Model Evaluation Based on the Distribution of Estimated Absolute Prediction Error

The construction of a reliable, practically useful prediction rule for future response is heavily dependent on the "adequacy" of the fitted regression model. In this article, we consider the absolute prediction error, the expected value of the absolute difference between the future and predicted responses, as the model evaluation criterion. This prediction error is easier to interpret than the average squared error and is equivalent to the mis-classification error for the binary outcome. We show that the distributions of the apparent error and its cross-validation counterparts are approximately normal even under a misspecified fitted model. When the prediction rule is "unsmooth", the variance of the above normal distribution can be estimated well via a perturbation-resampling method. We also show how to approximate the distribution of the difference of the estimated prediction errors from two competing models. With two real examples, we demonstrate that the resulting interval estimates for prediction errors provide much more information about model adequacy than the point estimates alone.

Spontaneous progression of ligature induced peri-implantitis at implants with different surface roughness: an experimental study in dogs

BACKGROUND: Peri-implantitis is associated with the presence of submarginal plaque, soft-tissue inflammation and advanced breakdown of the supporting bone. The progression of peri-implantitis following varying periods of continuing plaque accumulation has been studied in animal models. OBJECTIVE: The aim of the current experiment was to study the progression of peri-implantitis around implants with different surface roughness. MATERIAL AND METHODS: In five beagle dogs, three implants with either a sandblasted acid-etched surface (SLA) or a polished surface (P) were installed bilaterally in the edentulous premolar regions. After 3 months on a plaque control regimen, experimental peri-implantitis was induced by ligature placement and plaque accumulation was allowed to progress until about 40% of the height of the supporting bone had been lost. After this 4-month period, ligatures were removed and plaque accumulation was continued for an additional 5 months. Radiographs of all implant sites were obtained before and after 'active' experimental peri-implantitis as well as at the end of the experiment. Biopsies were harvested and the tissue samples were prepared for light microscopy. The sections were used for histometric and morphometric examinations. RESULTS: The radiographic examinations indicated that similar amounts of bone loss occurred at SLA and P sites during the active breakdown period, while the progression of bone loss was larger at SLA than at polished sites following ligature removal. The histological examination revealed that both bone loss and the size of the inflammatory lesion in the connective tissue were larger in SLA than in polished implant sites. The area of plaque was also larger at implants with an SLA surface than at implants with a polished surface. CONCLUSION: It is suggested that the progression of peri-implantitis, if left untreated, is more pronounced at implants with a moderately rough surface than at implants with a polished surface.

Calibration of a portal imaging device for high-precision dosimetry: a Monte Carlo study

Today electronic portal imaging devices (EPID's) are used primarily to verify patient positioning. They have, however, also the potential as 2D-dosimeters and could be used as such for transit dosimetry or dose reconstruction. It has been proven that such devices, especially liquid filled ionization chambers, have a stable dose response relationship which can be described in terms of the physical properties of the EPID and the pulsed linac radiation. For absolute dosimetry however, an accurate method of calibration to an absolute dose is needed. In this work, we concentrate on calibration against dose in a homogeneous water phantom. Using a Monte Carlo model of the detector we calculated dose spread kernels in units of absolute dose per incident energy fluence and compared them to calculated dose spread kernels in water at different depths. The energy of the incident pencil beams varied between 0.5 and 18 MeV. At the depth of dose maximum in water for a 6 MV beam (1.5 cm) and for a 18 MV beam (3.0 cm) we observed large absolute differences between water and detector dose above an incident energy of 4 MeV but only small relative differences in the most frequent energy range of the beam energy spectra. It is shown that for a 6 MV beam the absolute reference dose measured at 1.5 cm water depth differs from the absolute detector dose by 3.8%. At depth 1.2 cm in water, however, the relative dose differences are almost constant between 2 and 6 MeV. The effects of changes in the energy spectrum of the beam on the dose responses in water and in the detector are also investigated. We show that differences larger than 2% can occur for different beam qualities of the incident photon beam behind water slabs of different thicknesses. It is therefore concluded that for high-precision dosimetry such effects have to be taken into account. Nevertheless, the precise information about the dose response of the detector provided in this Monte Carlo study forms the basis of extracting directly the basic radiometric quantities photon fluence and photon energy fluence from the detector's signal using a deconvolution algorithm. The results are therefore promising for future application in absolute transit dosimetry and absolute dose reconstruction.

THE THREE DIMENSIONAL SHAPE AND ROUGHNESS OF MINERAL DUST

Mineral dust shape and roughness are important for a multitude of processes; it is known for aspherical shape but the true measurements in three dimensions are rare. Atomic Force Microscope was used for determine both 3D shape and roughness for two dust which are commonly used in laboratory experiments – Arizona Test Dust (ATD) and Kaolinite. We determined both of them are rather flat and round; an oblate spheroid would be a good model. Loess Filter was used to smooth the particles' surface and correlation analysis was used to examine the surfaces' properties of the dust; we found no features under 100nm scales. Also, our particles' surface area result is very similar to BET surface area.

Characterization of a Coal Fly Ash-Cement Slurry by the Absolute Foam Index

This dissertation established a standard foam index: the absolute foam index test. This test characterized a wide range of coal fly ash by the absolute volume of air-entraining admixture (AEA) necessary to produce a 15-second metastable foam in a coal fly ash-cement slurry in a specified time. The absolute foam index test was used to characterize fly ash samples having loss on ignition (LOI) values that ranged from 0.17 to 23.3 %wt. The absolute foam index characterized the fly ash samples by absolute volume of AEA, defined as the amount of undiluted AEA solution added to obtain a 15-minute endpoint signified by 15-second metastable foam. Results were compared from several foam index test time trials that used different initial test concentrations to reach termination at selected times. Based on the coefficient of variation (CV), a 15-minute endpoint, with limits of 12 to 18 minutes was chosen. Various initial test concentrations were used to accomplish consistent contact times and concentration gradients for the 15-minute test endpoint for the fly ash samples. A set of four standard concentrations for the absolute foam index test were defined by regression analyses and a procedure simplifying the test process. The set of standard concentrations for the absolute foam index test was determined by analyzing experimental results of 80 tests on coal fly ashes with loss on ignition (LOI) values ranging from 0.39 to 23.3 wt.%. A regression analysis informed selection of four concentrations (2, 6, 10, and 15 vol.% AEA) that are expected to accommodate fly ashes with 0.39 to 23.3 wt.% LOI, depending on the AEA type. Higher concentrations should be used for high-LOI fly ash when necessary. A procedure developed using these standard concentrations is expected to require only 1-3 trials to meet specified endpoint criteria for most fly ashes. The AEA solution concentration that achieved the metastable foam in the foam index test was compared to the AEA equilibrium concentration obtained from the direct adsorption isotherm test with the same fly ash. The results showed that the AEA concentration that satisfied the absolute foam index test was much less than the equilibrium concentration. This indicated that the absolute foam index test was not at or near equilibrium. Rather, it was a dynamic test where the time of the test played an important role in the results. Even though the absolute foam index was not an equilibrium condition, a correlation was made between the absolute foam index and adsorption isotherms. Equilibrium isotherm equations obtained from direct isotherm tests were used to calculate the equilibrium concentrations and capacities of fly ash from 0.17 to 10.5% LOI. The results showed that the calculated fly ash capacity was much less than capacities obtained from isotherm tests that were conducted with higher initial concentrations. This indicated that the absolute foam index was not equilibrium. Rather, the test is dynamic where the time of the test played an important role in the results. Even though the absolute foam index was not an equilibrium condition, a correlation was made between the absolute foam index and adsorption isotherms for fly ash of 0.17 to 10.5% LOI. Several batches of mortars were mixed for the same fly ash type increasing only the AEA concentration (dosage) in each subsequent batch. Mortar air test results for each batch showed for each increase in AEA concentration, air contents increased until a point where the next increase in AEA concentration resulted in no increase in air content. This was maximum air content that could be achieved by the particular mortar system; the system reached its air capacity at the saturation limit. This concentration of AEA was compared to the critical micelle concentration (CMC) for the AEA and the absolute foam index.

Remaining lifetime and absolute 10-year probabilities of osteoporotic fracture in Swiss men and women

SUMMARY: Remaining lifetime and absolute 10-year probabilities for osteoporotic fractures were determined by gender, age, and BMD values. Remaining lifetime probability at age 50 years was 20.2% in men and 51.3% in women and increased with advancing age and decreasing BMD. The study validates the elements required to populate a Swiss-specific FRAX model. INTRODUCTION: Switzerland belongs to high-risk countries for osteoporosis. Based on demographic projections, burden will still increase. We assessed remaining lifetime and absolute 10-year probabilities for osteoporotic fractures by gender, age and BMD in order to populate FRAX algorithm for Switzerland. METHODS: Osteoporotic fracture incidence was determined from national epidemiological data for hospitalised fractured patients from the Swiss Federal Office of Statistics in 2000 and results of a prospective Swiss cohort with almost 5,000 fractured patients in 2006. Validated BMD-associated fracture risk was used together with national death incidence and risk tables to determine remaining lifetime and absolute 10-year fracture probabilities for hip and major osteoporotic (hip, spine, distal radius, proximal humerus) fractures. RESULTS: Major osteoporotic fractures incidence was 773 and 2,078 per 100,000 men and women aged 50 and older. Corresponding remaining lifetime probabilities at age 50 were 20.2% and 51.3%. Hospitalisation for clinical spine, distal radius, and proximal humerus fractures reached 25%, 30% and 50%, respectively. Absolute 10-year probability of osteoporotic fracture increased with advancing age and decreasing BMD and was higher in women than in men. CONCLUSION: This study validates the elements required to populate a Swiss-specific FRAX model, a country at highest risk for osteoporotic fractures.