30-Second Takeaway
- New AASLD 2023 HCC surveillance algorithm substantially boosts sensitivity vs prior rules, at the cost of more recalls.
- Breast MRI and vacuum-assisted biopsy can reduce surgery and underestimation in selected high-risk breast lesions.
- Photon-counting CT advances (ultra–low-dose lung and high-Z contrast coronary CTA) enable major dose and artifact reductions.
- MRI-based nasopharyngeal screening outperforms endoscopy for early NPC in high-risk EBV–positive populations.
- LLM-based report assistants and low-field functional lung MRI show emerging roles in survivorship and workflow support.
Week ending January 24, 2026
Imaging innovations reshaping cancer surveillance, cardiothoracic protocols, and radiology workflows
AASLD 2023 HCC surveillance algorithm markedly increases sensitivity over LI-RADS and AASLD 2018
In 953 high-risk patients under US and AFP surveillance, 5% had hepatocellular carcinoma (HCC). The integrated 2023 AASLD algorithm, combining US visualization score, AFP trend, and lesion growth, achieved 94% sensitivity and 99.6% NPV for HCC detection. This sensitivity exceeded US LI-RADS v2017 (60%) and AASLD 2018 guidance (76%), including for early-stage HCC. Specificity dropped to 84% versus 90% for LI-RADS 2017 and 89% for AASLD 2018, implying more follow-up imaging or workup. Lower AFP (<20 ng/mL) and absence of cirrhosis were linked to false-positive classifications on multivariable analysis.
Contrast-enhanced breast MRI can safely exclude malignancy in selected B3 lesions
This meta-analysis pooled seven studies (493 high-risk B3 lesions; 17% malignancy prevalence) assessing contrast-enhanced breast MRI after image-guided biopsy. Pooled sensitivity was 91.3% and specificity 68.8% for detecting malignancy in high-risk lesions. Only six malignancies were missed by MRI, all small low-grade DCIS, indicating strong negative reassurance in this setting. Modeling suggested that for lesions with pretest malignancy probability ≤13.1%, a negative MRI keeps post-test risk ≤2%. These data support MRI-based surveillance rather than surgery for carefully selected B3 lesions with low to moderate pretest risk.
Ultra–low-dose photon-counting CT quantifies emphysema with 87% less radiation
In 152 emphysema patients, same-day low-dose and ultra–low-dose (ULD) photon-counting CT (PCCT) were compared. The ULD protocol reduced effective dose by 87% (0.20 vs 1.58 mSv) while maintaining similar overall image quality, sharpness, and artifact scores. Image noise was higher on ULD, but visual grading of centrilobular and paraseptal emphysema showed excellent agreement with low-dose scans (κ up to 0.98). Automated low-attenuation volume emphysema quantification showed excellent agreement between dose levels (ICC 0.96–0.98 across lungs and lobes). Visual emphysema grades correlated similarly with quantitative measures for both protocols, supporting ULD PCCT for longitudinal COPD monitoring.
Screening MRI outperforms endoscopy for early nasopharyngeal carcinoma in EBV-positive individuals
In 814 EBV-seropositive adults followed for 5 years, 32 nasopharyngeal carcinomas were detected. MRI identified 30 cancers, whereas endoscopy detected 21, yielding higher overall detection and better early-stage tumor detection rates. MRI sensitivity was 93.8% versus 65.6% for endoscopy, with a lower false-negative rate but lower specificity (84.0% vs 92.8%). Positive predictive value was also lower for MRI (19.6% vs 27.6%), implying more false-positive referrals and biopsies. The authors suggest a potential 5-year rescreening interval for MRI-negative high-risk individuals, pending broader validation and cost-effectiveness data.
References
Numbered in order of appearance. Click any reference to view details.
Additional Reads
Optional additional studies from this edition.