Normalized sensitivity of multi-dimensional body composition biomarkers for risk change prediction.

The limitations of BMI as a measure of adiposity and health risks have prompted the introduction of many alternative biomarkers. However, ranking diverse biomarkers from best to worse remains challenging. This study aimed to address this issue by introducing three new approaches: (1) a calculus-derived, normalized sensitivity score (NORSE) is used to compare the predictive power of diverse adiposity biomarkers; (2) multiple biomarkers are combined into multi-dimensional models, for increased sensitivity and risk discrimination; and (3) new visualizations are introduced that convey complex statistical trends in a compact and intuitive manner. Our approach was evaluated on 23 popular biomarkers and 6 common medical conditions using a large database (National Health and Nutrition Survey, NHANES, N ~ 100,000). Our analysis established novel findings: (1) regional composition biomarkers were more predictive of risk than global ones; (2) fat-derived biomarkers had stronger predictive power than weight-related ones; (3) waist and hip are always elements of the strongest risk predictors; (4) our new, multi-dimensional biomarker models yield higher sensitivity, personalization, and separation of the negative effects of fat from the positive effects of lean mass. Our approach provides a new way to evaluate adiposity biomarkers, brings forth new important clinical insights and sets a path for future biomarker research.

Machine learning for medical images analysis.

This article discusses the application of machine learning for the analysis of medical images. Specifically: (i) We show how a special type of learning models can be thought of as automatically optimized, hierarchically-structured, rule-based algorithms, and (ii) We discuss how the issue of collecting large labelled datasets applies to both conventional algorithms as well as machine learning techniques. The size of the training database is a function of model complexity rather than a characteristic of machine learning methods.

Encoding atlases by randomized classification forests for efficient multi-atlas label propagation.

We propose a method for multi-atlas label propagation (MALP) based on encoding the individual atlases by randomized classification forests. Most current approaches perform a non-linear registration between all atlases and the target image, followed by a sophisticated fusion scheme. While these approaches can achieve high accuracy, in general they do so at high computational cost. This might negatively affect the scalability to large databases and experimentation. To tackle this issue, we propose to use a small and deep classification forest to encode each atlas individually in reference to an aligned probabilistic atlas, resulting in an Atlas Forest (AF). Our classifier-based encoding differs from current MALP approaches, which represent each point in the atlas either directly as a single image/label value pair, or by a set of corresponding patches. At test time, each AF produces one probabilistic label estimate, and their fusion is done by averaging. Our scheme performs only one registration per target image, achieves good results with a simple fusion scheme, and allows for efficient experimentation. In contrast to standard forest schemes, in which each tree would be trained on all atlases, our approach retains the advantages of the standard MALP framework. The target-specific selection of atlases remains possible, and incorporation of new scans is straightforward without retraining. The evaluation on four different databases shows accuracy within the range of the state of the art at a significantly lower running time.

Regression forests for efficient anatomy detection and localization in computed tomography scans.

This paper proposes a new algorithm for the efficient, automatic detection and localization of multiple anatomical structures within three-dimensional computed tomography (CT) scans. Applications include selective retrieval of patients images from PACS systems, semantic visual navigation and tracking radiation dose over time. The main contribution of this work is a new, continuous parametrization of the anatomy localization problem, which allows it to be addressed effectively by multi-class random regression forests. Regression forests are similar to the more popular classification forests, but trained to predict continuous, multi-variate outputs, where the training focuses on maximizing the confidence of output predictions. A single pass of our probabilistic algorithm enables the direct mapping from voxels to organ location and size. Quantitative validation is performed on a database of 400 highly variable CT scans. We show that the proposed method is more accurate and robust than techniques based on efficient multi-atlas registration and template-based nearest-neighbor detection. Due to the simplicity of the regressor's context-rich visual features and the algorithm's parallelism, these results are achieved in typical run-times of only ∼4 s on a conventional single-core machine.

Evaluation of the apparent losses caused by water meter under-registration in intermittent water supply.

Apparent losses are usually caused by water theft, billing errors, or revenue meter under-registration. While the first two causes are directly related to water utility management and may be reduced by improving company procedures, water meter inaccuracies are considered to be the most significant and hardest to quantify. Water meter errors are amplified in networks subjected to water scarcity, where users adopt private storage tanks to cope with the intermittent water supply. The aim of this paper is to analyse the role of two variables influencing the apparent losses: water meter age and the private storage tank effect on meter performance. The study was carried out in Palermo (Italy). The impact of water meter ageing was evaluated in laboratory by testing 180 revenue meters, ranging from 0 to 45 years in age. The effects of the private water tanks were determined via field monitoring of real users and a mathematical model. This study demonstrates that the impact on apparent losses from the meter starting flow rapidly increases with meter age. Private water tanks, usually fed by a float valve, overstate meter under-registration, producing additional apparent losses between 15% and 40% for the users analysed in this study.