Random field models in image analysis*

Image models are useful in quantitatively specifying natural constraints and general assumptions about the physical world and the imaging process. This review paper explains how Gibbs and Markov random field models provide a unifying theme for many contemporary problems in image analysis. Random field models permit the introduction of spatial context into pixel labeling problems, such as segmentation and restoration. Random field models also describe textured images and lead to algorithms for generating textured images, classifying textures and segmenting textured images. In spite of some impressive model-based image restoration and texture segmentation results reported in the literature, a number of fundamental issues remain unexplored, such as the specification of MRF models, modeling noise processes, performance evaluation, parameter estimation, the phase transition phenomenon and the comparative analysis of alternative procedures. The literature of random field models is filled with great promise, but a better mathematical understanding of these issues is needed as well as efficient algorithms for applications. These issues need to be resolved before random field models will be widely accepted as general tools in the image-processing community.     

Modeling Two-state Disease Processes with Random Effects

Many chronic medical conditions are manifested by alternating sojourns in symptom-free and symptomatic states. In many cases, in addition to their relapsing and remitting nature, these conditions lead to worsening disease patterns over time and may exhibit seasonal trends. We develop a mixed-effect two-state model for such disease processes in which covariate effects are modeled multiplicatively on transition intensities. The transition intensities, in turn, are functions of three time scales: the semi-Markov scale involving the backward recurrence time for the cyclical component, the Markov scale for the time trend component, and a seasonal time scale. Multiplicative bivariate log-normal random effects are introduced to accommodate heterogeneity in disease activity between subjects and to admit a possible negative correlation between the transition intensities. Maximum likelihood estimation is carried out using Gauss-Hermite integration and a standard Newton-Raphson procedure. Tests of homogeneity are presented based on score statistics. An application of the methodology to data from a multi-center clinical trial of chronic bronchitis is provided for illustrative purposes.     

A Model of Migrants' Remittances with Human Capital Investment and Intrafimilial Transfers1

This study analyzes data on migrants' remittances using a two‐period theory of intergenerational transfers based on an informal, intrafamilial loan arrangement using “weak altruism,” a behavior between “strong altruism” and pure self‐interest. The model provides an integrated theory of migrants' remittances, human capital investment decisions, and intrafamilial transfers applicable to low‐income countries with no official pension schemes and imperfect capital markets. Propositions, derived from the theory, are tested, re‐analyzing original survey data on remittances of Pacific island migrants in Sydney. When weak altruism and strong altruism yield opposite predictions, the econometric results tend to confirm the former hypothesis and invalidate the latter.     

Bayesian utilitarianism and probability homogeneity

This paper investigates, within the axiomatic framework of Jeffrey-Bolker decision theory, two kinds of conditions and the relation between them: (1) The Utilitarian condition that social rankings of prospects be representable by an expected utility function that is a weighted sum of the expected utility functions representing individual rankings; and (2) Homogeneity conditions on the probabilities and preferences of individuals. In particular, we show that identity of individuals’ probabilities is necessary and sufficient for the Utilitarian condition to hold and that the homogeneity of individuals’ probabilities can be derived from a Pareto condition on the relation between individual and social rankings, provided that these rankings are separable in a particular sense.This paper has considerably benefited from comments by John Broome, Isaac Levi and an anonymous referee. Special thanks to Philippe Mongin who provided encouragement, help and careful criticism throughout the development of this paper.     

Conducting research with children: the limits of confidentiality and child protection protocols

This paper addresses the issue of child protection protocols within research conducted with children. Based partly on primary data this paper raises questions about the role of ethics committees in defining the limits of confidentiality in relation to child protection protocols within research, the perceptions of both children and parents about the meanings of confidentiality and notions of ‘harm’, and the role of the researcher in relation to legal and professional guidance about the protection of vulnerable people in society. We explore the concept of confidentiality within a research setting from a child protection perspective. In doing so we examine the legal and moral obligations of researchers to report child protection concerns, how children themselves perceive ‘harm’, and the need for much clearer guidance to researchers, as well as child participants and parents, about the limitations of confidentiality.     

The business portfolio approach—Where it falls down in practice

During the past decade the SBU (or portfolio) concept has enjoyed widespread popularity as a basis for corporate-wide strategic planning systems within large, diversified firms. This article discusses the implementation of such systems on the basis of 3 years of clinical and small sample research. A ‘basic model’ of implementation is used as a basis for comparison to progressive practice as observed in a sample of 13 firms. Furthermore, a new conceptual perspective for the portfolio concept is developed as a complement to the observed approach to implementation.     

The ethical aspects of gain sharing with physicians

Gain sharing arrangements involving physicians can be a model combination of ethical business practices and maximizing profits. Or, gain sharing can be as unethical as fee-splitting was at the turn of the century, and as corrupt as any conflict of interest. The devil is--or is not--in the details.     

Misbehaving physicians and professional ethics

Consider a true story about physician misbehavior and see how it was handled. Sometimes a gentle touch works well in situations of unexpected and unintended behavioral problems.