Maximum Likelihood Estimation of Mixture Densities for Binned and Truncated Multivariate Data

Binning and truncation of data are common in data analysis and machine learning. This paper addresses the problem of fitting mixture densities to multivariate binned and truncated data. The EM approach proposed by McLachlan and Jones (Biometrics, 44: 2, 571-578, 1988) for the univariate case is generalized to multivariate measurements. The multivariate solution requires the evaluation of multidimensional integrals over each bin at each iteration of the EM procedure. Naive implementation of the procedure can lead to computationally inefficient results. To reduce the computational cost a number of straightforward numerical techniques are proposed. Results on simulated data indicate that the proposed methods can achieve significant computational gains with no loss in the accuracy of the final parameter estimates. Furthermore, experimental results suggest that with a sufficient number of bins and data points it is possible to estimate the true underlying density almost as well as if the data were not binned. The paper concludes with a brief description of an application of this approach to diagnosis of iron deficiency anemia, in the context of binned and truncated bivariate measurements of volume and hemoglobin concentration from an individual's red blood cells.

Age-related lifecycles - Purpose variations

By describing and discussing Australian data on outbound tourism, this paper investigates the bimodal (double peaked) lifecycle pattern predicted and observed in the literature. These theories are primarily based on demand for holiday tourism, but tourists do not just go overseas for this purpose, with visiting friends and relatives, business, convention and conference, employment and educational tourism relevant purposes also. Trips overseas by Australians are split into these groups, enabling the age-related lifecycles of each to be examined separately. Unlike holiday tourism, all other purposes are unimodal, although they vary with respect to the age groups at which they peak and their relative positions.

Oral cancer in Australia: 1983-1996

Background: The aims of this study were to identify differences in oral cancer incidence and mortality between sexes, age groups, oral sites and Australian States and Territories and recent trends in oral cancer incidence, mortality and age-profile over time. Methods: Data were obtained from the Australian Institute for Health and Welfare and were age-standardized to the Australian 1991 Population Standard. Differences and trends were assessed with the Wilcoxon matched-pairs signed-ranks test and the Spearman correlation test, respectively. Results: In Australia in 1996, there were 2173 new oral cancers and 400 deaths due to oral cancer, the majority of oral cancers were in the 60+ age group, oral cancer affected men more than women (>2:1), lip cancer accounted for more than 50 per cent of oral cancers and the oral cancer mortality-to-incidence (M:I) ratio was greatest in ACT and NSW and least in QLD and SA. From 1983 to 1996, the annual incidence of lip cancer increased while the M:I ratio of lip cancer decreased. The annual incidence of cervical cancer decreased whereas the annual incidence of intra-oral cancer remained constant. The M:I ratio of cervical cancer was consistently lower than the MA ratio of intra-oral cancer. Conclusions; Reducing exposure to environmental carcinogens, increasing public awareness and population screening may reduce the incidence and mortality of oral cancer in Australia.

A comparison of Kodak Ultraspeed and Ektaspeed Plus dental X-ray films for the detection of dental caries

Background: Using the fastest dental X-ray film available is an easy way of reducing exposure to ionizing radiation. However, the diagnostic ability of fast films for the detection of proximal surface caries must be demonstrated before these films will become universally accepted. Methods: Extracted premolar and molar teeth were arranged to simulate a bitewing examination and radiographed using Ultraspeed and Ektaspeed Plus dental X-ray films. Three different exposure times were used for each film type. Six general dentists were used to determine the presence and depth of the decay in the proximal surfaces of the teeth radiographed. The actual extent of the decay in the teeth was determined by sectioning the teeth and examining them under a microscope. Results: There was no significant difference between the two films for the mean correct diagnosis. However, there was a significant difference between the means for the three exposure times used for Ultraspeed film. The practitioners used were not consistent in their ability to make a correct diagnosis, or for the film for which they got the highest correct diagnosis. Conclusions: Ektaspeed Plus dental X-ray film is just as reliable as Ultraspeed dental X-ray film for the detection of proximal surface decay. The effect of underexposure was significant for Ultraspeed, but not for Ektaspeed Plus. Patient exposure can be reduced significantly with no loss of diagnostic ability by changing from Ultraspeed X-ray film to Ektaspeed Plus X-ray film.

Chest pain in the dental surgery: A brief review and practical points in diagnosis and management

If a dental patient develops chest pain it must always be managed promptly and properly, i.e., the practitioner immediately stops the procedure and, being aware of the patients's medical history, questions the patient regarding the nature of the pain to help determine the likely diagnosis. It will most likely be a manifestation of coronary artery disease (synonymous with ischaemic heart disease), i.e., angina pectoris or acute myocardial infarction, most usually the former. Angina will usually resolve with proper intervention whereas up to about one-half of myocardial infarction cases will develop cardiac arrest, mostly in the first few hours, and this will be fatal in up to two-thirds of cases. As health care professions, dental practitioners have an inherent duty of care to be able to initiate appropriate care if such a medical emergency occurs.

Sites of dental erosion are saliva-dependent

Acid demineralization of teeth causes occlusal erosion and attrition and associated non-carious cervical lesions at sites relatively unprotected by saliva. Associations of occlusal pathology and cervical lesions were looked for in 450 patients with toothwear, and 174 subjects with cervical lesions were identified. Associations of occlusal attrition, or erosion, or no wear, with cervical lesions at 72 buccal and lingual sites were recorded from epoxy resin replicas of the subjects' dentitions (3241 teeth). Criteria used to discriminate occlusal erosion from attrition; and shallow from grooved and wedge-shaped cervical lesions were delineated by scanning electron microscopy (SEM). In the absence of occlusal pathology, cervical lesions were very rare (

Ethical dilemmas confronting dentists in Queensland, Australia

Background: This paper details contemporary ethical dilemmas encountered by Queensland dentists. Methods: An age-stratified sample of 499 dentists resident in Queensland was surveyed. The questionnaire contained scenarios of five common ethical dilemmas. In addition, open-ended questions sought the respondent's most frequent, difficult and recent ethical dilemmas, and where they would seek guidance in dealing with ethical problems. Results: Respondents acknowledged the patient's rights in treatment decisions and the dentist's right to refuse demands for inappropriate treatment. However, responses varied in the extent to which dentists may influence treatment decisions. Few respondents would ignore evidence of poor dental treatment but they are evenly divided in choosing to inform the patient, the dentist or both. Poor quality treatment is the most frequent and difficult dilemma, and half have experienced this problem recently. Requests by patients for fraudulent receipts occur in a third of responses. Dentists develop ethical values from multiple sources but for help with dental ethical problems, 90 per cent of respondents would consult another dentist. Conclusions: Of the ethical dilemmas discussed in this survey, those relating to poor quality treatment confronted most respondents. Also the actions of dentists in dealing with these dilemmas were most varied.