Reducing scan time in ¹⁷⁷Lu planar scintigraphy using convolutional neural network: A Monte Carlo simulation study

Abstract

Purpose

The aim of this study was to reduce scan time in ¹⁷⁷Lu planar scintigraphy through the use of convolutional neural network (CNN) to facilitate personalized dosimetry for ¹⁷⁷Lu‐based peptide receptor radionuclide therapy.

Methods

The CNN model used in this work was based on DenseNet, and the training and testing datasets were generated from Monte Carlo simulation. The CNN input images (IMG _input) consisted of ¹⁷⁷Lu planar scintigraphy that contained 10–90% of the total photon counts, while the corresponding full‐count images (IMG ^100%) were used as the CNN label images. Two‐sample t‐test was conducted to compare the difference in pixel intensities within region of interest between IMG ^100% and CNN output images (IMG _output).

Results

No difference was found in IMG _output for rods with diameters ranging from 13 to 33 mm in the Derenzo phantom with a target‐to‐background ratio of 20:1, while statistically significant differences were found in IMG _output for the 10‐mm diameter rods when IMG _input containing 10% to 60% of the total photon counts were denoised. Statistically significant differences were found in IMG _output for both right and left kidneys in the NCAT phantom when IMG _input containing 10% of the total photon counts were denoised. No statistically significant differences were found in IMG _output for any other source organs in the NCAT phantom.

Conclusion

Our results showed that the proposed method can reduce scan time by up to 70% for objects larger than 13 mm, making it a useful tool for personalized dosimetry in ¹⁷⁷Lu‐based peptide receptor radionuclide therapy in clinical practice.