, Switzerland, between 3-7 September.
Previous studies on Galen Breast established its AI algorithm’s accuracy in detecting cancer, distinguishing between multiple subtypes such as ducal and lobular carcinomas, grading DCIS (ductal carcinoma in-situ) and identifying rare tumours. Moreover, the solution successfully detected clinically important cancer-related and non-cancer features, including tumour infiltrating lymphocytes, angiolymphatic invasion, microcalcifications and more. Galen also enables automated pre-ordering of the breast immunohistochemistry (IHC) panel and other tests which may help shorten the turnaround time for diagnosis of cancer cases, maximising efficiency gains for laboratories and enabling patients to start treatment earlier. Ibex is also working to expand its technology to support improved prognosis of breast cancer, by providing automated quantification of HER2, Ki67, ER and PR receptors in immunohistochemistry stained slides as part of its product portfolio. These innovative AI algorithms are intended to assist pathologists in their diagnosis which may help further enhance diagnostic efficiency and enable more accurate and objective scoring of breast biomarkers, allowing for improved treatment decisions and patient care.
“We are impressed with the successful study outcomes and the performance of Galen Breast, that was evaluated in a diagnostic setting which is identical to how our pathologists review cases in their daily routine,” said Anne Vincent-Salomon, MD, Director of Pathology at Institut Curie and one of the principal investigators in the study. “Our team demonstrated that when pathologists use Ibex’s AI technology they achieve very high accuracy levels in diagnosing breast cancer over a broad range of subtypes, with higher quality than when using a microscope alone. With these results, and as more and more laboratories transition to digital pathology workflows, I look forward to seeing Artificial Intelligence broadly adopted in the field.”
