MLPET: a localized neural network approach for probabilistic post-reconstruction PET image analysis using informed priors.

PURPOSE: PET images often make small lesions difficult to identify because of noise and system blur. We address this by developing and evaluating MLPET , a fast localized machine-learning method that approximates a computationally expensive probabilistic sampling approach while reducing noise and increasing spatial resolution.

METHODS

Building on a probabilistic deconvolution framework with informed priors, MLPET replaces computationally demanding Markov chain Monte Carlo sampling with a localized neural network trained to directly estimate the posterior mean of voxel activity from small image neighborhoods. The method incorporates scanner-specific point spread functions (PSF), spatially correlated noise modeling, and flexible prior information. Performance was evaluated on NEMA phantom data acquired on three PET systems (GE Discovery MI, Siemens Biograph Vision 600, and Siemens Biograph Vision Quadra Edge) under varying reconstruction settings and acquisition times.

RESULTS

On NEMA IEC phantom data, MLPET produced contrast-recovery coefficients consistently higher than PET and frequently close to 1.0 (including the 10 mm sphere in several settings), while simultaneously reducing background noise and improving spatial definition. The effective point-spread function (PSF) full width at half maximum (FWHM) was on average reduced from about 2 mm in PET to below 1 mm with MLPET , corresponding to roughly a 2.5× decrease in effective blur. Comparable image quality was obtained at 40-80 s acquisition time using MLPET versus 900 s with conventional PET. A clinical example in a breast cancer patient illustrates potential clinical applicability.

CONCLUSIONS

MLPET provides a computationally efficient approach for quantitative probabilistic post-reconstruction PET image analysis. By combining informed priors with the speed of a neural network, it achieves both noise suppression and resolution enhancement. The method shows promise for improved small-lesion detectability and quantitative reliability in clinical PET imaging. Future clinical studies will evaluate its performance on patient data and quantify effects under realistic prior uncertainty.