How the O-RADS Calculator Works: Risk Stratification for Ovarian-Adnexal Lesions

The Problem with Adnexal Masses

Adnexal masses are found constantly. A pelvic ultrasound ordered for pelvic pain, an incidental finding on a CT done for something else, a cyst noted during a routine gynecologic visit — adnexal lesions turn up across nearly every imaging pathway, and most of them are benign. Functional cysts, hemorrhagic cysts, endometriomas, and dermoids account for the overwhelming majority of what radiologists see.

The clinical challenge is the same one that drives every reporting system: separating the small number of malignant or borderline lesions from the large number of benign ones, without over-referring patients to surgery and without under-triaging the cancers that matter. The stakes are high in both directions. Operating on a benign cyst exposes a patient to surgical risk and can compromise fertility. Missing an ovarian cancer, or sending it to a general gynecologist rather than a gynecologic oncologist, measurably worsens outcomes — staging and cytoreduction performed by a subspecialist are associated with better survival.

For decades, radiologists described adnexal masses in free-form prose: "a complex cystic lesion with some solid-appearing areas." That language left the referring clinician guessing about how worried to be and where to send the patient. The Ovarian-Adnexal Reporting and Data System, O-RADS, was developed by the American College of Radiology to replace that ambiguity with a defined lexicon, an ordered set of risk categories, and a management pathway attached to each one. O-RADS comes in two systems that share a 0-to-5 scale but rest on different evidence: O-RADS US v2022 for ultrasound and O-RADS MRI 2022 for magnetic resonance imaging.

O-RADS US v2022: Scoring from the Sonographic Lexicon

O-RADS US v2022, updated by Strachowski and colleagues on behalf of the ACR O-RADS US Committee and published in Radiology in 2023, scores a lesion from its ultrasound appearance. The system first classifies the lesion by type — physiologic, unilocular cyst, bilocular cyst, multilocular cyst, solid lesion, or one of the classic benign entities — and then refines the category using inner wall morphology, the presence of any solid component, the number of papillary projections, and a color score that grades vascularity from 1 (no flow) to 4 (very strong flow).

The categories run from 0 to 5, each with a malignancy risk band drawn from the IOTA-derived data underlying the system. O-RADS US 0 means the evaluation is incomplete. O-RADS US 1 is a normal ovary or a physiologic finding, carrying essentially 0% risk. O-RADS US 2 is almost certainly benign, with a malignancy risk under 1%. O-RADS US 3 is low risk, 1% to under 10%. O-RADS US 4 is intermediate risk, 10% to under 50%. O-RADS US 5 is high risk, 50% or higher.

A few decision rules carry most of the weight. The lesion type and the presence of a solid component protruding 3 mm or more off a wall or septum are the primary branch points. A unilocular cyst with a smooth inner wall is O-RADS 2 below 10 cm and O-RADS 3 at 10 cm or larger; an irregular inner wall (a projection under 3 mm) moves it to O-RADS 3. Once a true solid component appears, the lesion escalates: a unilocular cyst with a solid component and four or more papillary projections is O-RADS 5, while fewer projections or a non-papillary solid component is O-RADS 4. For bilocular and multilocular cysts with solid tissue, the color score decides — color score 1 to 2 is O-RADS 4, while color score 3 to 4 is O-RADS 5. A solid lesion with a smooth contour and no flow (color score 1) is O-RADS 3, while an irregular outer contour or color score 4 is O-RADS 5. Finally, ascites or peritoneal nodules not otherwise explained are themselves an O-RADS 5 descriptor, independent of the lesion classification.

The strength of the ultrasound system is that it formalizes a vocabulary radiologists already use, then anchors each combination of descriptors to a numeric category with a defined risk band. That reproducibility is what lets an O-RADS 2 read the same way in one institution as in another, and it is what makes the difference between a lesion that returns to routine care and one that is sent for further workup.

O-RADS MRI 2022: Scoring from Enhancing Solid Tissue

O-RADS MRI 2022, developed by Sadowski, Thomassin-Naggara, and colleagues and published in Radiology in 2022, takes a different path to the same 0-to-5 scale. Where ultrasound reads morphology, MRI reads tissue. The central question is whether the lesion contains enhancing solid tissue, and if so, how that tissue behaves on dynamic contrast-enhanced imaging.

The scores map to approximate positive predictive values for malignancy rather than the risk bands used on ultrasound. O-RADS MRI 1 is a normal ovary. O-RADS MRI 2 is almost certainly benign, with a PPV of approximately 0.5%. O-RADS MRI 3 is low risk, with a PPV of about 5%. O-RADS MRI 4 is intermediate risk, with a PPV of about 50%. O-RADS MRI 5 is high risk, with a PPV of about 90%.

The decision rules follow the tissue. Lesions without enhancing solid tissue — a unilocular cyst with simple or endometriotic fluid, a fat-containing dermoid with no enhancing component, a multilocular cyst with no solid tissue — fall into O-RADS MRI 1 to 3. Simple and endometriotic fluid are O-RADS MRI 2 regardless of wall enhancement; hemorrhagic or proteinaceous fluid becomes O-RADS MRI 3 once the wall enhances. A multilocular cyst with no enhancing solid tissue is O-RADS MRI 3. There is a dedicated benign exit: solid tissue that is homogeneously dark on T2-weighted imaging and on high-b-value diffusion-weighted imaging is fibrous, and is O-RADS MRI 2 even though it is solid. Once enhancing, non-fibrous solid tissue is present, the dynamic contrast-enhanced time-intensity curve sets the score — a low-risk curve is O-RADS MRI 3, an intermediate-risk curve is O-RADS MRI 4, and a high-risk curve is O-RADS MRI 5. When dynamic imaging is not available, the score falls back on the enhancement of the solid tissue relative to the myometrium at 30 to 40 seconds: enhancement at or below the myometrium is O-RADS MRI 4, and enhancement greater than the myometrium is O-RADS MRI 5. As on ultrasound, peritoneal, mesenteric, or omental nodularity or irregular thickening is an O-RADS MRI 5 descriptor in its own right.

The time-intensity curve is what gives MRI its discriminating power. A benign fibrous or stromal tumor and an early ovarian cancer can look similar on a single static image, but they enhance differently over time, and the dynamic curve captures that difference in a way morphology alone cannot.

How the Two Systems Work Together

The two O-RADS systems are not competitors; they answer different questions in sequence. Ultrasound is the first-line modality for characterizing an adnexal mass — it is widely available, requires no contrast, and resolves the majority of lesions definitively. A classic hemorrhagic cyst, a typical dermoid, or a simple cyst rarely needs anything more than the ultrasound report.

MRI enters when ultrasound leaves a lesion indeterminate. An O-RADS US 3 or 4 lesion — one with enough complexity to raise concern but not enough to be definitively malignant — is exactly the kind of problem MRI is built to solve. The superior soft-tissue contrast and the dynamic time-intensity curve let MRI reclassify many sonographically indeterminate lesions into a confidently benign category, sparing the patient surgery, or into a high-risk category that justifies immediate gynecologic oncology referral. This is why the ACR algorithm explicitly routes O-RADS US 4 lesions toward MRI characterization with an O-RADS MRI score.

Read this way, the shared 0-to-5 scale is a convenience for communication, not a claim that an O-RADS US 4 and an O-RADS MRI 4 mean exactly the same thing. They reflect different evidence and different precision — the MRI score, applied to the lesions that reach it, generally carries a sharper malignancy estimate.

Using the Aperivue O-RADS Calculator

To make both systems easier to apply at the workstation, we built a free online calculator at aperivue.com/rads/orads that supports O-RADS US v2022 and O-RADS MRI 2022.

The tool walks through the lexicon for whichever system you select. On the ultrasound side, you choose the lesion type, then refine with size, inner wall, solid component, papillary projection count, outer contour, and color score; the calculator applies the v2022 decision rules and returns the category, the risk band, and the management pathway. On the MRI side, you specify the lesion type, the fluid content or fat, whether solid tissue enhances, the dark-T2/dark-DWI fibrous exit, and the time-intensity curve (or the myometrium-relative fallback when dynamic imaging is unavailable). Switching between US and MRI takes a single click, which is useful when an ultrasound flags an indeterminate lesion and you want to model how the MRI score would land.

The output follows structured reporting conventions: the selected descriptors, the calculated O-RADS category, the malignancy risk or PPV, and the corresponding management recommendation, formatted to drop into a PACS report. Because multiple lesions are scored independently and management is driven by the highest score, the tool handles the multi-lesion case the way the ACR governing concepts specify.

One practical note: the calculator is a reference tool for medical professionals, not a diagnostic substitute. An O-RADS category, from either system, is one input into a clinical decision — it must be read against the patient's history, symptoms, tumor markers, and the full imaging examination, not in isolation.

Limitations of O-RADS

O-RADS is a substantial improvement over unstructured reporting, but it has limits that users should understand.

First, inter-observer agreement is good but not perfect. Distinctions like "smooth versus irregular inner wall," the 3 mm threshold that separates a solid component from a wall irregularity, and the color score itself all involve judgment that standardized definitions narrow but do not eliminate. Two experienced readers can still disagree on a borderline lesion.

Second, the systems were developed and validated on adult, predominantly non-pregnant populations. Specific scenarios — pregnancy, the postmenopausal patient in whom physiologic findings are not expected, pediatric ovarian lesions, or masses in a patient with a known primary cancer where metastasis is on the differential — may require management beyond what the O-RADS category alone prescribes.

Third, the MRI time-intensity curve depends on adequate dynamic contrast-enhanced acquisition. When the protocol does not include proper dynamic imaging, the system falls back on a coarser myometrium-relative comparison, which is less discriminating. The quality of the score is bounded by the quality of the acquisition.

Finally, an O-RADS category stratifies risk; it does not by itself dictate surgery, surveillance, or referral. Those decisions weigh the category against tumor markers such as CA-125, menopausal status, symptoms, family history, and patient preference. The category narrows the question — it does not answer it. For a broader view of how O-RADS fits alongside the other reporting systems, see our RADS systems guide.

Disclaimer

The information provided in this article and through the Aperivue O-RADS calculator is intended for educational purposes and as a reference for healthcare professionals. It does not constitute medical advice, diagnosis, or treatment recommendations. Clinical decisions regarding ovarian and adnexal lesion management should be made by qualified physicians based on the complete clinical context. The calculator is a support tool — not a replacement for professional medical judgment, and not a regulatory-cleared medical device.