Ibex Study Finds AI Highly Accurate in Cancer Identification

The latest study from Ibex Medical Analytics and the Institut Curie, published in the scientific journal Nature, demonstrates how AI implementation across pathology labs can significantly improve the outcome of cancer care.

Breast cancer is the most common malignant disease worldwide, with over 2.26 million new cases in 2020. The cumulative probability of a woman receiving at least one false-positive biopsy over 10 years is estimated to be between 4.8% and 9.4%, emphasizing the need for accurate, timely, and objective diagnosis of this disease. 

AI-based tools can support cancer detection and classification in breast biopsies ensuring rapid, accurate, and objective diagnosis.

Titled ‘Validation and real-world clinical application of an artificial intelligence algorithm for breast cancer detection in biopsies’, the study analyzed Ibex’s AI-based quality control solution for breast biopsy review on over 15,000 slides over a diverse cohort. The study found the AI algorithm achieved high accuracy in identifying various types of cancer and 51 different morphological features of breast tissue.

The algorithm also accurately identified invasive carcinoma and ductal carcinoma in situ (DCIS) with 93.79% specificity and 93.20% sensitivity, with the AI differentiating well between different grades and subtypes of breast cancer.

Join the HealthXL Webinar on ‘Retail Healthcare: Current and Future Opportunities for Digital Health’ on 26th January. Click here to Request to Join. 

Click here to read the complete study.