Radcalcs calculators are educational tools only. They do not constitute medical advice, diagnosis, or treatment. Always consult a qualified healthcare professional before making clinical decisions.
Nothing on this Site — including calculator outputs, interpretive text, scoring thresholds, reference ranges, or clinical guidance — constitutes medical advice. Medical advice is the application of medical knowledge to the specific circumstances of an individual patient by a licensed clinician who has evaluated that patient. This Site does not evaluate patients, does not know your patients, and does not provide medical advice of any kind.
Do not interpret any content on this Site as a recommendation for or against any diagnostic workup, treatment, medication, procedure, or clinical strategy for any individual patient. All clinical content on this Site exists for educational reference only.
All calculators, scoring tools, and clinical reference content on Radcalcs (radcalcs.org) are provided for informational and educational purposes only. These tools are designed to support clinician learning and workflow — they implement validated mathematical formulas from published peer-reviewed literature to help clinicians understand scoring systems, reference guidelines, and quantitative clinical tools.
In practical terms, "educational purposes only" means:
Use of this Site does not create a physician-patient relationship, provider-patient relationship, or any therapeutic relationship of any kind between you, any user, or any patient and Radcalcs or its operators, developers, or clinical advisors.
The absence of a physician-patient relationship is not a formality — it has real clinical implications. A physician-patient relationship carries obligations: to gather a full history, to examine the patient, to consider contraindications, to communicate findings, and to take responsibility for clinical decisions. None of those obligations exist here. Radcalcs is a reference tool, not a clinician.
If you are a member of the public (not a licensed clinician) seeking medical guidance for yourself or someone in your care, please consult a licensed healthcare provider. Do not use this Site to self-diagnose or make personal medical decisions.
Radcalcs calculators implement published equations and thresholds from authoritative clinical guidelines at the time of publication. However, clinical practice guidelines are living documents — they are updated as new evidence accumulates, as validation studies are published, and as professional societies revise their recommendations.
We revise calculators within 90 days of a known published guideline update. However, there may be a period during which a formula or threshold on this Site does not reflect the most current published recommendation. You are responsible for:
If you identify a discrepancy between a calculator on this Site and a recently published guideline, please contact us at radbridge@polsia.app. We investigate all reports within 48 hours.
Scores are tools, not diagnoses. A HEART score of 7 is not a diagnosis of acute coronary syndrome. A MELD score of 24 is not an organ allocation decision. A PI-RADS 4 is not a biopsy order. These are inputs to clinical reasoning by a qualified clinician who holds all the information about the patient — not outputs of clinical reasoning.
Specifically, these tools are not a substitute for:
Clinical judgment includes knowing when a validated scoring tool does not apply — when a patient falls outside the derivation cohort, has excluded comorbidities, or presents in a way the formula was not designed to handle. No calculator on this Site can tell you when it should not be used. That recognition is a clinical skill.
Qualified Professional Interpretation Required: All results generated by Radcalcs calculators must be reviewed, interpreted, and applied solely by a licensed, qualified healthcare professional. No calculator result should be used as the sole or primary basis for any clinical decision, diagnosis, treatment plan, or patient management strategy.
Beyond the general limitation that all calculators require clinical judgment, automated clinical calculators have specific structural limitations that clinicians should understand:
The following sections describe specific considerations for each category of calculator available on this Site.
Radiology reporting system calculators (TI-RADS, PI-RADS, BI-RADS) implement the published ACR lexicon and management guidelines. However, these systems assign categories based on imaging features that are inherently subject to interpreter variability. The same nodule may be scored differently by different radiologists; the calculator cannot resolve interpretive disagreement. ACR TI-RADS and PI-RADS management thresholds (size cutoffs for FNA, biopsy, and follow-up intervals) are population-level recommendations — individual patient factors (patient age, clinical context, prior imaging, patient preference) appropriately modify management. The Fleischner criteria apply to incidentally detected nodules in adults without prior lung cancer or immunocompromising conditions; they do not apply to known cancer patients, immunocompromised patients, or symptomatic patients. Coronary calcium scoring requires CT acquisition protocols optimized for calcium scoring — results from non-dedicated CT acquisitions may not be comparable to published Agatston percentile reference data.
Cardiac risk scores (HEART, CHA2DS2-VASc, ASCVD) are population-level risk stratification tools. A "low risk" score does not rule out the condition — it shifts pre-test probability. HEART score risk categories are based on 6-week MACE rates from derivation cohorts; local emergency department patient populations may have different baseline event rates. The ACC/AHA Pooled Cohort Equations for ASCVD risk were derived primarily from U.S. cohorts and have shown variable calibration in certain subpopulations; current ACC/AHA guidelines discuss these limitations and provide guidance on use in non-White patients. Wells scoring for PE and DVT produces pre-test probability estimates to guide diagnostic testing — it does not diagnose or exclude PE or DVT. D-dimer cutoffs and imaging thresholds in the Wells algorithms have been updated in various validation studies; verify that the thresholds used here match your institution's diagnostic protocol. Anticoagulation decisions based on CHA2DS2-VASc score must account for bleeding risk (e.g., HAS-BLED score) and patient preferences, which are not incorporated in this calculator.
CURB-65 and PSI/PORT Score are validated tools for severity stratification of community-acquired pneumonia — they were not derived or validated for hospital-acquired pneumonia, aspiration pneumonia, or pneumonia in immunocompromised hosts. Both scores are intended to support, not replace, clinical judgment about hospitalization. A low PSI class does not mandate outpatient treatment; high-risk social circumstances, inability to take oral medications, inadequate outpatient follow-up, or clinician concern are all legitimate reasons to hospitalize a low-score patient. The CURB-65 blood urea nitrogen threshold uses a UK reference (BUN > 7 mmol/L, equivalent to approximately > 19 mg/dL); verify your laboratory's units when interpreting this input. PSI/PORT Score is most validated in adults; it has not been validated for pediatric community-acquired pneumonia.
The CKD-EPI 2021 equation estimates GFR — it does not measure it. Measured GFR (mGFR) by gold-standard clearance methods (iohexol, inulin) is the definitive reference and may differ from eGFR in specific clinical scenarios. eGFR is less reliable at extreme body compositions (very low muscle mass, amputations, malnutrition, bodybuilding), in acute kidney injury (where creatinine is not at steady state), and in the setting of certain medications that affect creatinine secretion (e.g., trimethoprim, cimetidine). The CKD-EPI 2021 equation was developed to eliminate the race coefficient from the 2009 CKD-EPI equation; it was validated in predominantly U.S. cohorts and may perform differently in other populations. Kidney volume calculation for ADPKD uses the ellipsoidal formula; MRI-based stereology gives more accurate TKV than ultrasound-derived estimates, and imaging modality should be noted when tracking TKV over time.
MELD and MELD-Na scores are used by UNOS for liver transplant allocation and as a proxy for 90-day mortality in end-stage liver disease. MELD was derived in patients with cirrhosis undergoing TIPS procedures and has been validated broadly in patients with chronic liver disease; it performs less well in patients with hepatic malignancy and certain other conditions. Serum creatinine in patients with end-stage liver disease may be influenced by low muscle mass, leading to underestimation of renal dysfunction — MELD-Na partially addresses this but does not fully resolve it. Laboratory values used in MELD calculations (creatinine, bilirubin, INR) must be current (typically within 30 days for transplant listing purposes per UNOS policy); using outdated lab values produces a score that does not reflect current disease severity. Child-Pugh score incorporates semi-quantitative assessments of ascites and hepatic encephalopathy that carry inter-observer variability; the same patient may receive different Child-Pugh scores from different clinicians. Liver volume estimates from ellipsoidal measurements on cross-sectional imaging are approximations; volumetric software on dedicated 3D workstations provides more accurate pre-surgical planning volumes.
CT effective dose estimates using ICRP 103 conversion coefficients are population-average estimates — they do not represent the dose to any specific patient. Individual patient dose varies with body habitus, tube current modulation, and scanner-specific parameters. The DLP-to-effective-dose conversion coefficients used here are region-of-body averages from ICRP 103 and differ from scanner-specific Monte Carlo simulations; values should be used for comparative and quality assurance purposes, not precise dosimetry. ACR Dose Index Registry benchmarks represent national percentile distributions from contributing institutions and are updated periodically; the benchmarks used on this Site reflect the version in effect at time of publication. wRVU values reflect CMS Physician Fee Schedule published values for the specified year; actual contracted reimbursement rates depend on payer contracts, geographic adjustment factors (GPCI), and conversion factors that vary by payer and location. wRVU values are updated annually by CMS; values from prior years are not adjusted on this Site.
Organ volume calculators use the ellipsoidal formula (with organ-specific correction factors where applicable) applied to imaging measurements. Volumes calculated from two-dimensional measurements are approximations of true volumetric measurements; actual organ volumes may differ from calculated values due to irregular organ shapes. Reference ranges for organ sizes are derived from population studies and represent statistical distributions — an organ volume within the "normal range" does not rule out pathology, and a volume outside the range does not confirm it. Ovarian and uterine volume reference ranges vary significantly by age, hormonal status, parity, and menstrual phase; interpret in the full clinical and demographic context. Testicular volume reference values differ between the Lambert formula and the ellipsoidal formula; use a consistent method when tracking volume over time. All organ volume measurements are subject to measurement technique variability (probe angle, compression, identification of organ boundaries) that may produce variation between operators and imaging sessions.
Calculator formulas are implemented based on published peer-reviewed literature and recognized clinical guidelines (e.g., KDIGO, ACR, ICRP, CMS, AHA/ACC, Fleischner Society, UNOS). Our implementation process includes: selecting the primary source equation, implementing it directly from that source, and verifying outputs against published reference values or worked examples before deployment.
However:
Always cross-reference with current authoritative guidelines before applying any result in a clinical setting. Source citations are provided on each calculator page so you can access the primary literature directly.
To the fullest extent permitted by applicable law, Radcalcs and its operators disclaim all liability for any direct, indirect, incidental, or consequential harm arising from:
The user assumes all risk associated with applying any information from this site to clinical practice. See our full Terms of Service for the complete liability provisions.
Questions about this disclaimer or concerns about calculator accuracy? Contact us at radbridge@polsia.app. We welcome corrections from clinicians and investigate all error reports within 48 hours.