AI in 2D Mammography: Improving Breast Cancer Screening Accuracy DOI Creative Commons

Sebastian Ciurescu,

Simona Cerbu,

Ciprian Nicuşor Dima

et al.

Medicina, Journal Year: 2025, Volume and Issue: 61(5), P. 809 - 809

Published: April 26, 2025

Background and Objectives: Breast cancer is a leading global health challenge, where early detection essential for improving survival outcomes. Two-dimensional (2D) mammography the established standard breast screening; however, its diagnostic accuracy limited by factors such as density inter-reader variability. Recent advances in artificial intelligence (AI) have shown promise enhancing radiological interpretation. This study aimed to assess utility of AI lesion classification 2D mammography. Materials Methods: A retrospective analysis was performed on dataset 578 mammographic images obtained from single radiology center. The consisted 36% pathologic 64% normal cases, partitioned into training (403 images), validation (87 test (88 images) sets. Image preprocessing involved grayscale conversion, contrast-limited adaptive histogram equalization (CLAHE), noise reduction, sharpening. convolutional neural network (CNN) model developed using transfer learning with ResNet50. Model performance evaluated sensitivity, specificity, accuracy, area under receiver operating characteristic (AUC-ROC) curve. Results: achieved an overall 88.5% AUC-ROC 0.93, demonstrating strong discriminative capability between cases. Notably, exhibited high specificity 92.7%, contributing reduction false positives improved screening efficiency. Conclusions: AI-assisted holds potential enhance reducing false-positive findings. Although further optimization required minimize negatives. Future efforts should aim improve incorporate multimodal imaging techniques, validate results across larger, multicenter prospective cohorts ensure effective integration clinical workflows.

Language: Английский

Tumor-Agnostic Therapies in Practice: Challenges, Innovations, and Future Perspectives DOI Open Access
Sulin Wu, Rajat Thawani

Cancers, Journal Year: 2025, Volume and Issue: 17(5), P. 801 - 801

Published: Feb. 26, 2025

This review comprehensively analyzes the current landscape of tumor-agnostic therapies in oncology. Tumor-agnostic are designed to target specific molecular alterations rather than primary site tumor, representing a shift cancer treatment. We discuss recent approvals by regulatory agencies such as FDA and EMA, highlighting that have demonstrated efficacy across multiple types sharing common alterations. delve into trial methodologies underpin these approvals, emphasizing innovative designs basket trials umbrella trials. These present unique advantages, including increased efficiency patient recruitment ability assess drug diverse populations rapidly. However, they also entail certain challenges, need for robust biomarkers complexities requirements. Moreover, we examine promising prospects developing rare cancers exhibit targets typically associated with more prevalent malignancies. By synthesizing insights, this underscores transformative potential It offers pathway personalized treatment transcends conventional histology-based classification.

Language: Английский

Citations

0
AI in 2D Mammography: Improving Breast Cancer Screening Accuracy DOI Creative Commons

Sebastian Ciurescu,

Simona Cerbu,

Ciprian Nicuşor Dima

et al.

Medicina, Journal Year: 2025, Volume and Issue: 61(5), P. 809 - 809

Published: April 26, 2025

Background and Objectives: Breast cancer is a leading global health challenge, where early detection essential for improving survival outcomes. Two-dimensional (2D) mammography the established standard breast screening; however, its diagnostic accuracy limited by factors such as density inter-reader variability. Recent advances in artificial intelligence (AI) have shown promise enhancing radiological interpretation. This study aimed to assess utility of AI lesion classification 2D mammography. Materials Methods: A retrospective analysis was performed on dataset 578 mammographic images obtained from single radiology center. The consisted 36% pathologic 64% normal cases, partitioned into training (403 images), validation (87 test (88 images) sets. Image preprocessing involved grayscale conversion, contrast-limited adaptive histogram equalization (CLAHE), noise reduction, sharpening. convolutional neural network (CNN) model developed using transfer learning with ResNet50. Model performance evaluated sensitivity, specificity, accuracy, area under receiver operating characteristic (AUC-ROC) curve. Results: achieved an overall 88.5% AUC-ROC 0.93, demonstrating strong discriminative capability between cases. Notably, exhibited high specificity 92.7%, contributing reduction false positives improved screening efficiency. Conclusions: AI-assisted holds potential enhance reducing false-positive findings. Although further optimization required minimize negatives. Future efforts should aim improve incorporate multimodal imaging techniques, validate results across larger, multicenter prospective cohorts ensure effective integration clinical workflows.

Language: Английский

Citations

0