Automated quality control in image segmentation: application to the UK Biobank cardiovascular magnetic resonance imaging study.

BACKGROUND: The trend towards large-scale studies including population imaging poses new challenges in terms of quality control (QC). This is a particular issue when automatic processing tools such as image segmentation methods are employed to derive quantitative measures or biomarkers for further analyses. Manual inspection and visual QC of each segmentation result is not feasible at large scale. However, it is important to be able to automatically detect when a segmentation method fails in order to avoid inclusion of wrong measurements into subsequent analyses which could otherwise lead to incorrect conclusions. METHODS: To overcome this challenge, we explore an approach for predicting segmentation quality based on Reverse Classification Accuracy, which enables us to discriminate between successful and failed segmentations on a per-cases basis. We validate this approach on a new, large-scale manually-annotated set of 4800 cardiovascular magnetic resonance (CMR) scans. We then apply our method to a large cohort of 7250 CMR on which we have performed manual QC. RESULTS: We report results used for predicting segmentation quality metrics including Dice Similarity Coefficient (DSC) and surface-distance measures. As initial validation, we present data for 400 scans demonstrating 99% accuracy for classifying low and high quality segmentations using the predicted DSC scores. As further validation we show high correlation between real and predicted scores and 95% classification accuracy on 4800 scans for which manual segmentations were available. We mimic real-world application of the method on 7250 CMR where we show good agreement between predicted quality metrics and manual visual QC scores. CONCLUSIONS: We show that Reverse classification accuracy has the potential for accurate and fully automatic segmentation QC on a per-case basis in the context of large-scale population imaging as in the UK Biobank Imaging Study.

Acute liver effects, disposition and metabolic fate of [14C]-fenclozic acid following oral administration to normal and bile-cannulated male C57BL/6J mice.

The distribution, metabolism, excretion and hepatic effects of the human hepatotoxin fenclozic acid were investigated following single oral doses of 10 mg/kg to normal and bile duct-cannulated male C57BL/6J mice. Whole body autoradiography showed distribution into all tissues except the brain, with radioactivity still detectable in blood, kidney and liver at 72 h post-dose. Mice dosed with [14C]-fenclozic acid showed acute centrilobular hepatocellular necrosis, but no other regions of the liver were affected. The majority of the [14C]-fenclozic acid-related material recovered was found in the urine/aqueous cage wash, (49%) whilst a smaller portion (13%) was eliminated via the faeces. Metabolic profiles for urine, bile and faecal extracts, obtained using liquid chromatography and a combination of mass spectrometric and radioactivity detection, revealed extensive metabolism of fenclozic acid in mice that involved biotransformations via both oxidation and conjugation. These profiling studies also revealed the presence of glutathione-derived metabolites providing evidence for the production of reactive species by mice administered fenclozic acid. Covalent binding to proteins from liver, kidney and plasma was also demonstrated, although this binding was relatively low (less than 50 pmol eq./mg protein).

The metabolic fate of [14C]-fenclozic acid in the hepatic reductase null (HRN) mouse.

1. The distribution, metabolism, excretion and hepatic effects of fenclozic acid were investigated following a single oral dose of 10 mg/kg to hepatic reductase null (HRN) mice. 2. The majority of the [(14)C]-fenclozic acid was eliminated via the urine/aqueous cage wash, (55%) with a smaller portion excreted in the faeces, (5%). The total recovery of radioactivity in the excreta over the 72 h period studied was ca. 60%. 3. Metabolism of fenclozic acid in the HRN mice was entirely to the carboxylic acid function and was dominated by amino acid conjugation to glycine and taurine, with lesser amounts of an acyl glucuronide. 4. Whole body autoradiography of mice showed general distribution into all tissues except the brain. Radioactivity was still detectable in the kidney and liver of the HRN mice at 72 h post-dose. Covalent binding studies showed evidence of binding to kidney, liver and plasma proteins however, the degree of binding was less than 50 pmol equiv/mg protein for all tissues. 5. The HRN mouse appears to be a useful in vivo model for the study of the Phase II conjugation metabolism of fenclozic acid in the absence of hepatic cytochrome P450-related oxidative metabolism.

Diclofenac metabolism in the mouse: novel in vivo metabolites identified by high performance liquid chromatography coupled to linear ion trap mass spectrometry.

The metabolism of [(14)C]-diclofenac in mice was investigated following a single oral dose of 10 mg/kg. The majority of the drug-related material was excreted in the urine within 24 h of administration (49.7 %). Liquid chromatographic analyses of urine and faecal extracts revealed extensive metabolism to at least 37 components, with little unchanged diclofenac excreted. Metabolites were identified using a hybrid linear ion-trap mass spectrometer via exact mass determinations of molecular ions and subsequent multi-stage fragmentation. The major routes of metabolism identified included: 1) conjugation with taurine; and 2) hydroxylation (probably at the 4'-and 5-arene positions) followed by conjugation to taurine, glucuronic acid or glucose. Ether, rather than acyl glucuronidation, predominated. There was no evidence for p-benzoquinone-imine formation (i.e. no glutathione or mercapturic acid conjugates were detected). A myriad of novel minor drug-related metabolites were also detected, including ribose, glucose, sulfate and glucuronide ether-linked conjugates of hydroxylated diclofenac derivatives. Combinations of these hydroxylated derivatives with acyl conjugates (glucose, glucuronide and taurine) or N-linked sulfation or glucosidation were also observed. Acyl- or amide-linked-conjugates of benzoic acid metabolites and several indolinone derivatives with further hydroxylated and conjugated moieties were also evident. The mechanisms involved in the generation of benzoic acid and indolinone products indicate the formation reactive intermediates in vivo that may possibly contribute to hepatotoxicity.

The hepatic reductase null mouse as a model for exploring hepatic conjugation of xenobiotics: application to the metabolism of diclofenac.

The distribution, metabolism, excretion and hepatic effects of diclofenac were investigated following a single oral dose of 10 mg/kg to wild type and hepatic reductase null (HRN) mice. For the HRN strain the bulk of the [(14)C]-diclofenac-related material was excreted in the urine/aqueous cagewash within 12 h of administration (~82%) with only small amounts eliminated via the faeces (~2% in 24 h). Wild type mice excreted the radiolabel more slowly with ca. 52 and 15% of the dose recovered excreted in urine and faeces, respectively, by 24 h post dose. The metabolic profiles of the HRN mice were dominated by acyl conjugation to either taurine or glucuronic acid. Wild type mice produced relatively small amounts of the acyl glucuronide. Whole Body Autoradiography (WBA) of mice sacrificed at 24 h post dose indicated increased retention of radioactivity in the livers of HRN mice compared to wild type mice. Covalent binding studies showed no differences between the two strains. Metabolism of diclofenac in HRN mice involved mainly acyl glucuronide formation and taurine amide conjugation. This mouse model may find utility in understanding the impact of reactive metabolite formation via routes that involve the production of acyl-CoA or acyl glucuronides of acidic drugs.

Extended-Release Guanfacine for Hyperactivity in Children With Autism Spectrum Disorder.

OBJECTIVE: Hyperactivity, impulsiveness, and distractibility are common problems in children with autism spectrum disorder (ASD). Extended-release guanfacine is approved for children with attention deficit hyperactivity disorder but not well studied in ASD. METHOD: In a multisite, randomized clinical trial, extended-release guanfacine was compared with placebo in children with ASD accompanied by hyperactivity, impulsiveness, and distractibility. RESULTS: Sixty-two subjects (boys, N=53; girls, N=9; mean age=8.5 years [SD=2.25]) were randomly assigned to guanfacine (N=30) or placebo (N=32) for 8 weeks. The guanfacine group showed a 43.6% decline in scores on the Aberrant Behavior Checklist-hyperactivity subscale (least squares mean from 34.2 to 19.3) compared with a 13.2% decrease in the placebo group (least squares mean from 34.2 to 29.7; effect size=1.67). The rate of positive response (much improved or very much improved on the Clinical Global Impression-Improvement scale) was 50% (15 of 30) for guanfacine compared with 9.4% (3 of 32) for placebo. A brief cognitive battery tapping working memory and motor planning showed no group differences before or after 8 weeks of treatment. The modal dose of guanfacine at week 8 was 3 mg/day (range: 1-4 mg/day), and the modal dose was 3 mg/day (range: 2-4 mg/day) for placebo. Four guanfacine-treated subjects (13.3%) and four placebo subjects (12.5%) exited the study before week 8. The most common adverse events included drowsiness, fatigue, and decreased appetite. There were no significant changes on ECG in either group. For subjects in the guanfacine group, blood pressure declined in the first 4 weeks, with return nearly to baseline by endpoint (week 8). Pulse rate showed a similar pattern but remained lower than baseline at endpoint. CONCLUSIONS: Extended-release guanfacine appears to be safe and effective for reducing hyperactivity, impulsiveness, and distractibility in children with ASD.

Computational study on the origin of the stereoselectivity for the photochemical denitrogenation of diazabicycloheptene.

The origin of the inversion stereoselectivity of housane formation via photochemical nitrogen extrusion of diazabicycloheptene (DBH) has been investigated using reaction path computations and multireference second-order perturbation theory within a CASPT2//CASSCF scheme. We show that the primary photoproduct of the reaction is an exo-axial conformer of the diazenyl diradical ((1) DZ) which displays a cyclopenta-1,3-diyl moiety with a Cs-like structure. (1) DZ is selectively generated via decay at a linear-axial conical intersection, and it is located in a shallow region of the ground state potential energy surface that provides access to five different reaction pathways. Reaction path analysis (including probing with classical trajectories) indicates that production of inverted housane can only occur via impulsive population of an axial-to-equatorial pathway, and it is thus inconsistent with thermal equilibration of the primary (1) DZ conformer. Similarly, according to the same analysis, the decrease of inversion stereoselectivity and even the retention (stereochemical memory effect) observed for suitably substituted DBHs are explained by dynamics effects where the axial-to-equatorial impulsive motion is restrained by the inertia and/or steric hindrance of the substituents. These results shade light on the poorly understood mechanisms that allow a photochemical reaction, in which a large amount of energy is deposited in the reactant by photon absorption, to show a high degree of stereoselectivity.