Highlights d Imaging transcriptomics recovers plausible correlates of several neuroimaging markers d Its use to capture the neuroimmune axis in the healthy brain remains inconclusive d This approach might help to dissect regional variability in neuroimaging findings Authors

The Brain Imaging Data Structure (BIDS) is a standard for organizing and describing neuroimaging datasets, serving not only to facilitate the process of data sharing and aggregation, but also to simplify the application and development of new methods and software for working with neuroimaging data. Here, we present an extension of BIDS to include positron emission tomography (PET) data, also known as PET-BIDS, and share several open-access datasets curated following PET-BIDS along with tools for conversion, validation and analysis of PET-BIDS datasets.

Introduction. With biomedical imaging research increasingly using large datasets, it becomes critical to find operator-free methods to quality control the data collected and the associated analysis. Attempts to use artificial intelligence (AI) to perform automated quality control (QC) for both single-site and multi-site datasets have been explored in some neuroimaging techniques (e.g. EEG or MRI), although these methods struggle to find replication in other domains. The aim of this study is to test the feasibility of an automated QC pipeline for brain [ 18 F]-FDOPA PET imaging as a biomarker for the dopamine system. Methods. Two different Convolutional Neural Networks (CNNs) were used and combined to assess spatial misalignment to a standard template and the signal-to-noise ratio (SNR) relative to 200 static [ 18 F]-FDOPA PET images that had been manually quality controlled from three different PET/CT scanners. The scans were combined with an additional 400 scans, in which misalignment (200 scans) and low SNR (200 scans) were simulated. A cross-validation was performed, where 80% of the data were used for training and 20% for validation. Two additional datasets of [ 18 F]-FDOPA PET images (50 and 100 scans respectively with at least 80% of good quality images) were used for out-of-sample validation. Results . The CNN performance was excellent in the training dataset (accuracy for motion: 0.86 ± 0.01, accuracy for SNR: 0.69 ± 0.01), leading to 100% accurate QC classification when applied to the two out-of-sample datasets. Data dimensionality reduction affected the generalizability of the CNNs, especially when the classifiers were applied to the out-of-sample data from 3D to 1D datasets. Conclusions . This feasibility study shows that it is possible to perform automatic QC of [ 18 F]-FDOPA PET imaging with CNNs. The approach has the potential to be extended to other PET tracers in both brain and non-brain applications, but it is dependent on the availability of large datasets necessary for the algorithm training.

The Brain Imaging Data Structure (BIDS) is a standard for organizing and describing neuroimaging datasets. It serves not only to facilitate the process of data sharing and aggregation, but also to simplify the application and development of new methods and software for working with neuroimaging data. Here, we present an extension of BIDS to include positron emission tomography (PET) data (PET-BIDS). We describe the PET-BIDS standard in detail and share several open-access datasets curated following PET-BIDS. Additionally, we highlight several tools which are already available for converting, validating and analyzing PET-BIDS datasets.