Search

Background: The reason for declining risk to psychosis across individuals assessed and meeting Ultra High Risk (UHR) criteria is still unclear. No studies have investigated the potential substantial role of the underlying risk enrichment across all the individuals undergoing an UHR assessment. Methods: Cohort study including all non-psychotic subjects who were assessed on suspicion of psychosis risk by the OASIS UHR service in the period 2001 to 2015. Posttest (after UHR assessment) and pretest risk (before UHR assessment) of psychosis were stratified and compared across three time periods (2001–2005, 2006–2010, 2011–2015) with Cox analysis and modulating factors were investigated. Results: The posttest risk of psychosis at the OASIS service has increased from the initial pilot years of the service (2001–2005) and then stabilised and not declined over the following decade (2006–2010 and 2011–2015). This was paralleled by a similar course of pretest risk for psychosis. Stability of pretest risk for psychosis over the past decade was associated with a lack of change in ethnicity and to counterweighting changes in the type of referral sources over different time periods. Conclusions: The time course of transition risk to psychosis in UHR services is strictly associated with the time course of pretest risk enrichment. If the latter remains stable over time, as for the OASIS service, no declining transition risk is observed over the most recent years. Pretest risk enrichment is determined by recruitment and sampling strategies. This study confirms the need to control these factors in the UHR field.

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 [18F]-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 [18F]-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 [18F]-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 [18F]-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.