Whole-Body Parametric PET: Patlak Ki Imaging
Whole-body parametric PET converts a dynamic FDG acquisition into images of the metabolic influx rate constant Ki, using Patlak graphical analysis to map — voxel by voxel, across the whole body — the rate at which tracer is irreversibly trapped, rather than the single static snapshot that a standardized uptake value (SUV) provides. The payoff is improved lesion-to-background contrast in high-uptake organs and a quantitative measure of metabolism that is more directly tied to the underlying physiology.
For two decades, static whole-body FDG PET/CT with SUV has been the workhorse of oncologic imaging. It is fast, robust, and clinically validated. But SUV is a single number captured at one moment, and it mixes together trapped tracer, tracer still circulating in blood, and reversible background. Dynamic, kinetic imaging separates these components — and until recently it was confined to a single bed position because PET scanners could not image the whole body dynamically. Multi-bed dynamic protocols and, more recently, total-body PET have changed that, bringing Patlak parametric imaging into the whole-body clinical domain. 3, 4
This guide explains the Patlak model, how a whole-body dynamic acquisition is built, how Ki relates to SUV, and what the evidence says the method adds. DRPS supports quantitative PET/CT programs across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware through its PET/CT and nuclear medicine physics services.
Introduction
FDG is taken up by cells through glucose transporters, phosphorylated by hexokinase to FDG-6-phosphate, and then — unlike glucose — largely trapped, because it is a poor substrate for the next metabolic step. That trapping is what makes FDG such a useful tracer: the signal accumulates in metabolically active tissue over time.
A static SUV image captures the net result at a single time point. But the rate of trapping is the quantity most directly linked to glucose metabolism, and extracting that rate requires watching the tracer over time. Parametric imaging does exactly this: it fits a kinetic model to the time course of activity in each voxel and produces an image whose intensity is a physiological rate constant rather than a concentration. 1, 2
The Patlak plot is the graphical tool that makes this practical for an irreversibly trapped tracer like FDG. It avoids full compartmental fitting, is computationally light enough to run voxel by voxel across the whole body, and yields the net influx rate constant Ki directly as a slope. 1, 2 For the compartmental background that underlies it, see our guide to dynamic PET tracer kinetic modeling, and for the static metric it complements, see PET SUV quantification.
Topic Explanation
From dynamic data to a parametric image
A dynamic PET acquisition records activity as a series of time frames rather than one static image. For each voxel this produces a time-activity curve (TAC): how the measured concentration rises and falls from injection onward. Parametric imaging fits a model to every voxel's TAC and stores the fitted parameter — here, Ki — as the pixel value. The output is a "map" of a physiological rate across the body. 3, 4
Two inputs are needed for every voxel: the tissue TAC (from the images) and the input function, the time course of tracer concentration in plasma that drives uptake. In most whole-body implementations the input function is image-derived, sampled from a large blood pool such as the ascending aorta or left ventricle during an early dynamic phase over the heart, rather than from arterial blood sampling. 3
The irreversible-trapping assumption
The Patlak model applies when, after an initial equilibration period, the tracer is effectively trapped — flux into the irreversible compartment dominates and efflux is negligible over the measurement window. FDG satisfies this well in most tissues because dephosphorylation is slow. When efflux is not negligible (some tumors and tissues show measurable FDG dephosphorylation), a generalized Patlak model that includes an efflux rate constant improves accuracy, at some cost in noise. 5
Key Technical Principles
The Patlak equation
The Patlak graphical analysis linearizes the kinetics of an irreversibly trapped tracer. If
Plotting
Worked example. Suppose, on the linear portion of a lesion's Patlak plot, two late samples give:
- at normalized time
min, - at normalized time
min,
The influx rate constant is the slope:
and the intercept is:
So this voxel traps FDG at 0.05 mL of plasma cleared per minute per mL of tissue, with a reversible/vascular distribution volume of about 1.0. Ki can be converted to a metabolic rate of glucose by multiplying by the plasma glucose concentration and dividing by the lumped constant, which links the physical measurement to glucose metabolism. 1
Why Ki can beat SUV on contrast
Because the Patlak slope isolates the irreversibly trapped component, it subtracts away the reversible and vascular background that inflates SUV in well-perfused organs. In the liver — where high background FDG makes SUV lesion detection difficult — Ki images frequently show markedly higher tumor-to-background and contrast-to-noise ratios than SUV. 3, 5 This is the single most cited practical advantage of whole-body Patlak imaging.
| Property | Static SUV image | Patlak Ki parametric image |
|---|---|---|
| Data required | Single late time frame | Dynamic multi-frame acquisition |
| Physiological meaning | Tracer concentration (semi-quantitative) | Net irreversible influx rate constant |
| Input function | None | Plasma input (usually image-derived) |
| Background handling | Includes reversible + vascular signal | Separates trapped tracer from background |
| Acquisition time | Short (a few minutes per bed) | Longer (tens of minutes, protocol-dependent) |
| Noise | Lower | Higher (mitigated by model/denoising) |
| Typical strength | Robust, fast, validated | Improved contrast in high-background organs |
Standard, generalized, and relative Patlak
Three variants of the Patlak approach are relevant to whole-body work:
| Variant | Key feature | Trade-off |
|---|---|---|
| Standard Patlak (sPatlak) | Assumes complete irreversible trapping; slope = Ki | Simple and low-noise, but biased where efflux occurs |
| Generalized Patlak (gPatlak) | Adds an efflux rate constant |
More accurate for reversible uptake; ~10% noisier |
| Relative Patlak (RP) | Eliminates the need for the early-time input function; slope is Ki scaled by a per-subject factor | Enables short late-time scans; gives a scaled rather than absolute Ki |
A hybrid of standard and generalized Patlak has been proposed to balance quantification against lesion-detection contrast. 5 The relative Patlak method is what makes very short dynamic acquisitions feasible on total-body systems: it has produced high-quality parametric images from as little as a 20-minute list-mode scan, with the RP slope highly correlated with the standard Patlak Ki and the RP intercept equivalent to an SUV-relative-to-blood ratio. 10
Clinical Impact
Oncology
The most-studied application is oncologic FDG imaging. A prospective comparison of 18 oncologic studies reported that whole-body Patlak Ki imaging achieved equivalent or superior lesion detectability with fewer false positives than standard-of-care SUV, with one proven malignant lesion visible only on the Ki image and overall sensitivity and accuracy improving when Ki complemented SUV. 6 A dynamic whole-body study assessing lesion detectability with compartmental and Patlak mapping found that parametric metrics correlated strongly with reference kinetic estimates and that full compartmental analysis of the blood-pool bed could reveal lesions not visible on static SUV or Patlak images, underscoring the complementary nature of the methods. 7
Tumor characterization is a second theme. In primary breast cancer, mean Ki from dynamic whole-body parametric imaging was significantly higher in higher-grade and node-positive lesions and tracked biological markers more closely than SUV in several comparisons, suggesting Ki carries characterization information beyond uptake magnitude. 9
Shorter and smarter acquisitions
A practical barrier to clinical adoption has always been scan time and workflow. Research has steadily chipped away at this: acquisition protocols were optimized from about 45 minutes toward roughly 35 minutes without loss of image quality 3; deep-learning methods have been shown to infer Ki-Patlak images from fewer dynamic passes, shortening acquisition while preserving lesion detectability 8; and relative-Patlak plus denoising has produced parametric images from 20-minute total-body scans. 10 Motion over a long dynamic acquisition is a recognized challenge, and inter-pass motion correction has been shown to reduce Patlak fitting error and sharpen parametric lesion quantification. 11 The move to total-body PET with long axial fields of view is a natural fit, because the whole body can be imaged dynamically at once.
Practical Optimization Tips
- Use an adequate input function. Image-derived input from a large blood pool (aorta or left ventricle) during an early dynamic phase over the heart is standard; partial-volume and motion effects in the blood-pool ROI directly bias Ki. 3
- Respect the steady-state assumption. Only fit the linear (post-equilibration) portion of the Patlak plot; including early non-linear points biases the slope. 1, 2
- Choose the model to the task. Use generalized Patlak where efflux is expected and quantitative accuracy matters; favor standard or hybrid Patlak when lesion-detection contrast and noise are the priority. 5
- Plan for noise. Parametric images are noisier than SUV; appropriate reconstruction, smoothing, or model-based/deep-learning denoising is part of a usable Ki workflow. 8, 10
- Manage motion. For long dynamic acquisitions, account for inter-pass patient motion, which otherwise degrades parametric quantification. 11
- Report Ki as a complement, not a replacement. The strongest evidence supports Ki alongside SUV, especially for lesions in high-background organs. 6
- Standardize before comparing. As with SUV, consistent injection-to-scan timing, reconstruction, and analysis are prerequisites for comparing Ki across scans; see EARL harmonization.
Regulatory Considerations
Whole-body parametric PET uses an already-approved radiopharmaceutical (18F-FDG) and standard PET/CT hardware, so it does not create a new regulatory category, but it does raise quantitative accuracy and standardization expectations that the medical physicist supports. FDG is an FDA-approved radiopharmaceutical, and its medical use under a radioactive material license falls under NRC or Agreement State regulation (10 CFR Part 35 for NRC licensees). Parametric imaging does not change the administered activity or the licensing framework; it changes how the acquired data are processed.
What it does add is a heightened dependence on quantitative integrity: scanner calibration, cross-calibration with the dose calibrator, reconstruction consistency, and input-function accuracy all propagate directly into Ki. A quantitative PET program should maintain the same calibration and harmonization discipline used for SUV — see quantitative SPECT/CT and PET calibration practices for the analogous calibration logic — and should validate parametric workflows during commissioning. For the dose side of any PET/CT protocol, the CT component should still be optimized as covered in CT dose in SPECT/CT and PET/CT.
Facilities operating under NRC or Agreement State licenses should confirm that any protocol change is consistent with their license conditions and radiation safety program; of the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are Agreement States, while Washington DC and Delaware are regulated directly by the NRC.
Frequently Asked Questions (FAQs)
What is whole-body parametric PET?
Whole-body parametric PET is a quantitative imaging method that acquires a dynamic (time-resolved) FDG PET scan over the whole body and converts it into parametric images of a kinetic parameter — most often the metabolic influx rate constant Ki — using Patlak graphical analysis. Instead of a single static uptake value, each voxel reports the rate at which tracer is irreversibly trapped.
How is a Patlak Ki image different from an SUV image?
A standardized uptake value (SUV) image is a single snapshot of tracer concentration at one time point, normalized to injected dose and body weight. A Patlak Ki image uses multiple time frames and a blood input function to estimate the net irreversible influx rate of FDG, separating trapped tracer from reversible blood and tissue background. Ki often improves lesion-to-background contrast in organs with high background such as the liver.
What does the Patlak plot actually measure?
The Patlak plot linearizes the tracer kinetics for an irreversibly trapped tracer. When tissue-to-plasma concentration is plotted against the normalized time (the integral of plasma concentration divided by instantaneous plasma concentration), the data become linear once the reversible compartments reach steady state. The slope of that line is the net influx rate constant Ki and the intercept reflects the distribution volume of the reversible and vascular spaces.
Does whole-body parametric PET require an arterial blood sample?
Not usually. Most clinical implementations use an image-derived input function, sampling a large blood pool such as the aorta or left ventricle from an early dynamic scan over the heart, rather than arterial blood sampling. Newer relative-Patlak methods can estimate a scaled influx rate from a short late-time acquisition without the full early input function.
How long does a whole-body dynamic PET scan take?
Early whole-body Patlak protocols used total acquisitions of roughly 45 minutes, later optimized toward about 35 minutes while preserving image quality. On total-body PET systems with very long axial fields of view, and with relative-Patlak methods, high-quality parametric images have been generated from much shorter dynamic data, including from a 20-minute list-mode acquisition.
What are the main clinical advantages of Ki imaging?
Studies suggest Ki imaging can improve lesion detectability and reduce false positives relative to SUV in selected oncologic settings, particularly for lesions in high-background organs such as the liver, and can provide quantitative tumor characterization. It is best viewed as a complement to standard SUV imaging rather than a replacement.
Key Takeaways
- Patlak parametric imaging maps the net influx rate constant Ki — the rate of irreversible FDG trapping — voxel by voxel, rather than the single static concentration that SUV reports.
- The Patlak plot's slope is Ki and its intercept is the reversible/vascular distribution volume, valid on the linear portion after the reversible compartments reach steady state.
- Ki imaging improves lesion contrast in high-background organs such as the liver and can reduce false positives, complementing rather than replacing SUV.
- Most whole-body implementations use an image-derived input function, avoiding arterial sampling.
- Variants matter: generalized Patlak handles efflux, and relative Patlak enables short acquisitions by removing the need for the early input function.
- Quantitative discipline is essential: calibration, reconstruction consistency, input-function accuracy, and motion management all propagate directly into Ki.
Conclusion
Whole-body parametric PET is the maturation of a long-standing idea: that the rate of metabolism, not just a snapshot of uptake, is what FDG is best positioned to measure. Patlak graphical analysis makes that rate computable across the whole body, and advances in acquisition protocols, total-body scanners, relative-Patlak methods, and denoising have moved it from a single-bed research technique toward clinically feasible whole-body imaging.
The method is not a replacement for SUV, and it is not free — it demands dynamic acquisition, an accurate input function, careful modeling, and attention to noise and motion. But in the settings where it helps, especially lesions in high-background organs, the contrast and quantitative gains are real. A PET program considering parametric imaging should treat it as a quantitative capability to be commissioned and validated, with the same calibration rigor that underpins trustworthy SUV.
How DRPS Can Help
Diagnostic Radiation Physics Services supports nuclear medicine and PET/CT facilities in building and validating quantitative imaging programs. This includes PET/CT and nuclear medicine physics support for scanner calibration and cross-calibration, reconstruction and harmonization review, commissioning of advanced acquisition and processing workflows, and medical physicist consulting to align new protocols with license conditions and the radiation safety program.
DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware.
Related Resources
- Dynamic PET tracer kinetic modeling
- PET SUV quantification
- Total-body PET and long axial field of view
- EARL PET SUV harmonization
- PERCIST tumor response quantification
- Deep learning PET reconstruction and denoising
- PET/CT and nuclear medicine physics
- Medical physicist consulting
References
- Patlak CS, Blasberg RG, Fenstermacher JD. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. J Cereb Blood Flow Metab. 1983;3(1):1-7. doi:10.1038/jcbfm.1983.1. PubMed
- Patlak CS, Blasberg RG. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. Generalizations. J Cereb Blood Flow Metab. 1985;5(4):584-590. doi:10.1038/jcbfm.1985.87. PubMed
- Karakatsanis NA, Lodge MA, Tahari AK, Zhou Y, Wahl RL, Rahmim A. Dynamic whole-body PET parametric imaging: I. Concept, acquisition protocol optimization and clinical application. Phys Med Biol. 2013;58(20):7391-7418. doi:10.1088/0031-9155/58/20/7391. PubMed
- Karakatsanis NA, Lodge MA, Zhou Y, Wahl RL, Rahmim A. Dynamic whole-body PET parametric imaging: II. Task-oriented statistical estimation. Phys Med Biol. 2013;58(20):7419-7445. doi:10.1088/0031-9155/58/20/7419. PubMed
- Karakatsanis NA, Zhou Y, Lodge MA, Casey ME, Wahl RL, Zaidi H, Rahmim A. Generalized whole-body Patlak parametric imaging for enhanced quantification in clinical PET. Phys Med Biol. 2015;60(22):8643-8673. doi:10.1088/0031-9155/60/22/8643. PubMed
- Fahrni G, Karakatsanis NA, Di Domenicantonio G, Garibotto V, Zaidi H. Does whole-body Patlak 18F-FDG PET imaging improve lesion detectability in clinical oncology? Eur Radiol. 2019;29(9):4812-4821. doi:10.1007/s00330-018-5966-1. PubMed
- Zaker N, Kotasidis F, Garibotto V, Zaidi H. Assessment of lesion detectability in dynamic whole-body PET imaging using compartmental and Patlak parametric mapping. Clin Nucl Med. 2020;45(5):e221-e231. doi:10.1097/RLU.0000000000002954. PubMed
- Zaker N, Haddad K, Faghihi R, Arabi H, Zaidi H. Direct inference of Patlak parametric images in whole-body PET/CT imaging using convolutional neural networks. Eur J Nucl Med Mol Imaging. 2022;49(12):4048-4063. doi:10.1007/s00259-022-05867-w. PubMed
- Sundaraiya S, Raja T, Nangia S, Sirohi B, Patil S. Role of dynamic and parametric whole-body FDG PET/CT imaging in molecular characterization of primary breast cancer: a single institution experience. Nucl Med Commun. 2022;43(9):1015-1025. doi:10.1097/MNM.0000000000001596. PubMed
- Li S, Abdelhafez YG, Nardo L, Cherry SR, Badawi RD, Wang G. Total-body parametric imaging using relative Patlak plot. J Nucl Med. 2025;66(4):654-661. doi:10.2967/jnumed.124.268496. PubMed
- Guo X, Wu J, Chen MK, et al. Inter-pass motion correction for whole-body dynamic PET and parametric imaging. IEEE Trans Radiat Plasma Med Sci. 2023;7(4):344-353. doi:10.1109/trpms.2022.3227576. PubMed
Related Articles
-
F-18 PSMA PET/CT: Piflufolastat Imaging
F-18 PSMA agents (piflufolastat, flotufolastat) bring a 110-minute half-life and short positron range to prostate cancer PET. Physics, SUV, and QC.
-
F-18 Sodium Fluoride Bone PET/CT: Physics & QC
F-18 sodium fluoride (NaF) PET/CT images bone metabolism with high resolution. Learn the tracer physics, SUV quantification, dosimetry, and QC essentials.
-
MR Elastography: Physics and Quality Control
MR elastography maps liver stiffness in kilopascals to stage fibrosis noninvasively. Learn the shear-wave physics, worked math, and QC that keep it reliable.