Apparatus and method for medical image reconstruction using deep learning to improve image quality in positron emission tomography (PET)

An apparatus, comprising:

• processing circuitry configured to
• acquire a reconstructed positron emission tomography (PET) image, the reconstructed PET image being reconstructed from emission data representing coincidence counts of respective pairs of gamma rays arising from electron-positron annihilation events, the coincidence counts being detected at a plurality of detector elements, and
• acquire a neural network including weighting coefficients of connections between neuronal nodes of respective layers of a plurality of layers between an input layer and an output layer of the neural network, the neural network having been trained using a training dataset that, for a given noise-minimized reconstructed data, includes two or more noise-exhibiting reconstructed data having greater noise levels than in the noise-minimized reconstructed data, the two or more noise-exhibiting reconstructed data being reconstructed using subsets of a full PET dataset used to reconstruct the given noise-minimized reconstructed data, and the training of the neural network including
• reducing a value produced by a loss function;
• and
• applying the reconstructed PET image to the acquired neural network to generate a noise reduced image, wherein
• the processing circuitry trains using a combination of a low-quality PET image and a low-quality image of another medical imaging modality to generate a high-quality PET image, and
• the processing circuitry incorporates high resolution anatomical information obtained from the another medical imaging modality into the neural network to correct for partial volume effects.

An apparatus, comprising:

• processing circuitry configured to
• acquire a reconstructed positron emission tomography (PET) image, the reconstructed PET image being reconstructed from emission data representing coincidence counts of respective pairs of gamma rays arising from electron-positron annihilation events, the coincidence counts being detected at a plurality of detector elements, and
• acquire a neural network including weighting coefficients of connections between neuronal nodes of respective layers of a plurality of layers between an input layer and an output layer of the neural network, the neural network having been trained using a training dataset that, for a given noise-minimized reconstructed data, includes two or more noise-exhibiting reconstructed data having greater noise levels than in the noise-minimized reconstructed data, the two or more noise-exhibiting reconstructed data being reconstructed using subsets of a full PET dataset used to reconstruct the given noise-minimized reconstructed data, and the training of the neural network including
• reducing a value produced by a loss function;
• and
• applying the reconstructed PET image to the acquired neural network to generate a noise reduced image, wherein
• the processing circuitry is further configured to acquire the neural network wherein the neural network has been trained using a plurality of noise-minimized reconstructed data, each of the plurality of noise-minimized reconstructed data being paired with each of two or more noise-exhibiting reconstructed data that are respectively generated by reconstructing PET images using subsets of the full PET dataset that is used to reconstruct the each of plurality of noise-minimized reconstructed data,
• the subsets are selected to produce the two or more noise-exhibiting reconstructed data that span a predefined range of statistical properties corresponding to images reconstructed based on PET scans performed using the plurality of detector elements, and
• when the reconstructed image to which the acquired neural network is applied is within the predefined range of statistical properties, an image quality of the noise reduced image is less affected by statistical properties of the reconstructed PET image than if the noise reduced image were generated using a neural network trained using noise-exhibiting reconstructed data that all had a same value for the statistical properties.

A method, comprising:

• acquiring a reconstructed positron emission tomography (PET) image, the reconstructed PET image reconstructed from emission data representing coincidence counts of respective pairs of gamma rays arising from electron-positron annihilation events, the coincidence counts being detected at a plurality of detector elements;
• and
• acquiring a neural network including weighting coefficients of connections between neuronal nodes of respective layers of a plurality of layers between an input layer and an output layer of the neural network, the neural network having been trained using a training dataset that, for a given noise-minimized reconstructed data, includes two or more noise-exhibiting reconstructed data having greater levels of noise than in the noise-minimized reconstructed data, the two or more noise-exhibiting reconstructed data being reconstructed using subsets of a full PET dataset used to reconstruct the given noise-minimized reconstructed data, and the training of the neural network including
• reducing a value produced by a loss function;
• optimizing a loss function representing respective differences between the given noise-minimized reconstructed data and each of the two or more noise-exhibiting reconstructed data;
• and
• applying the reconstructed image to the acquired neural network to generate a noise reduced image,
• wherein the method comprises training using a combination of a low-quality PET image and a low-quality image of another medical imaging modality to generate a high-quality PET image and the loss function is calculated by comparing the low-quality PET image and the low-quality image of the another medical imaging modality to respective high-quality PET image and high-quality image of the another medical imaging modality.

A method, comprising:

• acquiring a reconstructed positron emission tomography (PET) image, the reconstructed PET image reconstructed from emission data representing coincidence counts of respective pairs of gamma rays arising from electron-positron annihilation events, the coincidence counts being detected at a plurality of detector elements;
• and
• acquiring a neural network including weighting coefficients of connections between neuronal nodes of respective layers of a plurality of lavers between an input layer and an output layer of the neural network, the neural network having been trained using a training dataset that, for a given noise-minimized reconstructed data, includes two or more noise-exhibiting reconstructed data having greater levels of noise than in the noise-minimized reconstructed data, the two or more noise-exhibiting reconstructed data being reconstructed using subsets of a full PET dataset used to reconstruct the given noise-minimized reconstructed data, and the training of the neural network including
• reducing a value produced by a loss function;
• optimizing a loss function representing respective differences between the given noise-minimized reconstructed data and each of the two or more noise-exhibiting reconstructed data, and
• applying the reconstructed image to the acquired neural network to generate a noise reduced image, wherein
• the method comprises training using a combination of a low-quality PET image and a low-quality image of another medical imaging modality to generate a high-quality PET image and
• the method comprises incorporating high resolution anatomical information obtained from the another medical imaging modality into the neural network to correct for partial volume effects.