What’s new at RSNA: risk from the image itself

RSNA’s breast-imaging preview highlights sessions on image-only, 5-year breast cancer risk models, external validation work, and how MRI adds value in multi-modal AI. It also calls out global screening updates and a deeper look at background parenchymal enhancement (BPE) on MRI. RSNA

In parallel, the FDA granted De Novo authorization to the first image-only AI risk platform that predicts 5-year risk directly from a screening mammogram—an inflection point that makes risk-adjusted pathways far more scalable. Coverage from Radiology Business and BCRF explains the authorization and clinical intent. Radiology Business

Why it matters: average-risk guidance in the U.S. now begins screening at age 40 (USPSTF, 2024). Programs can layer image-based risk on top of that baseline to triage who needs annual vs. short-interval follow-up and who merits supplemental imaging. USPSTF

CEM is earning a seat next to MRI

Expect exhibits and sessions positioning CEM as a cost-effective, accessible adjunct—particularly for dense-breast populations and diagnostic workups. RSNA News recently framed CEM as a practical alternative to MRI in some screening/diagnostic scenarios, and new peer-review literature is refining technique (e.g., lower volume/higher-iodine contrast while preserving diagnostic performance). RSNA

On outcomes, the RACER trial in The Lancet Regional Health – Europe reported that using CEM as primary imaging for recalled women improved the accuracy and efficiency of the work-up compared with conventional imaging—evidence that will influence protocols beyond the show floor. The Lancet

MRI still leads for sensitivity—BPE is your underused signal

Breast MRI remains the sensitivity champion for high-risk patients and for problem solving. This year’s RSNA content spotlights BPE—how the level of background enhancement relates to tumor biology and outcomes. Recent reviews (2024–2025) synthesize BPE’s predictive/prognostic value, including associations with pathologic complete response after neoadjuvant therapy and survival in certain subtypes. SpringerLink

Practical move: standardize how you document BPE and incorporate it into structured reports and risk conferences; it’s becoming more than a descriptive footnote.

What to ask vendors at RSNA

1. Risk engine proof: “Show external validation and calibration plots by density and race; how does your image-only model integrate into our mammography worklist and letters?”
2. CEM logistics: “Demonstrate CEM acquisition workflows, contrast protocols, and how your viewer handles subtraction/kinetics alongside priors.”
3. MRI + BPE analytics: “Can we standardize BPE capture in structured reports and trend it across treatment?”

As risk-first screening, CEM, and MRI gain real traction, the winners will be the programs that operationalize them quickly and consistently. If you’re planning your 2026 breast-imaging playbook, stop by Vesta at RSNA to see how our subspecialists, standardized templates, and overflow routing make risk-adjusted pathways usable on day one.

The post Breast Imaging 2025–26: Risk Models, CEM/MRI Momentum — RSNA Preview first appeared on Vesta Teleradiology.

1) AI that graduates from pilot to practice

This year’s agenda emphasizes real outcomes over proofs of concept: reader-in-the-loop tools, bias monitoring, and governance. In breast imaging alone, RSNA previews spotlight external validation for image-only risk models and integration of MRI signals into multimodal AI—clear signals that “personalization” is landing in routine workflows. Bring vendor questions that force specifics: external validation cohorts, drift detection, and how metrics (TAT, recalls, rework) appear in your dashboard. RSNA

What to set up before RSNA: define 3–5 outcome metrics and insist every demo shows pre/post performance tied to those measures. Use QIBA concepts to push for standardized inputs/outputs so results are reproducible across scanners and sites. QIBA Wiki

2) Personalization that reaches the reading room

Personalization isn’t only radiogenomics. RSNA’s preview points to risk-stratified pathways you can actually run: e.g., image-only 5-year breast cancer risk at the point of screening to route patients into annual vs. short-interval follow-up or supplemental imaging (CEM/MRI). That pairs well with updated U.S. recommendations: screening beginning at age 40 for average-risk women, then adjusting based on risk and local policy. Build routing rules, templates, and letters now, so RSNA demos can plug into your plan.

Operational checklist:

• Map risk thresholds → next steps (annual vs. short-interval, CEM/MRI).
• Standardize templates so risk outputs appear consistently in reports and patient letters.
• Decide who reviews outlier risk flags and how quickly (SLA).

3) Equity you can instrument—not just endorse

RSNA is foregrounding health equity, with sessions on encoding equity in AI and addressing access gaps for underserved communities. Equity becomes real when you can see it in your data: turnaround times by language, missed-appointment patterns by zip code, recall rates by screening site, and AI performance by subgroup. Build those slices into your analytics now; then ask vendors to show subgroup performance in their dashboards.

Practical moves:

• Add demographic and language filters to your TAT and recall reports.
• Require AI vendors to show calibration and error analysis by subgroup.
• Stand up multilingual patient letter templates to support new screening starts at 40. USPSTF

4) CEM/MRI momentum: choose the lever that fits your service line

RSNA coverage calls out CEM as an increasingly practical adjunct—especially useful for dense-breast populations and diagnostic workups where capacity or cost limits MRI. The RACER trial reported higher accuracy and efficiency for CEM as the primary exam for recalled women vs. conventional imaging—evidence that can justify protocol changes and equipment planning. Meanwhile, MRI retains the sensitivity crown, with renewed attention on background parenchymal enhancement (BPE) as a signal worth documenting consistently.

Action items:

• Decide where CEM fits: diagnostic recall pathway, dense-breast supplemental strategy, or both.
• Add BPE level to structured MRI reports and trend it during therapy response clinics.

5) Governance, not guesswork

If personalization is the “what,” governance is the “how.” Use QIBA ideas—claim definitions, acquisition standards, and profile adherence—to control variability across devices and shifts. Tie RSNA learnings to a written governance plan with three parts: 1) protocol book (who owns it, update cadence), 2) quality book (metrics, subgroup views), and 3) AI book (approval process, monitoring, rollback).

6) Where teleradiology extends your capacity

Personalization increases complexity at peaks (recalls, dense-breast seasons, MR backlogs). A teleradiology partner helps you keep individualized pathways moving: standardized templates, subspecialty over-reads, and after-hours coverage that adheres to your risk rules and equity metrics—so “Imaging the Individual” doesn’t stop at 5 p.m.

Headed to RSNA?

Visit Vesta at Booth 1346 (South Hall) to see how we make “Imaging the Individual” work in real clinics—then enter to win a 1-year Medality CME subscription. Don’t wait: email “RSNA CME Entry” to info@vestarad.com now for a reserved entry, and show your confirmation at the booth for a bonus entry.

The post Imaging the Individual — In the Trenches: AI, Personalization & Equity at RSNA 2025 first appeared on Vesta Teleradiology.

What changed—and why it matters

The FDA has expanded the label for flutemetamol F 18 (Vizamyl), enabling quantification of amyloid plaque burden and long-term therapy monitoring in Alzheimer’s disease. This shift moves amyloid PET beyond a qualitative “positive/negative” decision toward objective, longitudinal assessment that can inform treatment choice, dose intervals, and discontinuation decisions. Business Wire

Professional groups report the update aligns amyloid PET with the clinical era of disease-modifying anti-amyloid therapies (e.g., lecanemab, donanemab), clarifying roles for baseline confirmation, on-treatment monitoring, and response tracking in routine care. Notably, SNMMI stated the FDA granted supplemental indications—including quantitative measurement and use for therapy monitoring—to three amyloid PET agents (flutemetamol F-18/Vizamyl, florbetapir F-18, and florbetaben F-18). SNMMI

Operational updates for radiology leaders

• Protocols & quant pipelines: Build or validate a quant workflow (SUVr or comparable metrics) that’s scanner-calibrated and reproducible across sites. If you operate multi-vendor fleets, document harmonization steps in your SOPs.
• Structured reports: Add fields for quantified burden at baseline, change from baseline, and interpretive guidance tied to therapeutic decisions (initiation, continuation, or discontinuation).
• Scheduling & throughput: Expect rising referral volume from neurology and geriatrics as therapy monitoring enters routine practice; protect access with extended hours or overflow capacity.
• Quality & governance: Define thresholds for biologically meaningful change, reader training for quant review, and reconciliation rules when quant and visual impressions diverge.

For additional context, trade coverage underscores that the updated label formally removes previous limitations around therapy monitoring and permits quant analysis in routine reporting. Empr

How Vesta Teleradiology helps

Vesta’s subspecialty neuro and nuclear medicine radiologists provide:

• Amyloid PET expertise: Visual+quant reads with structured templates aligned to your therapy pathway.
• Coverage when you need it: After-hours, weekends, or daytime overflow—without sacrificing turnaround time.
• Interoperability: Seamless delivery to your PACS/RIS and EMR; clear flags for therapy decisions and recall intervals.
• QA you can see: Peer review, consistency checks across readers, and optional double-reads during program ramp-up.

If you’re standing up or scaling amyloid PET services, we can supply immediate subspecialty coverage and templates tuned to your neurologists’ needs.

The post Vizamyl’s New PET Label: Quantify & Monitor Amyloid—What It Means for Imaging Teams first appeared on Vesta Teleradiology.

As teams lock in Q4 budgets and head into RSNA season, the FDA’s AI lifecycle draft (Jan 2025) and the now-final PCCP (Dec 2024) have reset what buyers should expect from AI in imaging—devices, software, and workflows. Vendors are updating claims and governance; this issue distills a practical buyer’s checklist—multisite validation with subgroup results, drift monitoring and version control, clear in-viewer transparency—and how pairing those tools with Vesta’s subspecialty coverage and QA turns promise into measurable gains across CT/MRI/US/mammography.

A practical buyer’s checklist

Use this when evaluating AI for your service lines:

1. Intended use fit: Verify indications, inputs/outputs, and claims match your pathway and patient mix.
2. Validation depth: Prefer multisite, diverse datasets; stratified results; pre-specified endpoints; documented data lineage and splits.
3. Bias mitigation: Demand subgroup performance (sex, age, race/ethnicity when available), scanner/vendor variability analyses, and site-transfer testing.
4. TPLC plan: Require drift monitoring, retraining triggers, versioning, and how updates are communicated.
5. Human factors & transparency: Ensure limitations, failure modes, and interpretable outputs are presented in-viewer without slowing reads.
6. Security & support: Patch cadence, vulnerability disclosure, SOC2/ISO posture, uptime SLAs, and rollback paths for version issues.
7. Governance: Define metrics owners, review cadence, and thresholds to pause or roll back a model.

Implementation playbook: pilot → scale without disruption

Start with a 60–90 day pilot in one high-impact line (e.g., ED stroke CT or mammography triage) and lock in baselines: median TAT, positive/negative agreement, recall rate, PPV/NPV, and discrepancy rate. Set guardrails—when to auto-triage vs. force human review—and document escalation paths for model failures. Require case-level confidence and structured outputs your radiologists can verify quickly. Stand up a model governance huddle (modality lead, QA, IT security, and your teleradiology partner) that meets biweekly to review drift signals, subgroup performance, and near-misses. Bake in a rollback plan (version pinning) and a quiet-hours change window so updates don’t collide with peak volumes. As results stabilize, scale by cohort (e.g., expand to non-contrast head CT, then CTA) and keep training “micro-bursts” for techs/readers—short videos or checklists in-workflow. Tie vendor SLAs to uptime, support response, and clinical KPIs so the AI program stays accountable to operational value.

Where teleradiology fits

AI only delivers when it’s welded to coverage, quality, and speed. A teleradiology partner should provide:

• 24/7 subspecialty + surge capacity: Vesta absorbs volume peaks so AI never becomes a bottleneck.
• QA you can see: We benchmark pre/post-AI performance, add targeted second looks for edge cases, and feed variance data back to your team.
• Standardized outputs: Structured reports that integrate model outputs with radiologist findings—no black-box surprises.
• Smooth rollout: Pilot by service line (stroke CT, mammo triage, PE workups), then scale with tracked KPIs (TAT, PPV, recalls).
• Interoperability & security: Seamless PACS/RIS/EMR integration with strict access controls, audit trails, and support for change-controlled updates.

Bottom line: Pairing AI with Vesta Teleradiology gives you round-the-clock subspecialty reads, measurable QA, and operational breathing room while you pilot and scale responsibly. If you’re mapping your AI roadmap under the FDA’s 2025 draft guidance, we’ll be your coverage and quality backbone—so your clinicians see faster answers and your patients see safer care. Visit vestarad.com to get started.

The post FDA’s 2025 AI Draft Guidance: A Buyer’s Checklist for Imaging Leaders first appeared on Vesta Teleradiology.