Differences Between MR Brain Region Segmentation Methods: Impact on Single-Subject Analysis

For the segmentation of magnetic resonance brain images into anatomical regions, numerous fully automated methods have been proposed and compared to reference segmentations obtained manually. However, systematic differences might exist between the resulting segmentations, depending on the segmentation method and underlying brain atlas. This potentially results in sensitivity differences to disease and can further complicate the comparison of individual patients to normative data. In this study, we aim to answer two research questions: 1) to what extent are methods interchangeable, as long as the same method is being used for computing normative volume distributions and patient-specific volumes? and 2) can different methods be used for computing normative volume distributions and assessing patient-specific volumes? To answer these questions, we compared volumes of six brain regions calculated by five state-of-the-art segmentation methods: Erasmus MC (EMC), FreeSurfer (FS), geodesic information flows (GIF), multi-atlas label propagation with expectation–maximization (MALP-EM), and model-based brain segmentation (MBS). We applied the methods on 988 non-demented (ND) subjects and computed the correlation (PCC-v) and absolute agreement (ICC-v) on the volumes. For most regions, the PCC-v was good (>0.75), indicating that volume differences between methods in ND subjects are mainly due to systematic differences. The ICC-v was generally lower, especially for the smaller regions, indicating that it is essential that the same method is used to generate normative and patient data. To evaluate the impact on single-subject analysis, we also applied the methods to 42 patients with Alzheimer’s disease (AD). In the case where the normative distributions and the patient-specific volumes were calculated by the same method, the patient’s distance to the normative distribution was assessed with the z-score. We determined the diagnostic value of this z-score, which showed to be consistent across methods. The absolute agreement on the AD patients’ z-scores was high for regions of thalamus and putamen. This is encouraging as it indicates that the studied methods are interchangeable for these regions. For regions such as the hippocampus, amygdala, caudate nucleus and accumbens, and globus pallidus, not all method combinations showed a high ICC-z. Whether two methods are indeed interchangeable should be confirmed for the specific application and dataset of interest.

Towards automated extraction of 2D standard fetal head planes from 3D ultrasound acquisitions: A clinical evaluation and quality assessment comparison

Introduction

Clinical evaluation of deep learning (DL) tools is essential to compliment technical accuracy metrics. This study assessed the image quality of standard fetal head planes automatically-extracted from three-dimensional (3D) ultrasound fetal head volumes using a customised DL-algorithm.

Methods

Two observers retrospectively reviewed standard fetal head planes against pre-defined image quality criteria. Forty-eight images (29 transventricular, 19 transcerebellar) were selected from 91 transabdominal fetal scans (mean gestational age = 26 completed weeks, range = 20⁺⁵–32⁺³ weeks). Each had two-dimensional (2D) manually-acquired (2D-MA), 3D operator-selected (3D-OS) and 3D-DL automatically-acquired (3D-DL) images. The proportion of adequate images from each plane and modality, and the number of inadequate images per plane was compared for each method. Inter and intra-observer agreement of overall image quality was calculated.

Results

Sixty-seven percent of 3D-OS and 3D-DL transventricular planes were adequate quality. Forty-five percent of 3D-OS and 55% of 3D-DL transcerebellar planes were adequate.

Seventy-one percent of 3D-OS and 86% of 3D-DL transventricular planes failed with poor visualisation of intra-cranial structures. Eighty-six percent of 3D-OS and 80% of 3D-DL transcerebellar planes failed due to inadequate visualisation of cerebellar hemispheres. Image quality was significantly different between 2D and 3D, however, no significant difference between 3D-modalities was demonstrated (p < 0.005). Inter-observer agreement of transventricular plane adequacy was moderate for both 3D-modalities, and weak for transcerebellar planes.

Conclusion

The 3D-DL algorithm can automatically extract standard fetal head planes from 3D-head volumes of comparable quality to operator-selected planes. Image quality in 3D is inferior to corresponding 2D planes, likely due to limitations with 3D-technology and acquisition technique.

Implications for practice

Automated image extraction of standard planes from US-volumes could facilitate use of 3DUS in clinical practice, however image quality is dependent on the volume acquisition technique.

Role of regenerating islet-derived proteins in inflammatory bowel disease

Inflammatory bowel disease (IBD) is an inflammatory disorder of the gastrointestinal tract that affects millions of patients worldwide. It has a complex and multifactorial etiology leading to excessive exposure of intestinal epithelium to microbial antigens, inappropriate activation of the immune system and ultimately to the damage of intestinal tissues. Although numerous efforts have been made to improve the disease management, IBD remains persistently recurring and beyond cure. This is due largely to the gaps in our understanding of the pathogenesis of IBD that hamper the development of timely diagnoses and effective treatment. However, some recent discoveries, including the beneficial effects of interleukin-22 (IL-22) on the inflamed intestine, have shed light on a self-protective mechanism in IBD. Regenerating islet-derived (REG/Reg) proteins are small secretory proteins which function as IL-22's downstream effectors. Mounting studies have demonstrated that IBD patients have significantly increased REG expressions in the injured intestine, but with undefined mechanisms and roles. The reported functions of REG/Reg proteins in intestinal homeostasis, such as those of antibacterial, anti-inflammatory and tissue repair, lead us to discuss their potential mechanisms and clinical relevance in IBD in order to advance IBD research and management.

The formation of pattern on the wing of the moth, Ephestia kuhniella

Transverse bands of coloured scales are a common feature of lepidopteran wing patterns and it has been suggested that their positions are specified by a propagated ‘Determination Wave’ (Kuhn & von Englehardt), or a ‘Gradient’ of diffusible morphogen (Nijhout). We have assessed these models in an experimental study of the formation of bands in the moth, Ephestia kuhniella. The banding pattern can be altered by microcautery of the pupal wing between 1h and 48h postpupation (at 20 degrees C) and effects are of two types: Local modifications follow early (1–3h post-pupation) cautery located on or between the presumptive sites of the bands; operations more proximally or distally on the wing have no effect. Patterns consist of a loop deflecting the nearest band medially, around the site of cautery; or an isolated ring of band scales surrounding the operation site, with the two bands in their normal positions. The type of pattern formed depends only on the location of the cautery, with rings following medial cautery (midway between the bands). Global modifications follow a 36–48h cautery anywhere on the wing surface and they consist of a medial displacement of both bands. The degree of displacement is very variable (with band separation reduced to 33–87% of that on the contralateral control wing) but this does not depend on location of the cautery. Furthermore, the degree of modification does not depend on the precise age at cautery, as an equivalent range of effects is produced at 36h, at 42h and at 48h. The Determination Wave model is disproved since it explicitly assumes that the different degrees of global modification form a temporal sequence (from severe early to mild late) and they do not. Furthermore, this model cannot explain adequately the formation of loop or ring local modifications. The Gradient model can account for many features of the results, but it does not readily explain the ring local modifications or the effect of (36–48h) cautery in causing a global change in the banding pattern, independent of the site or precise time of operation.