Radiological.ai

Platform & software · PACS software

PACS software: PACS system vendors and cloud PACS systems compared

The short answer

PACS software, short for picture archiving and communication system, is the software that stores, retrieves and displays medical images, and it is what radiologists actually read from. There is no single best PACS, and the clearest evidence for that is that KLAS does not rank the category as one list. In the 2026 Best in KLAS awards announced February 4, 2026, Sectra ranked first for practices reading more than 300,000 studies a year, its thirteenth consecutive win in that band, while Agfa HealthCare ranked first below 300,000 studies and INFINITT PACS won both the ambulatory and community categories. So the first number a buyer needs is their own annual study volume, because it decides which shortlist even applies. Cost splits the same way: federal prime contracts naming "picture archiving and communication" carry a median of $250,438, while contracts naming "PACS software" carry a median of $33,725, because the first is a platform deal and the second is usually a license or module line item.

Every PACS shortlist starts the same way. Somebody searches for the best PACS software, lands on a ranked list of ten vendors, and brings that list to the next meeting. The list is almost always wrong for the practice holding it, and not because the vendors on it are bad.

KLAS, which is the closest thing this market has to an independent scoreboard, does not publish one PACS ranking. It publishes several, split by how many studies a practice reads in a year, and a different vendor takes the top spot in each band. In the awards announced on February 4, 2026, Sectra won above 300,000 studies per year for the thirteenth consecutive time, Agfa HealthCare won below 300,000, and INFINITT PACS won ambulatory and community. Agfa also ranked first for universal viewer and vendor neutral archive, which are related purchases people often fold into the same project.

Read that as a buying instruction rather than a trophy list. A three-site outpatient imaging center reading 90,000 studies a year and a health system reading 600,000 are shopping in different categories, and the vendor that satisfies one is frequently the wrong fit for the other. Volume drives how the product scales, how it is priced, and how painful it is to leave.

The cost question has the same shape. We pulled every federal prime contract award since October 2016 that names PACS in four different ways, and the numbers are not close to each other. Platform procurements run to a median of $250,438 and a 75th percentile of $1,340,540. Line items that literally say "PACS software" run to a median of $33,725. Both are real, they are just answers to different questions. The tables below work through which band you are in, who ranks first there, and what the contract record suggests you should budget.

Last updated August 2026

The Reading Station

Worklist

SERIES 1 · AX
SLICE 24/64
SAMPLE STUDY
NOT FOR DIAGNOSTIC USE
W 80 · L 40
ILLUSTRATIVE SAMPLE

Structured report

Draft

Run the assistant to draft this report for review.

You review & sign

Illustrative sample · not a real patient study, not a diagnosis

Drafted in · you review & sign Worklist re-prioritized

Decision support for qualified clinicians. Radiological.ai does not provide a diagnosis and is not a substitute for professional judgment.

Run the assistant

Flag · prioritize · draft · you review and sign

X-RAY CT MRI BUILT WITH RADIOLOGISTS

Decision support not a diagnosis

You review & sign

PACS is one half of the operational stack and the other half is the schedule, which is covered in radiology information system software, including how RIS and PACS divide the work. If your question is narrower, namely how an AI assistant actually reaches the reading station and returns anything useful to it, that is radiology AI PACS integration over DICOM and HL7. Groups reading across multiple sites usually hit the multi-site problem first, which teleradiology software works through.

For the money side, radiology software pricing breaks federal contract data down by software category, and the contract terms worth negotiating covers renewal uplift and the fully loaded first-year total, which matters more on a platform deal than the license price does. Once the images are handled, the report is the next bottleneck: see the radiology reporting software comparison and the radiology AI alternatives hub for named vendors.

Side by side

Which PACS band you are actually buying in

KLAS segments PACS by annual study volume and a different vendor ranks first in each band. Find your row count first, then build the shortlist. A ranked list that ignores volume is answering someone else's question.

What you are comparing Large practice, above 300,000 studies a year Small practice, at or below 300,000 studies Ambulatory and community imaging Teleradiology and multi-site groups
Who ranked first in the 2026 Best in KLAS awards Sectra, its thirteenth consecutive year in this band Agfa HealthCare INFINITT PACS, which took both the ambulatory and the community category No dedicated KLAS category; these groups usually buy from the band matching their total read volume
What you are actually buying An enterprise imaging platform, normally with a vendor neutral archive and a universal viewer attached A PACS that has to run without a dedicated imaging IT team behind it A system tuned for outpatient throughput and referrer access rather than inpatient breadth Image exchange and reading access across sites you do not control, which is a distribution problem before it is a storage one
Typical deployment On premises or hybrid is still common at this scale, with cloud migration usually staged over years Cloud or vendor hosted, because there is rarely a data center to put it in Cloud first in most new purchases Cloud effectively by necessity, since the readers are not on your network
Integrations you will be maintaining Modalities, RIS, EHR, VNA, reporting platform, plus every downstream viewer Modalities, the RIS or the practice management system, and the reporting platform Modalities, scheduling, referrer portal and reporting Every sending facility separately, which is where the real cost sits
What the federal contract record shows Awards naming "picture archiving and communication" total $68,836,793 across 48 prime awards, median $250,438, 75th percentile $1,340,540 Awards naming "PACS software" total $35,367,110 across 40 awards, median $33,725, which reads like licenses and modules rather than platforms Awards naming "radiology PACS" total $18,602,086 across 72 awards, median $8,573, consistent with support and add-on line items Only 3 prime awards name "cloud PACS" at all, totaling $6,577,444, so federal buying is not a useful guide to the commercial cloud market here
Who actually holds the contract Frequently an integrator: Four Points Technology holds $16,122,656 of the platform awards, ahead of Philips Healthcare Informatics at $13,050,750 Peraton alone holds $24,628,496 of the "PACS software" awards, with Four Points second at $7,898,208 Philips Healthcare Informatics leads at $7,810,350, with Agfa HealthCare at $2,830,106 Resellers again; the named PACS vendor is often not the counterparty on the paper
The exit question to ask before signing What migration of the full archive costs, and who owns the study data in a readable form Whether you can export DICOM in bulk without a professional services engagement How referrer access survives a vendor change Whether each sending facility has to be re-onboarded if you switch
Where an AI assistant attaches Reads the study from PACS and returns flags to it; the report is drafted in the reporting platform Same attachment, though a smaller site usually wants one vendor rather than three Same attachment, with turnaround for referrers as the usual driver Attaches once centrally rather than per site, which is the main reason multi-site groups look at it
Regulatory status Display and storage software; some diagnostic viewers carry their own FDA clearances Same, and worth confirming the viewer clearance if reads are diagnostic Same Decision support in our case; we make no regulatory-status claims on this site

Compiled August 2026. Best in KLAS placements are from the 2026 awards announced February 4, 2026, as reported by Radiology Business and by Sectra's own release; category names are KLAS category names. Contract figures are prime awards from the USAspending.gov API, award type codes A, B, C and D, award start dates from October 1 2016 through August 23 2026, searched on the exact keyword shown. Keyword matching on USAspending is literal, so these figures are a floor rather than a full market picture, and per-award statistics drift slightly between pulls while totals stay stable. Vendor names are examples of who sells into each band and are not an endorsement or a shortlist. Confirm current products, ownership and pricing with any vendor directly.

Why it works

What your group gets with PACS software

Start with your study volume, not with a vendor list

The single most useful thing in the public record on PACS is that KLAS refuses to rank it as one category. Above and below roughly 300,000 studies a year are different markets with different winners. Count your annual studies before the first demo, because that number, not the feature grid, decides which three or four vendors are worth a conversation.

The contract is usually signed with an integrator

Across four different ways of naming PACS in the federal award data, the largest single recipients are repeatedly resellers and integrators rather than the PACS vendors themselves. Four Points Technology appears near the top of three of the four slices and Peraton holds $24,628,496 on its own. That changes who holds discount authority, who owns the renewal uplift clause and who you escalate to when an interface breaks.

AI does not require replacing PACS

Radiological.ai works alongside whatever PACS you land on. It reads the study, flags suspected findings for a second look, pushes urgent studies up the worklist and drafts the structured report into your template. It is decision support, not a viewer and not an archive, and the radiologist reviews, edits and signs every study.

What it handles

Flagged, prioritized and drafted for your review

The assistant pre-reads each study, surfaces a region of interest for review, re-prioritizes the worklist, and drafts the structured report in your template. You confirm, edit and sign.

  • Works with your existing PACS rather than replacing it
  • Connects over standard DICOM and HL7 interfaces
  • Flags suspected findings for a second look
  • Reorders the worklist so urgent studies surface first
  • Drafts the structured report into your template
  • Attaches once for multi-site groups instead of per facility
  • Radiologist reviews, edits and signs every study
PACS SOFTWARE STAT

Region of interest flagged for review

A focal region is surfaced on the sample study for the radiologist to review. The assistant does not characterize it as a diagnosis.

Draft impression

Suspected finding flagged for radiologist review. Correlate clinically and confirm. Draft for review and sign-off.

Illustrative sample · not for diagnostic use You review & sign

Why Radiological.ai

One assistant across the whole read

Not three vendors stitched together. Flag, prioritize and draft in one calm pane, on X-ray, CT and MRI, with the radiologist signing every study.

Flags suspected findings

A second set of eyes surfaces regions of interest for review on every study, so a suspected finding is less likely to slip past late in a shift.

Prioritizes the worklist

Suspected-critical studies move to the top, so urgent reads surface ahead of routine follow-ups across your sites and shifts.

Drafts the report

A structured draft arrives in your template, ready to edit and sign. The draft saves the typing and the measuring, never the judgment.

Good questions

Questions about PACS software

There is no single best PACS, and the 2026 Best in KLAS results show why. KLAS ranks the category in bands by annual study volume. Sectra ranked first above 300,000 studies a year, its thirteenth consecutive win in that band, Agfa HealthCare ranked first at or below 300,000, and INFINITT PACS won the ambulatory and community categories. Your study volume decides which of those results applies to you.
Nobody publishes list pricing, but federal contract data gives a usable range. Prime awards naming "picture archiving and communication" total $68,836,793 across 48 awards, with a median of $250,438 and a 75th percentile of $1,340,540. Awards that say "PACS software" instead have a median of $33,725, because those are typically license or module line items rather than full platform procurements.
The names that recur across KLAS categories and the federal contract record include Sectra, Agfa HealthCare, INFINITT, Philips, Fujifilm, GE HealthCare, Intelerad, Carestream, Merative and Visage. Which of them belongs on your list depends on your study volume and setting, since the large practice, small practice and ambulatory categories are ranked separately and rarely share a winner.
In practice there are four buying shapes: an enterprise imaging platform bundled with a vendor neutral archive and universal viewer, a standalone departmental PACS, an integrated RIS/PACS sold as one product by a single vendor, and a cloud or vendor hosted PACS. The choice is usually driven by whether you have imaging IT staff and a data center, not by clinical features.
Neither is better in general; they fail differently. Cloud removes the data center, the storage refresh cycle and most of the upgrade work, which is why nearly every new small and ambulatory purchase is cloud. On premises keeps image retrieval independent of your internet connection and keeps the archive physically in your control, which still matters at high volume. The deciding factors are usually bandwidth, in-house IT staffing and how your archive migration would work.
Cloud PACS is normally priced per study or per user per month rather than as a capital purchase, which shifts the cost from an upfront project to an operating line that grows with volume. Federal contract data is not a useful guide here: only 3 prime awards since 2016 name "cloud PACS" at all, totaling $6,577,444. Ask any cloud vendor for the fully loaded three-year total including storage growth and egress on exit.
It is a single product from one vendor covering both the radiology information system, which handles scheduling, worklists and billing, and the PACS, which handles the images. The appeal is that the interface between them disappears and there is one support number. The tradeoff is concentration risk, since a future change to either half means changing both. Sectra and Intelerad are among the vendors that sell it this way.
For a center below 300,000 studies a year, the 2026 Best in KLAS result in that band went to Agfa HealthCare, and INFINITT PACS won the ambulatory and community categories that most outpatient centers actually sit in. Beyond the ranking, weight cloud or vendor hosted delivery, referrer access and bulk DICOM export heavily, because a small center rarely has the IT staff to compensate for a product that assumes one.
A vendor neutral archive stores studies in a standard format independent of the PACS that created them, so you can change PACS without migrating the archive. It matters most where you have multiple PACS instances or expect to switch vendors. Agfa HealthCare ranked first in both the universal viewer and vendor neutral archive categories in the 2026 Best in KLAS awards. Below a few hundred thousand studies a year, a separate VNA is often more architecture than the problem needs.
No. Radiological.ai is decision support that works alongside the PACS and RIS you already run. It reads the study, flags suspected findings for a second look, prioritizes the worklist and drafts the structured report for review. It does not store your archive, it is not a diagnostic viewer, and it does not provide a diagnosis. The responsible radiologist reviews, edits and signs every study.

Explore more

More ways teams read with Radiological.ai

From the blog: best AI radiology software in 2026, how much radiology AI costs, how to implement AI in a radiology workflow, and AI radiology companies.

Read more studies, with the assistant alongside you

Flag suspected findings, prioritize the worklist, and draft the structured report. You review and sign every study.

See pricing

Radiological.ai is a workflow and decision-support tool for qualified clinicians. It does not provide a diagnosis and is not a substitute for professional medical judgment.