Sahar Mansour: First prospective study in Egypt to evaluate AI in double reading for breast cancer screening
Sahar Mansour, Professor of Radiology in the Women’s Imaging Unit at Kasr El Ainy Hospital, Cairo University, shared a post on LinkedIn about a paper she co-authored with colleagues published in Egyptian Journal of Radiology and Nuclear Medicine:
“To date, no research has been conducted on this issue in Egypt, the Middle East, or Africa.
Our latest study at Baheya Foundation Center for early breast cancer detection and treatment delves into the effectiveness of combining AI with radiologists in the routine double reading of mammograms. to screen breast cancer.
We analyzed 32,822 screening mammograms, comparing the performance of two radiologists versus one radiologist paired with AI.
For screening mammograms, the Youden index (J) was 0.7395 for a cut-off AI scoring value of > 14% to set a recall not for cancer detection showing the specificity of 95.42%.
The mean AI score for malignant lesions was 68% (95% confidence interval, 66% to 70%)
The key strengths of our groundbreaking research:
- AI Independent Reader: We incorporated AI as a junior reader, tasked with sorting negative cases from those requiring recall. This innovative approach ensures that AI supports radiologists by identifying potential issues early on, streamlining the screening process.
- Expertise of Radiologists: A team of highly experienced radiologists, 15 to 30 years of expertise in breast imaging. Their profound knowledge and skill set provided a robust foundation for the study, ensuring accurate and reliable results.
- Comprehensive Prospective Study: This is the first prospective study to include biopsy results for suspicious and malignant lesions, along with a two-year follow-up for normal and benign cases. The availability of tissue sampling and extended follow-up allowed us to directly calculate sensitivity and negative predictive value, providing high certainty that no cancer was present in women who did not undergo biopsy.
- Centre of Excellence: The study was performed at a premier breast imaging center equipped with ultrasound and biopsy facilities within the same unit as the screening mammogram. This setup facilitated a seamless breast cancer screening experience, enabling interpretation, recall, and biopsy (if required) to be completed in a single session.
As we continue to explore the integration of AI in medical imaging, we are optimistic about the future of breast cancer screening and the lives it can save.”
Authors: Sahar Mansour, Enas Sweed, Mohammed Mohammed Mohammed Gomaa, Samar Ahmed Hussein, Engy Abdallah, Yassmin Mohamed Nada, Rasha Kamal, Ghada Mohamed, Sherif Nasser Taha, Amr Farouk Ibrahim Moustafa.
More posts featuring Sahar Mansour.
-
Challenging the Status Quo in Colorectal Cancer 2024
December 6-8, 2024
-
ESMO 2024 Congress
September 13-17, 2024
-
ASCO Annual Meeting
May 30 - June 4, 2024
-
Yvonne Award 2024
May 31, 2024
-
OncoThon 2024, Online
Feb. 15, 2024
-
Global Summit on War & Cancer 2023, Online
Dec. 14-16, 2023