A CT procurement comparison should be built from documented physics measurements, regulatory reporting requirements and accreditation standards rather than marketing specifications. The defensible sequence is to fix the clinical use case first, then compare documented acceptance-test and annual-evaluation parameters across candidate models, and only then weight commercial factors such as service, training and total cost. Federal rules require manufacturers to supply dose and imaging-performance information in the user documentation, while accreditation bodies define the minimum equipment and quality-control expectations a facility must meet [2][1]. A concrete example from a published North Country Healthcare request for proposals shows how one organization translated these obligations into weighted evaluation criteria for three hospital sites [6].

The central difficulty is that comparable scanner data exists in different formats. Console-displayed CTDIvol values are manufacturer-reported and consistent within a given make and model, but cross-vendor comparisons require matching phantom size, tube potential, collimation and other technique factors [4]. A medical physicist can turn those values into measured acceptance-test results, but procurement teams should not assume that a higher slice count or a faster rotation time alone predicts acceptable imaging. The comparison method described below treats source documents as the evidence base and names what remains unknown.

Fix the examination scope before scoring equipment

Accreditation standards make the clinical examination types the starting point. IAC accreditation applies to specific categories such as neurological CT, coronary CTA, body CT and low dose CT lung cancer screening, and the 2026 version states that all CT systems used for diagnostic studies must be FDA compliant for the specific examination or procedure being performed [5]. A buyer comparing scanners for a cardiac program therefore needs different minimum capabilities than a buyer selecting a unit for routine abdomen and pelvis work. Without a fixed examination list, feature comparisons become a scorecard of unrelated specifications.

The equipment question follows the protocol question. The IAC standard requires that instruments be appropriate for the organ systems being imaged and approved for the specific imaging task [1]. Comparing scanners therefore means comparing documented compliance for the planned examinations, not comparing general promotional tiers. A 64-slice system may be enough for routine body imaging at one site but unsuitable for a site required to perform acute stroke brain or coronary CTA at standardized speed. The source documents do not supply model-specific clinical performance claims, so the comparison must rely on the manufacturer's published performance specifications and the physicist's verification of them for the planned protocols [1][3].

Use acceptance testing and annual physics criteria as the comparison spine

The ACR–AAPM technical standard specifies the minimum parameters that make scanners comparable. Acceptance tests must include compliance with local regulatory requirements, compliance with special contractual terms, evaluation of compliance with manufacturer's specifications as relevant, evaluation of room radiation shielding, and the tests performed during the annual performance evaluation [3]. Annual evaluation must include alignment light accuracy, image localization, table travel accuracy, radiation beam width, high-contrast resolution, low-contrast detectability, image uniformity, noise, artifact evaluation, CT number accuracy, acquisition workstation display quality, and radiation output compared to baseline [3].

A procurement team should request these measurements in a common format for each candidate system. The IAC physicist report requirements add detail: radiation dose reports must compare measured dosimetry with a reference standard using the same dose units and indicate pass or fail, while image quality reports must record contrast scale, mean CT number of water and reference materials, CT number linearity, laser light alignment, gantry tilt, slice localization accuracy and table incrementation accuracy [1]. Comparing candidate models on these measures separates documented performance from sales descriptions. Where a vendor cannot supply the underlying test data, that gap is itself a procurement risk.

A concrete comparison worksheet

A suggested workflow starts with five fixed fields: planned examination categories from the IAC accreditation scope, FDA-cleared indications from manufacturer documentation, measured annual-evaluation parameters from the ACR–AAPM list, dose documentation under matched CTDI phantom conditions, and the vendor's completed commercial response form. For a hypothetical three-hospital replacement purchasing one scanner per site, the procurement team would first list the highest-acuity protocol at each site, then request identical acceptance-test reporting for alignment, CT number accuracy, noise, uniformity and dosimetry. Next, the team would request normalized CTDIvol data only for matched phantom and technique combinations. Finally, the team would apply the organization's own weighted commercial criteria to the systems that pass the physics review. This sequence is a suggestion, not a validated testing protocol; the source documents do not establish that it improves procurement outcomes.

The main limitation is that comparing scanners on accreditation criteria does not produce a clinical ranking. IAC standards and ACR–AAPM technical standards set minimum performance and monitoring expectations, and compliance indicates that a system meets professional baseline requirements. They do not certify that one compliant model produces better diagnostic outcomes than another compliant model. Procurement teams should record which candidate features lack direct source support and avoid inventing a winner when the available documentation supports only a pass or fail judgment on physics and compliance measures.

Compare dose information under identical measurement conditions

Federal rules create a standard dose-reporting obligation. Under 21 CFR 1020.33, manufacturers must provide purchasers with CT conditions of operation, dose information obtained using the CT dosimetry phantom, and separate head and body dose information for systems designed to image both applications [2]. The required measurements include CTDI at locations along the axis of rotation and one centimeter interior to the phantom surface [2]. This requirement means procurement teams can demand the same dose documentation from every vendor, provided they ask for it in the same phantom and technique terms.

Cross-vendor dose comparison requires careful matching. A survey of normalized CTDIvol values across four major vendors found strong agreement under similar technique factors, with differences between 0% and 12% for the compared scanners [4]. The authors caution that the CTDIvol values of the desired scanner and the reference scanner must both correspond to the same sized PMMA phantom [4]. Normalized CTDIvol in mGy per 100 mAs is therefore a useful procurement metric only when tube potential, collimation, bowtie filter size and phantom size are held constant. Comparing raw vendor brochures without these controls can produce falsely large or small apparent dose differences.

Separate documentation gaps from comparative weaknesses

A formal procurement structure shows what to score and what to leave unknown. The North Country Healthcare RFP issued March 1, 2026 for three CT scanners assigned weights of 30 percent to technical compliance, 25 percent to total cost of ownership, 20 percent to service and support, 10 percent to implementation timeline, 10 percent to vendor experience, and 5 percent to training [6]. The RFP also required references from at least three hospital clients, itemized pricing, and a minimum five-year service agreement [6]. This structure is a single organization's decision, not a market standard, but it illustrates how physics-based criteria can be embedded in a broader commercial comparison.

The same documentation also reveals what an RFP may not verify. Vendor response forms commonly ask for number of slices, gantry bore size, table weight capacity, rotation time, reconstruction technology, included software and optional module costs [6]. None of those fields by itself proves dose performance or image quality. A comparison should therefore treat the vendor-completed specification form as a screening document and reserve the comparative assessment for physicist-reviewed measurements, dosimetry reports and protocol validation. Claims that cannot be tied to a measurement or manufacturer document should be marked as unverified rather than converted into evaluation points.