Improving the accuracy of fatty liver index to reflect liver fat content with predictive regression modelling.

BACKGROUND: The fatty liver index (FLI) is frequently used as a non-invasive clinical marker for research, prognostic and diagnostic purposes. It is also used to stratify individuals with hepatic steatosis such as non-alcoholic fatty liver disease (NAFLD), and to detect the presence of type 2 diabetes or cardiovascular disease. The FLI is calculated using a combination of anthropometric and blood biochemical variables; however, it reportedly excludes 8.5-16.7% of individuals with NAFLD. Moreover, the FLI cannot quantitatively predict liver fat, which might otherwise render an improved diagnosis and assessment of fatty liver, particularly in longitudinal studies. We propose FLI+ using predictive regression modelling, an improved index reflecting liver fat content that integrates 12 routinely-measured variables, including the original FLI. METHODS AND FINDINGS: We evaluated FLI+ on a dataset from the UK Biobank containing 28,796 individual estimates of proton density fat fraction derived from magnetic resonance imaging across normal to severe levels and interpolated to align with the original FLI range. The results obtained for FLI+ outperform the original FLI by delivering a lower mean absolute error by approximately 47%, a lower standard deviation by approximately 20%, and an increased adjusted R2 statistic by approximately 49%, reflecting a more accurate representation of liver fat content. CONCLUSIONS: Our proposed model predicting FLI+ has the potential to improve diagnosis and provide a more accurate stratification than FLI between absent, mild, moderate and severe levels of hepatic steatosis.

Improving Automatic Renal Segmentation in Clinically Normal and Abnormal Paediatric DCE-MRI via Contrast Maximisation and Convolutional Networks for Computing Markers of Kidney Function.

There is a growing demand for fast, accurate computation of clinical markers to improve renal function and anatomy assessment with a single study. However, conventional techniques have limitations leading to overestimations of kidney function or failure to provide sufficient spatial resolution to target the disease location. In contrast, the computer-aided analysis of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) could generate significant markers, including the glomerular filtration rate (GFR) and time-intensity curves of the cortex and medulla for determining obstruction in the urinary tract. This paper presents a dual-stage fully modular framework for automatic renal compartment segmentation in 4D DCE-MRI volumes. (1) Memory-efficient 3D deep learning is integrated to localise each kidney by harnessing residual convolutional neural networks for improved convergence; segmentation is performed by efficiently learning spatial-temporal information coupled with boundary-preserving fully convolutional dense nets. (2) Renal contextual information is enhanced via non-linear transformation to segment the cortex and medulla. The proposed framework is evaluated on a paediatric dataset containing 60 4D DCE-MRI volumes exhibiting varying conditions affecting kidney function. Our technique outperforms a state-of-the-art approach based on a GrabCut and support vector machine classifier in mean dice similarity (DSC) by 3.8% and demonstrates higher statistical stability with lower standard deviation by 12.4% and 15.7% for cortex and medulla segmentation, respectively.

3D Deep Learning for Anatomical Structure Segmentation in Multiple Imaging Modalities.

Accurate, quantitative segmentation of anatomical structures in radiological scans, such as Magnetic Resonance Imaging (MRI) and Computer Tomography (CT), can produce significant biomarkers and can be integrated into computer-aided assisted diagnosis (CADx) systems to support the interpretation of medical images from multi-protocol scanners. However, there are serious challenges towards developing robust automated segmentation techniques, including high variations in anatomical structure and size, the presence of edge-based artefacts, and heavy un-controlled breathing that can produce blurred motion-based artefacts. This paper presents a novel computing approach for automatic organ and muscle segmentation in medical images from multiple modalities by harnessing the advantages of deep learning techniques in a two-part process. (1) a 3D encoder-decoder, Rb-UNet, builds a localisation model and a 3D Tiramisu network generates a boundary-preserving segmentation model for each target structure; (2) the fully trained Rb-UNet predicts a 3D bounding box encapsulating the target structure of interest, after which the fully trained Tiramisu model performs segmentation to reveal detailed organ or muscle boundaries. The proposed approach is evaluated on six different datasets, including MRI, Dynamic Contrast Enhanced (DCE) MRI and CT scans targeting the pancreas, liver, kidneys and psoas-muscle and achieves quantitative measures of mean Dice similarity coefficient (DSC) that surpass or are comparable with the state-of-the-art. A qualitative evaluation performed by two independent radiologists verified the preservation of detailed organ and muscle boundaries.

A Framework for Automatic Morphological Feature Extraction and Analysis of Abdominal Organs in MRI Volumes.

The accurate 3D reconstruction of organs from radiological scans is an essential tool in computer-aided diagnosis (CADx) and plays a critical role in clinical, biomedical and forensic science research. The structure and shape of the organ, combined with morphological measurements such as volume and curvature, can provide significant guidance towards establishing progression or severity of a condition, and thus support improved diagnosis and therapy planning. Furthermore, the classification and stratification of organ abnormalities aim to explore and investigate organ deformations following injury, trauma and illness. This paper presents a framework for automatic morphological feature extraction in computer-aided 3D organ reconstructions following organ segmentation in 3D radiological scans. Two different magnetic resonance imaging (MRI) datasets are evaluated. Using the MRI scans of 85 adult volunteers, the overall mean volume for the pancreas organ is 69.30 ± 32.50cm3, and the 3D global curvature is (35.23 ± 6.83) × 10-3. Another experiment evaluates the MRI scans of 30 volunteers, and achieves mean liver volume of 1547.48 ± 204.19cm3 and 3D global curvature (19.87 ± 3.62) × 10- 3. Both experiments highlight a negative correlation between 3D curvature and volume with a statistical difference (p < 0.0001). Such a tool can support the investigation into organ related conditions such as obesity, type 2 diabetes mellitus and liver disease.

Morphological and multi-level geometrical descriptor analysis in CT and MRI volumes for automatic pancreas segmentation.

Automatic pancreas segmentation in 3D radiological scans is a critical, yet challenging task. As a prerequisite for computer-aided diagnosis (CADx) systems, accurate pancreas segmentation could generate both quantitative and qualitative information towards establishing the severity of a condition, and thus provide additional guidance for therapy planning. Since the pancreas is an organ of high inter-patient anatomical variability, previous segmentation approaches report lower quantitative accuracy scores in comparison to abdominal organs such as the liver or kidneys. This paper presents a novel approach for automatic pancreas segmentation in magnetic resonance imaging (MRI) and computer tomography (CT) scans. This method exploits 3D segmentation that, when coupled with geometrical and morphological characteristics of abdominal tissue, classifies distinct contours in tight pixel-range proximity as "pancreas" or "non-pancreas". There are three main stages to this approach: (1) identify a major pancreas region and apply contrast enhancement to differentiate between pancreatic and surrounding tissue; (2) perform 3D segmentation via continuous max-flow and min-cuts approach, structured forest edge detection, and a training dataset of annotated pancreata; (3) eliminate non-pancreatic contours from resultant segmentation via morphological operations on area, structure and connectivity between distinct contours. The proposed method is evaluated on a dataset containing 82 CT image volumes, achieving mean Dice Similarity coefficient (DSC) of 79.3 ±â€¯4.4%. Two MRI datasets containing 216 and 132 image volumes are evaluated, achieving mean DSC 79.6 ±â€¯5.7% and 81.6 ±â€¯5.1% respectively. This approach is statistically stable, reflected by lower metrics in standard deviation in comparison to state-of-the-art approaches.