Research has demonstrated that the accuracy achieved through artificial intelligence algorithms is comparable to that of radiologists, and optimal results may be obtained by combining radiologist's assessments with artificial intelligence methods, particularly in the role of the first reader.
Abstract
Breast cancer represents a leading public health problem due to its high incidence, morbidity, and mortality. Significant geographic differences in disease burden are influenced not only by socioeconomic and behavioural factors but also by the resources of health systems, access to healthcare, and the quality of diagnostic programs. Modern mammography screening programs substantially contribute to early disease detection, while existing challenges include limited resources, restricted access and inequalities in healthcare, high costs, variability in result interpretation, and the need for additional and invasive diagnostic procedures. To address these challenges, artificial intelligence systems have been developed, employing complex algorithms, machine learning, and deep learning for the automated analysis of mammographic images. These systems can improve diagnostic accuracy, reduce reading time, lower false-negative rates, and decrease the need for repeat diagnostic procedures. Research has demonstrated that the accuracy achieved through artificial intelligence algorithms is comparable to that of radiologists, and optimal results may be obtained by combining radiologist's assessments with artificial intelligence methods, particularly in the role of the first reader. Key limitations of these methods include insufficient adaptability to diverse populations and the risk of over-reliance on algorithm results by less experienced radiologists. Beyond technical challenges, an important public health aspect concerns women's attitudes toward the use of artificial intelligence, as the level of trust in this technology may affect screening participation rates. The integration of artificial intelligence into screening programs requires careful evaluation of benefits and limitations, transparency in technology use, and assurance of human oversight. Proper implementation and public education can contribute to improved early diagnosis, greater screening accuracy, and more efficient use of health system resources.
This comprehensive review delves into the fundamental concepts of radiomics and AI, summarises their current applications in BC imaging, and explores their evolving roles in clinical practice, highlighting recent advances and presenting case studies demonstrating the clinical impact.
Priyanka Dutta· Karnataka Journal of Surgery· 0 citations
Introduction: Breast cancer remains the most frequently diagnosed malignancy and one of the leading causes of cancer mortality among women worldwide. Despite the implementation of organized mammography screening programmes, considerable disparities in mortality persist, particularly in lowand middle-income countries. Artificial intelligence (AI) has emerged as a promising tool for improving the accuracy, efficiency, and accessibility of breast cancer screening. To review current evidence regarding the role of artificial intelligence in breast cancer screening and early detection, with particular emphasis on its potential application within the healthcare system of the Republic of Serbia. Methodology: A comprehensive literature review was conducted using the MEDLINE, PubMed, and KOBSON databases. Literature published between 2012 and 2026 were identified using the keywords breast cancer, artificial intelligence, mammography, screening, and machine learning. After applying predefined inclusion criteria, 23 relevant publications were included in the final analysis. Topic: Current evidence demonstrates that AI-assisted mammography achieves diagnostic performance comparable to or exceeding that of experienced breast radiologists, while substantially reducing reading workload. Prospective clinical trials have shown that AI-supported screening increases early cancer detection rates without compromising patient safety. Beyond lesion detection, AI enables individualized risk assessment, prediction of interval cancers, and supports the transition toward risk-adapted screening strategies. For Serbia, AI-assisted centralized triage systems could optimize limited radiology resources, improve access to expert interpretation, and strengthen organized screening programmes. Successful implementation, however, requires local validation studies, interoperable digital infrastructure, appropriate regulatory frameworks, and continuous professional education. Conclusion: Artificial intelligence has the potential to substantially improve breast cancer screening by increasing diagnostic accuracy, reducing radiologists' workload, and supporting personalized screening approaches. Strategic integration aligned with European quality standards and national healthcare priorities could promote earlier diagnosis and reduced breast cancer mortality in Serbia.
Ancy Edwin, Kristina Stamenković, V. Mijatović-Jovanović et al.· Hospital Pharmacology· 0 citations
Abstract Artificial intelligence (AI) has emerged as a clinically significant and rapidly evolving technology in breast imaging, with applications spanning cancer detection, risk prediction, workflow optimization, and supplemental imaging modalities. The evidence base has matured rapidly, transitioning from retrospective accuracy studies to prospective randomized controlled trials. This narrative review synthesizes the most recent evidence (2023–2026) on AI applications in breast cancer screening and imaging, including landmark trial results, systematic reviews, and society recommendations. Current data demonstrate that AI-supported mammography screening can increase cancer detection rates by 10 to 29% and reduce interval cancer rates by up to 12% while reducing radiologist workload by 31 to 64% in double-reading screening settings, without increasing false-positive rates. However, the evidence base remains predominantly derived from nondiverse, high-income country populations, and long-term outcome data including breast cancer mortality are not yet available. Challenges related to generalizability, algorithmic bias, overdiagnosis, and regulatory frameworks remain. As the field moves toward clinical implementation, rigorous post-market surveillance, diverse dataset validation, and cost-effectiveness analyses will be essential.
Amrita Kumar, Gerald Lip· Indian Journal of Radiology...· 0 citations
The World Health Organization (WHO) reports 2.3 million new cases and 670,000 fatalities annually, and breast cancer remains one of the most prevalent and life-threatening diseases on a global scale. These poor outcomes highlight the critical need for new diagnostic techniques that can identify breast cancer in its early stages, allowing for early detection and increasing survival rates. Traditional screening methods, such as mammography, are frequently restricted by their high false-positive rates, despite the fact that early and precise diagnosis is essential for enhancing survival rates. To enhance diagnostic accuracy, we recommend a Modified Logistic Regression (MLR) model that employs different activation functions (sigmoid and modified
$$\tanh$$
). In addition, we employed feature selection techniques to reduce redundancy and eradicate highly correlated features in the Wisconsin Diagnostic Breast Cancer (WDBC) dataset and Wisconsin breast cancer (WBC) dataset. The proposed MLR model was compared using standard machine learning algorithms, including Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), K-Nearest Neighbors (KNN), XGBoost, LightGBM, CatBoost and these models are integrated into a soft voting classifier (SVC) technique to enhance predictive accuracy. All machine learning models in this study were implemented using the scikit-learn library, except for the modified logistic regression model, which was manually coded without using the LogisticRegression class from sklearn. Each model’s performance is individually evaluated and then compared against MLR. The proposed MLR attains an accuracy that is competitive with the strongest baselines (98.25% on WDBC and 97.81% on WBC); a paired McNemar’s test confirms that this accuracy is statistically on par with standard logistic regression rather than significantly superior. The distinctive benefits of MLR lie elsewhere: it produces better-calibrated probability estimates (consistently lower Brier scores than logistic regression and far below the no-skill prevalence baseline), a lower false-negative rate, and more compact feature attributions with consistent ranking, within a flexible architecture that admits any differentiable activation. We further show analytically that the modified
$$\tanh$$
used by MLR is equivalent to a logistic sigmoid evaluated at twice the logit,
$$\tfrac{1+\tanh (z)}{2}=\sigma (2z)$$
, so the training objective remains convex in the effective logits and the learnable bias weight introduces no spurious minima. These results position MLR as a reliable, well-calibrated instrument for supporting early breast cancer diagnosis.
A. Sudhagar, S. Athithan, V. Sivakumar et al.· Scientific Reports· 0 citations
Breast cancer continues to be a significant health challenge for women worldwide, and early detection is essential for improving survival outcomes. Over the years, Computer-Aided Diagnosis (CAD) systems have supported radiologists by highlighting suspicious regions in mammograms, ultrasound images, and magnetic resonance imaging (MRI). However, traditional CAD systems evaluate image features based on predefined rules and are associated with high false-positive rates, limiting their ability to adapt to new or complex cases. This study aimed to examine the role of deep learning (DL) in the development of computer-aided diagnosis (CAD) systems and to evaluate the diagnostic accuracy and clinical integration of AI-based approaches in breast cancer detection. A systematic review was conducted by analyzing 13 studies published between 2000 and 2024. Relevant studies were identified through a comprehensive search of the PubMed, IEEE Xplore, and Scopus databases. The included studies were evaluated according to their methodology, deep learning architectures including Convolutional Neural Networks (CNNs), ResNet, EfficientNet, and Vision Transformers and their reported diagnostic performance in terms of accuracy, sensitivity, specificity, and false-negative reduction. The findings were synthesized narratively to compare the advantages and limitations of AI-based diagnostic systems with traditional CAD systems. The findings demonstrate that DL-based models outperform traditional CAD systems. These models contribute to earlier diagnosis by detecting subtle lesions, microcalcifications, and asymmetrical structures that may be overlooked by the human eye. However, several challenges remain, including the availability of large, high-quality datasets, the interpretability of AI decisions, ethical considerations, and integration into routine clinical practice. Deep learning-enhanced CAD systems represent a promising advancement in breast cancer diagnosis by improving diagnostic performance and supporting clinical decision-making. However, their successful implementation depends on high-quality data, transparent and interpretable models, and appropriate ethical and regulatory frameworks to ensure that AI serves as a supportive tool that complements, rather than replaces, radiologists.
Şeyma Betül Samsa, Hina Zahoor· İstanbul Gelişim Üniversites...· 0 citations
: Breast cancer is routinely encountered in screening programs and continues to impose a heavy burden on health systems. Shifting diagnosis toward earlier stages is the primary driver of mortality reduction in screening, yet traditional diagnostic pathways — mammography interpretation, ultrasound/MRI assessment, and biopsy — can be expensive, invasive, and vulnerable to human error. Machine learning (ML) is used to prioritize borderline cases and to assign calibrated malignancy risk for expedited review. Among many algorithms, logistic regression (LR) stands out for its simplicity, computational efficiency, and interpretability, which are critical for clinical acceptance. This review synthesizes recent work on breast tumor classification with an emphasis on LR and compares it with Support Vector Machines (SVM), Decision Trees (DT), XGBoost, and Neural Networks (NN). This paper summarizes common datasets and metrics, highlight reported performance ranges, and discuss practical challenges — including interpretability, privacy, dataset shift, and external validation. This paper concludes with suggestions for future research on explainable AI, multi‑modal feature fusion, and privacy‑preserving learning pipelines that are suitable for real‑world healthcare deployments.
Xusen Zheng· Proceedings of the 3rd Inter...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
