Algorithms for Hidden Markov Models With Imprecisely Specified Parameters

Hidden Markov models (HMMs) are widely used models for sequential data. As with other probabilistic graphical models, they require the specification of precise probability values, which can be too restrictive for some domains, especially when data are scarce or costly to acquire. We present a generalized version of HMMs, whose quantification can be done by sets of, instead of single, probability distributions. Our models have the ability to suspend judgment when there is not enough statistical evidence, and can serve as a sensitivity analysis tool for standard non-stationary HMMs. Efficient inference algorithms are developed to address standard HMM usage such as the computation of likelihoods and most probable explanations. Experiments with real data show that the use of imprecise probabilities leads to more reliable inferences without compromising efficiency.

High temperature thermal stability studies of ultrathin Al2O3 layers deposited on native oxide and sulphur passivated InGaAs surfaces

High resolution synchrotron radiation core level photoemission measurements have been used to undertake a comparative study of the high temperature stability of ultrathin Al2O3 layers deposited by atomic layer deposition (ALD) on both sulphur passivated and native oxide covered InGaAs. The residual interfacial oxides between sulphur passivated InGaAs and the ultrathin Al2O3 layer can be substantially removed at high temperature (up to 700 °C) without impacting on the InGaAs stoichiometry while significant loss of indium was recorded at this temperature on the native oxide InGaAs surface.

Plasmon enhanced fluorescence studies from aligned gold nanorod arrays modified with SiO2 spacer layers

Here, we demonstrate that quasi self-standing Au nanorod arrays prepared with plasma polymerisation deposited SiO2 dielectric spacers support surface enhanced fluorescence (SEF) while maintaining high signal reproducibility. We show that it is possible to find a balance between enhanced radiative and non-radiative decay rates at which the fluorescent intensity is maximized. The SEF signal optimised with a 30 nm spacer layer thickness showed a 3.5-fold enhancement with a signal variance of <15% thereby keeping the integrity of the nanorod array. We also demonstrate the decreased importance of obtaining resonance conditions when localized surface plasmon resonance is positioned within the spectral region of Au interband transitions. Procedures for further increasing the SEF enhancement factor are also discussed.

Nonreciprocal nonlinear scattering by stacked magnetoactive semiconductor layers

The combinatorial frequency generation by the periodic stacks of magnetically biased semiconductor layers has been modelled in the self-consistent problem formulation, taking into account the nonlinear dynamics of carriers. It has been shown that the nonlinear response of the magnetoactive semiconductor periodic structure is strongly enhanced by magnetic bias and combinations of the layer physical and geometrical parameters. The effects of the pump wave nonreciprocal reflectance and field displacement on the efficiency of three-wave mixing process is illustrated by the simulation results

Quasi-periodic stacks of semiconductor layers: Combinatorial frequency generation

The nonlinear scattering and combinatorial frequency generation by the quasi-periodic Fibonacci and Thue-Morse stacks of semiconductor layers have been investigated taking into account the nonlinear charge dynamics. It has been shown that the mixing processes in passive semiconductor structures are driven by the competitive effects of the collision of charges and resonance interactions of carriers with pump waves. The effects of the stack arrangements and constituent layer parameters on the efficiency of the combinatorial frequency generation are discussed. 

Diabetic retinopathy:morphometric analysis of basement membrane thickening of capillaries in different retinal layers within arterial and venous environments

AIMS: To assess quantitatively variations in the extent of capillary basement membrane (BM) thickening between different retinal layers and within arterial and venous environments during diabetes.

METHODS: One year after induction of experimental (streptozotocin) diabetes in rats, six diabetic animals together with six age-matched control animals were sacrificed and the retinas fixed for transmission electron microscopy (TEM). Blocks of retina straddling the major arteries and veins in the central retinal were dissected out, embedded in resin, and sectioned. Capillaries in close proximity to arteries or veins were designated as residing in either an arterial (AE) or a venous (VE) environment respectively, and the retinal layer in which each capillary was located was also noted. The thickness of the BM was then measured on an image analyser based two dimensional morphometric analysis system.

RESULTS: In both diabetics and controls the AE capillaries had consistently thicker BMs than the VE capillaries. The BMs of both AE and VE capillaries from diabetics were thicker than those of capillaries in the corresponding retinal layer from the normal rats (p < or = 0.005). Also, in normal AE and VE capillaries and diabetic AE capillaries the BM in the nerve fibre layer (NFL) was thicker than that in either the inner (IPL) or outer (OPL) plexiform layers (p < or = 0.001). However, in diabetic VE capillaries the BMs of capillaries in the NFL were thicker than those of capillaries in the IPL (p < or = 0.05) which, in turn, had thicker BMs than capillaries in the OPL (p < or = 0.005).

CONCLUSIONS: The variation in the extent of capillary BM thickening between different retinal layers within AE and VE environments may be related to differences in levels of oxygen tension and oxidative stress in the retina around arteries compared with that around veins.

Deciphering the performance hints hidden in J. S. Bach’s quaver beams

Bach often beamed his quavers beyond the beat-unit indicated by the time-signature, as if to indicate the way he perceived, phrased or articulated a musical line. In my previous study in 2011 I demonstrated the significance for both performers and editors of critical editions of quaver beaming against a broader background of Bach’s notational practice. In this article, I go one step further and demonstrate that Bach’s quaver beams reflect how Bach responded to his pieces in composition and performance, which sheds light on how he engaged with and perceived his music from motivic to structural levels. This enquiry has never before been pursued thoroughly and it promises to uncover the ideas that guided his notational practice as well as his spontaneous responses to the challenges he faced while writing out his music on paper

Using the hidden Markov model to capture quality of care in Lombardy geriatric wards

Quality of care is an important aspect of healthcare monitoring, which is used to ensure that the healthcare system is delivering care of the highest standard. With populations growing older there is an increased urgency in making sure that the healthcare delivered is of the highest standard. Healthcare providers are under increased pressure to ensure that this is the case with public and government demand expecting a healthcare system of the highest quality. Modelling quality of care is difficult to measure due to the many ways of defining it. This paper introduces a potential model which could be used to take quality of care into account when modelling length of stay. The Coxian phase-type distribution is used to model length of stay and the associated quality of care incorporated into the Coxian using a Hidden Markov model. Covariates are also introduced to determine their impact on the hidden level to find out what potentially can affect quality of care. This model is applied to geriatic patient data from the Lombardy region of Italy. The results obtained highlighted that bed numbers and the type of hospital (public or private) can have an effect on the quality of care delivered.

Hidden Markov Models With Set-Valued Parameters

Hidden Markov models (HMMs) are widely used probabilistic models of sequential data. As with other probabilistic models, they require the specification of local conditional probability distributions, whose assessment can be too difficult and error-prone, especially when data are scarce or costly to acquire. The imprecise HMM (iHMM) generalizes HMMs by allowing the quantification to be done by sets of, instead of single, probability distributions. iHMMs have the ability to suspend judgment when there is not enough statistical evidence, and can serve as a sensitivity analysis tool for standard non-stationary HMMs. In this paper, we consider iHMMs under the strong independence interpretation, for which we develop efficient inference algorithms to address standard HMM usage such as the computation of likelihoods and most probable explanations, as well as performing filtering and predictive inference. Experiments with real data show that iHMMs produce more reliable inferences without compromising the computational efficiency.

Fermi level de-pinning of aluminium contacts to n-type germanium using thin atomic layer deposited layers

Fermi-level pinning of aluminium on n-type germanium (n-Ge) was reduced by insertion of a thin interfacial dielectric by atomic layer deposition. The barrier height for aluminium contacts on n-Ge was reduced from 0.7 eV to a value of 0.28 eV for a thin Al2O3 interfacial layer (∼2.8 nm). For diodes with an Al2O3 interfacial layer, the contact resistance started to increase for layer thicknesses above 2.8 nm. For diodes with a HfO2 interfacial layer, the barrier height was also reduced but the contact resistance increased dramatically for layer thicknesses above 1.5 nm.

Using a Discrete Hidden Markov Model Kernel for lip-based biometric identification

In this paper, a novel and effective lip-based biometric identification approach with the Discrete Hidden Markov Model Kernel (DHMMK) is developed. Lips are described by shape features (both geometrical and sequential) on two different grid layouts: rectangular and polar. These features are then specifically modeled by a DHMMK, and learnt by a support vector machine classifier. Our experiments are carried out in a ten-fold cross validation fashion on three different datasets, GPDS-ULPGC Face Dataset, PIE Face Dataset and RaFD Face Dataset. Results show that our approach has achieved an average classification accuracy of 99.8%, 97.13%, and 98.10%, using only two training images per class, on these three datasets, respectively. Our comparative studies further show that the DHMMK achieved a 53% improvement against the baseline HMM approach. The comparative ROC curves also confirm the efficacy of the proposed lip contour based biometrics learned by DHMMK. We also show that the performance of linear and RBF SVM is comparable under the frame work of DHMMK.

Diabetes Distress: Understanding the Hidden Struggles of Living with Diabetes and Exploring Intervention Strategies

Diabetes Distress is a rational emotional response to the threat of a life-changing illness. Distinct from depression, it is rooted in the demands of diabetes management and is a product of psychological adjustment. Diabetes distress has been found to be significantly associated with HbA1c and self-care, which demonstrates its clinical use in treatment outcomes. Interpersonal factors such as perceived support and protectiveness of partners significantly contribute to elevated distress, suggesting that these are valued areas of focus for interventions. Pioneering large-scale research, DAWN2, gives voices to the families of those with diabetes and reaffirms the need to consider psychosocial factors in routine diabetes care. Structured diabetes education programmes are the most widely used in helping individuals cope with diabetes, but they fail to consider the psychological or interpersonal aspects of diabetes management. Psycho-educational approaches are found to be effective in reducing diabetes distress while also improving HbA1c. Certain limitations in the current literature are discussed, along with future directions. Of utmost importance is the need for health practitioners, irrespective of background, to demonstrate an understanding of diabetes distress and actively engage in discussion with individuals struggling to cope with diabetes; to normalize this and integrate it into routine diabetes practice.