PT - JOURNAL ARTICLE AU - Sari, Hasan AU - Reaungamornrat, Ja AU - Catalano, Onofrio A. AU - Vera-Olmos, Javier AU - Izquierdo-Garcia, David AU - Morales, Manuel A. AU - Torrado-Carvajal, Angel AU - Ng, Thomas S.C. AU - Malpica, Norberto AU - Kamen, Ali AU - Catana, Ciprian TI - Evaluation of Deep Learning–Based Approaches to Segment Bowel Air Pockets and Generate Pelvic Attenuation Maps from CAIPIRINHA-Accelerated Dixon MR Images AID - 10.2967/jnumed.120.261032 DP - 2022 Mar 01 TA - Journal of Nuclear Medicine PG - 468--475 VI - 63 IP - 3 4099 - http://jnm.snmjournals.org/content/63/3/468.short 4100 - http://jnm.snmjournals.org/content/63/3/468.full SO - J Nucl Med2022 Mar 01; 63 AB - Attenuation correction remains a challenge in pelvic PET/MRI. In addition to the segmentation/model-based approaches, deep learning methods have shown promise in synthesizing accurate pelvic attenuation maps (μ-maps). However, these methods often misclassify air pockets in the digestive tract, potentially introducing bias in the reconstructed PET images. The aims of this work were to develop deep learning–based methods to automatically segment air pockets and generate pseudo-CT images from CAIPIRINHA-accelerated MR Dixon images. Methods: A convolutional neural network (CNN) was trained to segment air pockets using 3-dimensional CAIPIRINHA-accelerated MR Dixon datasets from 35 subjects and was evaluated against semiautomated segmentations. A separate CNN was trained to synthesize pseudo-CT μ-maps from the Dixon images. Its accuracy was evaluated by comparing the deep learning–, model-, and CT-based μ-maps using data from 30 of the subjects. Finally, the impact of different μ-maps and air pocket segmentation methods on the PET quantification was investigated. Results: Air pockets segmented using the CNN agreed well with semiautomated segmentations, with a mean Dice similarity coefficient of 0.75. The volumetric similarity score between 2 segmentations was 0.85 ± 0.14. The mean absolute relative changes with respect to the CT-based μ-maps were 2.6% and 5.1% in the whole pelvis for the deep learning–based and model-based μ-maps, respectively. The average relative change between PET images reconstructed with deep learning–based and CT-based μ-maps was 2.6%. Conclusion: We developed a deep learning–based method to automatically segment air pockets from CAIPIRINHA-accelerated Dixon images, with accuracy comparable to that of semiautomatic segmentations. The μ-maps synthesized using a deep learning–based method from CAIPIRINHA-accelerated Dixon images were more accurate than those generated with the model-based approach available on integrated PET/MRI scanners.