Advances in Breast Imaging: Quantitative Assessment, AI Applications, and Imaging Biomarkers

Artificial intelligence in breast cancer imaging: a systematic review of diagnostic performance, predictive modeling, and clinical translation.

Precision Diagnostics in Prostate Cancer: Integrating Biomarkers, Imaging, Genomics, and Artificial Intelligence in Contemporary United States Practice

Prostate cancer is the most commonly diagnosed non-cutaneous malignancy among men in the United States and remains a leading cause of cancer-related mortality. Its marked biological, molecular, and histopathological heterogeneity creates a central diagnostic challenge: identifying clinically significant disease while limiting unnecessary biopsy and overdiagnosis of tumors unlikely to affect survival or quality of life. Although prostate-specific antigen (PSA) remains the foundation of early detection, its limited cancer specificity has driven the development of increasingly risk-adapted diagnostic pathways. Contemporary evaluation integrates clinical risk assessment and PSA-derived measures with selectively used blood- and urine-based biomarkers, multiparametric magnetic resonance imaging (mpMRI), image-guided biopsy, histopathological classification, genomic risk assessment, and molecular imaging. Biomarkers such as the Prostate Health Index, 4Kscore, IsoPSA, MiCheck, SelectMDx, and ExoDx may refine biopsy decisions in appropriately selected patients but should be interpreted according to the clinical setting, decision threshold, and surrounding diagnostic pathway. Prostate MRI and PI-RADS-based assessment have become central to pre-biopsy evaluation, while MRI-targeted biopsy improves detection of Grade Group ≥ 2 disease. Increasing use of the transperineal biopsy route offers comparable cancer detection with a lower infectious risk. Following diagnosis, Grade Group, adverse histological features, clinical risk models, and selected tissue-based genomic classifiers provide complementary prognostic information. PSMA PET/CT has further improved staging of selected patients with higher-risk disease and localization of biochemical recurrence. Precision diagnostics must also account for disease phenotypes that may not be adequately represented by conventional PSA- and imaging-based pathways, including intraductal carcinoma, cribriform architecture, ductal adenocarcinoma, and neuroendocrine prostate cancer. Emerging approaches, including artificial intelligence-assisted MRI interpretation, digital pathology, high-frequency micro-ultrasound, liquid biopsy, alternative molecular radiotracers, and multi-omic integration, show increasing potential but remain at different stages of validation and clinical adoption. This review critically examines contemporary prostate cancer diagnostics within United States clinical practice, distinguishing established guideline-supported approaches from selectively used adjuncts and emerging technologies. Particular emphasis is placed on diagnostic performance in context, clinical utility, external validation, healthcare equity, regulatory considerations, and the need to demonstrate that increasing diagnostic complexity translates into meaningful improvements in patient care.

Artificial Intelligence in Breast Imaging and Screening: Current Evidence, Applications, and Future Directions

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.

A COMPREHENSIVE EVALUATION OF BREAST CANCER DETECTION AND DIAGNOSTIC METHODS: TRADITIONAL AND EMERGING TECHNOLOGIES

ABSTRACT: Background: Breast cancer is the most frequently diagnosed malignancy in women worldwide, with over 2.3 million new cases annually. Despite advances in treatment, early detection remains the primary determinant of survival. Objective: To systematically evaluate the diagnostic performance of Digital Mammography (DM), Digital Breast Tomosynthesis (DBT), Ultrasound, MRI, Contrast-Enhanced Mammography (CEM), and AI-assisted imaging for breast cancer detection in women aged 30–70 years. Methods: A PRISMA-compliant systematic review of peer-reviewed literature (2010–2024) was conducted. Studies sourced from PubMed/MEDLINE, Cochrane Library, Embase, and Scopus. QUADAS-2 was applied for quality assessment. Outcome measures: cancer detection rate (CDR), sensitivity, specificity, stage at diagnosis, and recall rate. Results: Of 142 included studies (>4.2 million examinations), DBT demonstrated superior CDR (4.0–6.5/1000; 95% CI: 3.7–6.9) over DM (3.4–5.0/1000; 95% CI: 3.1–5.4) with lower recall rates. MRI achieved the highest sensitivity (90–99%) but lowest specificity (72–89%). CEM and AI-assisted imaging showed clinically promising performance, though evidence remains emerging. Conclusion: No single modality is universally optimal. A risk-stratified, personalised screening approach is recommended, with DBT as preferred standard for average-risk populations and MRI for high-risk individuals. Future prospective trials should address equity, AI validation, and access.

Deep Learning Methods for Breast Cancer Detection, Classification, and Segmentation Using MRI Scans: A Systematic Review

Breast cancer remains one of the leading causes of cancer-related mortality among women worldwide, underscoring the importance of early and accurate diagnosis to improve patient outcomes. Magnetic resonance imaging (MRI) is a highly sensitive imaging modality for detecting breast malignancies, particularly in patients with dense breast tissue or those at high risk, where conventional imaging techniques may have limited sensitivity. Recent advances in deep learning (DL) have demonstrated considerable potential for improving the automated analysis of breast MRI, including tumour classification, prediction, and segmentation. This systematic review synthesises peer-reviewed studies published between 2014 and 2025 that exclusively applied DL techniques to breast MRI for cancer classification, prediction, or segmentation. The included studies were critically evaluated with respect to model architectures, dataset characteristics, image preprocessing methods, validation strategies, and reported performance metrics. The reviewed literature demonstrates that DL models consistently achieve high diagnostic performance and have the potential to enhance radiological workflows by supporting automated lesion detection and clinical decision-making. However, several challenges continue to limit their translation into routine clinical practice, including limited access to large, diverse, and well-annotated datasets, inadequate external validation, variability in MRI acquisition protocols, and concerns regarding model interpretability and generalisability. Future research should prioritise the development of robust, explainable, and clinically validated DL models trained on multicentre datasets using standardised evaluation frameworks. Addressing these challenges will be essential to improve the reliability, reproducibility, and clinical applicability of AI-assisted breast cancer diagnosis using MRI.

Ultrafast breast MRI to differentiate between molecular subtypes of breast cancer: a systematic review and meta-analysis.

BACKGROUND Ultrafast dynamic contrast-enhanced MRI (UF-MRI) offers high temporal resolution and early kinetic information that may extend beyond lesion discrimination to molecular subtype differentiation in breast cancer. Conventional MRI has been more extensively studied, but whether UF-MRI provides comparable or superior diagnostic accuracy remains unclear. OBJECTIVE To systematically review and synthesize the evidence on the diagnostic accuracy of UF-MRI in differentiating breast cancer molecular subtypes with pooled estimates of sensitivity, specificity, diagnostic odds ratio (DOR), and key kinetic parameters. EVIDENCE ACQUISITION A systematic search of PubMed, Scopus, and Embase through 2025 identified studies reporting UF-MRI performance stratified by subtype (Luminal, HER2-enriched, TNBC) in both screening and diagnostic extent-of-disease (EOD) breast MRI populations. Data extraction included study characteristics, diagnostic outcomes, and kinetic parameters (time to enhancement, TTE, and maximum slope, MaxSlope). Meta-analyses were performed using random-effects models to account for between-study heterogeneity with heterogeneity assessed by I². Meta-regression examined covariates including scanner vendor, sequence type, and reader. EVIDENCE SYNTHESIS Nine studies including 1,371 lesions met inclusion criteria. Luminal cancers showed the highest pooled accuracy (sensitivity 0.90, 95% CI: 0.81-0.95; specificity 0.79, 95% CI: 0.65-0.89; DOR 25.80, 95% CI: 10.15-65.59; area under the curve [AUC] 0.87, 95% CI: 0.81-0.92), followed by HER2-enriched (sensitivity 0.85, 95% CI: 0.68-0.94; specificity 0.67, 95% CI: 0.10-0.97; DOR 15.03, 95% CI: 7.72-29.28; AUC 0.81, 95% CI: 0.69-0.89) and TNBC (sensitivity 0.82, 95% CI: 0.74-0.88; specificity 0.72, 95% CI: 0.69-0.75; DOR 10.09, 95% CI: 6.44-15.82, AUC 0.81, 95% CI: 0.72-0.85). Meta-regression identified scanner vendor and sequence type as contributors to heterogeneity. TTE was shortest for TNBC (7.86 s, 95% CI: 4.85-10.86) compared with HER2-enriched (9.67 s, 95% CI: 7.01-12.33) and Luminal (9.78 s, 95% CI: 8.00-11.56). MaxSlope showed extreme heterogeneity with poor reproducibility, precluding reliable interpretation of pooled estimates. CONCLUSION UF-MRI demonstrates subtype-specific diagnostic performance for differentiating breast cancer molecular subtypes, with the strongest discriminatory accuracy in Luminal tumors and diagnostic accuracy patterns and kinetic parameters aligned with known differences in enhancement behavior across subtypes, high sensitivity but modest specificity in TNBC, and more variable performance in HER2-enriched cancers. Time to enhancement (TTE) appears to be a reliable and interpretable kinetic marker, whereas maximum slope (MaxSlope) shows extreme heterogeneity and limited reproducibility, restricting its clinical utility. CLINICAL IMPACT This meta-analysis supports UF-MRI as a rapid imaging technique with potential utility for molecular subtype assessment in breast cancer. UF-MRI may complement existing breast MRI protocols by providing early kinetic information without substantially increasing scan time. Standardization of protocols is essential to reduce heterogeneity, improve reproducibility, and strengthen its role in precision breast imaging. KEY FINDING UF-MRI demonstrated the highest overall diagnostic accuracy in Luminal tumors, reflected by the highest pooled sensitivity (0.90), specificity (0.79), and diagnostic odds ratio (25.8). Diagnostic performance was intermediate in HER2-enriched tumors (sensitivity 0.85, specificity 0.67, DOR 15.0) and lowest in TNBC (sensitivity 0.82, specificity 0.72, DOR 10.1). Time to enhancement (TTE) consistently reflected subtype-specific kinetic differences, whereas MaxSlope showed high variability and limited reliability. IMPORTANCE UF-MRI offers important subtype-specific diagnostic accuracy without costing additional scan time, making it a valuable addition to the evaluation of breast cancer on MRI.