Related Experiment Videos
Maintaining SUV accuracy in low-count PET with PETfectior: a deep learning denoising solution
Yamila Rotstein Habarnau1, Nicolás Bustos1, Paola Corona1
1Fundación Centro Diagnóstico Nuclear (FCDN), Buenos Aires C1417CVE, Argentina.
Biomedical Physics & Engineering Express
|July 14, 2026
Summary
PETfectior, an AI software, enhances low-count positron emission tomography (PET) scans. It achieves high diagnostic image quality with reduced radiotracer activity, lowering patient radiation exposure and costs.
Area of Science:
- Nuclear Medicine
- Medical Imaging
- Artificial Intelligence
Background:
- Positron emission tomography (PET) image quality is limited by radiotracer activity and acquisition time, impacting radiation exposure and costs.
- Reducing these parameters is crucial for patient safety and economic viability in PET examinations.
- Current standards, like EARL, guide quantitative PET imaging, but achieving them often requires higher activity and longer scans.
Purpose of the Study:
- To clinically validate PETfectior, an AI software designed to improve signal-to-noise ratio in low-count PET scans.
- To assess PETfectior's performance in lesion detection, quantitative accuracy, and image quality using reduced counting statistics.
- To evaluate the potential of PETfectior in enabling high-quality PET imaging with halved radiation dose and cost.
Main Methods:
- Prospective inclusion of 258 patients undergoing 18F-FDG PET/CT scans.
- Acquisition and reconstruction of standard-of-care scans (100% statistics) following EARL standards.
- Generation of half-counting-statistics (50%) PET images processed by PETfectior.
- Evaluation of lesion detectability, SUVmax quantification, and subjective image quality by experienced physicians.
Main Results:
- PETfectior processed 50% counting statistics images demonstrated high lesion detection sensitivity (99.9%) with minimal false positives.
- Quantitative analysis showed good concordance for SUVmax between 100% and 50% + PETfectior images (±12.5% 95% LOA, -1.01% bias).
- Subjective assessment revealed no significant difference in image quality between standard and PETfectior-processed reduced-statistics images.
Conclusions:
- PETfectior can be safely implemented in clinical practice for 18F-FDG PET imaging at half counting statistics.
- The AI software maintains high sensitivity and specificity, with acceptable quantitative accuracy and excellent subjective image quality.
- PETfectior offers a promising solution for reducing radiation dose and costs in PET examinations without compromising diagnostic performance.
Related Concept Videos
Deconvolution
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Reducing Line Loss
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss in...
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss in...
Downsampling
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
Upsampling
Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...
Differential Leveling
Differential leveling is a precise method in surveying used to determine the elevation difference between two points. Its primary goal is to establish accurate vertical measurements to create level surfaces or grade lines critical for designing and constructing infrastructures such as roads, bridges, and buildings.The procedure for differential leveling begins with setting up and leveling the instrument at a point where the benchmark can be seen. The level rod is held on the benchmark (BM), and...