Magnus Borga

Evidence for two types of brown adipose tissue in humans

The previously observed supraclavicular depot of brown adipose tissue (BAT) in adult humans was commonly believed to be the equivalent of the interscapular thermogenic organ of small mammals. This view was recently disputed on the basis of the demonstration that this depot consists of beige (also called brite) brown adipocytes, a newly identified type of brown adipocyte that is distinct from the classical brown adipocytes that make up the interscapular thermogenic organs of other mammals. A combination of high-resolution imaging techniques and histological and biochemical analyses showed evidence for an anatomically distinguishable interscapular BAT (iBAT) depot in human infants that consists of classical brown adipocytes, a cell type that has so far not been shown to exist in humans. On the basis of these findings, we conclude that infants, similarly to rodents, have the bona fide iBAT thermogenic organ consisting of classical brown adipocytes that is essential for the survival of small mammals in a cold environment.

Adaptive analysis of fMRI data

This article introduces novel and fundamental improvements of fMRI data analysis. Central is a technique termed constrained canonical correlation analysis, which can be viewed as a natural extension and generalization of the popular general linear model method. The concept of spatial basis filters is presented and shown to be a very successful way of adaptively filtering the fMRI data. A general method for designing suitable hemodynamic response models is also proposed and incorporated into the constrained canonical correlation approach. Results that demonstrate how each of these parts significantly improves the detection of brain activity, with a computation time well within limits for practical use, are provided.

Learning multidimensional signal processing

This paper presents our general strategy for designing learning machines as well as a number of particular designs. The search for methods allowing a sufficient level of adaptivity are based on two main principles: 1) simple adaptive local models; and 2) adaptive model distribution. Particularly important concepts in our work is mutual information and canonical correlation. Examples are given on learning feature descriptors, modeling disparity, synthesis of a global 3-mode model and a setup for reinforcement learning of online video coder parameter control.

Exploratory fMRI Analysis by Autocorrelation Maximization

A novel and computationally efficient method for exploratory analysis of functional MRI data is presented. The basic idea is to reveal underlying components in the fMRI data that have maximum autocorrelation. The tool for accomplishing this task is Canonical Correlation Analysis. The relation to Principal Component Analysis and Independent Component Analysis is discussed and the performance of the methods is compared using both simulated and real data.

Small baseline volume of left hippocampus is associated with subsequent conversion of MCI into dementia: The Göteborg MCI study

BACKGROUND Earlier studies have reported that hippocampal atrophy can to some extent predict which patients with mild cognitive impairment (MCI) will subsequently convert to dementia, and that converters have an enhanced rate of hippocampal volume loss. OBJECTIVE To further validate the hypothesis that hippocampal atrophy predicts conversion from MCI to dementia, to relate baseline hippocampal volume to different forms of dementia, and to investigate the role of hippocampal side differences and rate of volume loss over time. PATIENTS The subjects (N=68) include patients with MCI at baseline and progression to dementia at the two-year follow-up (N=21), stable MCI patients (N=21), and controls (N=26). Among the progressing patients, 13 were diagnosed as having AD. METHODS The Göteborg MCI study is a clinically based longitudinal study with biannual clinical assessments. Hippocampal volumetry was performed manually on the MRI investigations at baseline and at the two-year follow-up. RESULTS Hippocampal volumetry could predict conversion to dementia in both the AD and the non-AD subgroup of converters. Left hippocampal volume in particular discriminated between converting and stable MCI. Cut off points for individual discrimination were shown to be potentially useful. The converting MCI group had a significantly higher rate of hippocampal volume loss as compared to the stable MCI group. CONCLUSIONS In MCI patients, hippocampal volumetry at baseline gives prognostic information about possible development of AD and non-AD dementia. Contrary to earlier studies, we found that left hippocampal volume has the best predictive power. Reliable predictions appear to be possible in many individual cases.

Automatic and quantitative assessment of regional muscle volume by multi-atlas segmentation using whole-body water–fat MRI

To develop and demonstrate a rapid whole‐body magnetic resonance imaging (MRI) method for automatic quantification of total and regional skeletal muscle volume.

Detection and detrending in fMRI data analysis.

This article addresses the impact that colored noise, temporal filtering, and temporal detrending have on the fMRI analysis situation. Specifically, it is shown why the detection of event-related designs benefit more from pre-whitening than blocked designs in a colored noise structure. Both theoretical and empirical results are provided. Furthermore, a novel exploratory method for producing drift models that efficiently capture trends and drifts in the fMRI data is introduced. A comparison to currently employed detrending approaches is presented. It is shown that the novel exploratory model is able to remove a major part of the slowly varying drifts that are abundant in fMRI data. The value of such a model lies in its ability to remove drift components that otherwise would have contributed to a colored noise structure in the voxel time series.

Quantitative abdominal fat estimation using MRI

This paper introduces a new method for automatic quantification of subcutaneous, visceral and non-visceral internal fat from MR-images acquired using the two point Dixon technique in the abdominal region. The method includes (1) a three dimensional phase unwrapping to provide water and fat images, (2) an image intensity inhomogeneity correction, and (3) a morphon based registration and segmentation of the tissue. This is followed by an integration of the corrected fat images within the different fat compartments that avoids the partial volume effects associated with traditional fat segmentation methods. The method was tested on 18 subjects before and after a period of fast-food hyper-alimentation showing high stability and performance in all analysis steps.

Blood vessel segmentation using multi-scale quadrature filtering

The segmentation of blood vessels is a common problem in medical imaging and various applications are found in diagnostics, surgical planning, training and more. Among many different techniques, the use of multiple scales and line detectors is a popular approach. However, the typical line filters used are sensitive to intensity variations and do not target the detection of vessel walls explicitly. In this article, we combine both line and edge detection using quadrature filters across multiple scales. The filter result gives well defined vessels as linear structures, while distinct edges facilitate a robust segmentation. We apply the filter output to energy optimization techniques for segmentation and show promising results in 2D and 3D to illustrate the behavior of our method. The conference version of this article received the best paper award in the bioinformatics and biomedical applications track at ICPR 2008.

Detection of Neural Activity in fMRI Using Maximum Correlation Modeling

A technique for detecting neural activity in functional MRI data is introduced. It is based on a novel framework termed maximum correlation modeling. The method employs a spatial filtering approach that adapts to the local activity patterns, which results in an improved detection sensitivity combined with good specificity. A spatially varying hemodynamic response is simultaneously modelled by a sum of two gamma functions. Comparisons to traditional analysis methods are made using both synthetic and real data. The results indicate that the maximum correlation modeling approach is a strong alternative for analyzing fMRI data.