Assessing clinical utility of machine learning and artificial intelligence approaches to analyze speech recordings in multiple sclerosis: A pilot study.

BACKGROUND: An early diagnosis together with an accurate disease progression monitoring of multiple sclerosis is an important component of successful disease management. Prior studies have established that multiple sclerosis is correlated with speech discrepancies. Early research using objective acoustic measurements has discovered measurable dysarthria. METHOD: The objective was to determine the potential clinical utility of machine learning and deep learning/AI approaches for the aiding of diagnosis, biomarker extraction and progression monitoring of multiple sclerosis using speech recordings. A corpus of 65 MS-positive and 66 healthy individuals reading the same text aloud was used for targeted acoustic feature extraction utilizing automatic phoneme segmentation. A series of binary classification models was trained, tuned, and evaluated regarding their Accuracy and area-under-the-curve. RESULTS: The Random Forest model performed best, achieving an Accuracy of 0.82 on the validation dataset and an area-under-the-curve of 0.76 across 5 k-fold cycles on the training dataset. 5 out of 7 acoustic features were statistically significant. CONCLUSION: Machine learning and artificial intelligence in automatic analyses of voice recordings for aiding multiple sclerosis diagnosis and progression tracking seems promising. Further clinical validation of these methods and their mapping onto multiple sclerosis progression is needed, as well as a validating utility for English-speaking populations.

Study of the Automatic Detection of Parkison's Disease Based on Speaker Recognition Technologies and Allophonic Distillation.

The use of new tools to detect Parkinson's Disease (PD) from speech articulatory movements can have a considerable impact in the diagnosis of patients. In this study, a novel approach involving speaker recognition techniques with allophonic distillation is proposed and tested separately in four parkinsonian speech databases (205 patients and 186 controls in total). This new scheme provides values between 72% and 94% of accuracy in the automatic detection of PD, depending on the database, and improvements up to 9% respect to baseline techniques. Results not only point towards the importance of the segmentation of the speech for the differentiation of parkinsonian and control speakers but confirm previous findings about the relevance of plosives and fricatives in the detection of parkinsonian dysarthria.

Enhanced spin-orbit coupling in tetragonally strained Fe-Co-B films.

Tetragonally strained interstitial Fe-Co-B alloys were synthesized as epitaxial films grown on a 20 nm thick Au0.55Cu0.45 buffer layer. Different ratios of the perpendicular to in-plane lattice constant c/a = 1.013, 1.034 and 1.02 were stabilized by adding interstitial boron with different concentrations 0, 4, and 10 at.%, respectively. Using ferromagnetic resonance (FMR) and x-ray magnetic circular dichroism (XMCD) we found that the total orbital magnetic moment significantly increases with increasing c/a ratio, indicating that reduced crystal symmetry and interstitial B leads to a noticeable enhancement of the effect of spin-orbit coupling (SOC) in the Fe-Co-B alloys. First-principles calculations reveal that the increase in orbital magnetic moment mainly originates from B impurities in octahedral position and the reduced symmetry around B atoms. These findings offer the possibility to enhance SOC phenomena-namely the magnetocrystalline anisotropy and the orbital moment-by stabilizing anisotropic strain by doping 4 at.% B. Results on the influence of B doping on the Fe-Co film microstructure, their coercive field and magnetic relaxation are also presented.

GABA spectra and remote distractor effect in progressive supranuclear palsy: A pilot study.

Disturbances of the gamma-aminobutyric-acid (GABA) system have been suspected of contributing to the pathophysiology of progressive supranuclear palsy (PSP). The ability to rapidly resolve competitive action decisions, such as shifting the gaze to one particular stimulus rather than another, can be predicted by the concentration of GABA in the region of the frontal cortex relevant to eye movements. For this reason, our study measured GABA levels in seven PSP patients and eight healthy controls, using proton magnetic resonance spectroscopy, and assessed the relationship of these measurements to the remote distractor effect (RDE), an eye-movement paradigm investigating competitive action decisions. No significant differences were found in either frontal-eye-field GABA levels or RDE between PSP patients and controls.

Automatic detection of Parkinson's disease in running speech spoken in three different languages.

The aim of this study is the analysis of continuous speech signals of people with Parkinson's disease (PD) considering recordings in different languages (Spanish, German, and Czech). A method for the characterization of the speech signals, based on the automatic segmentation of utterances into voiced and unvoiced frames, is addressed here. The energy content of the unvoiced sounds is modeled using 12 Mel-frequency cepstral coefficients and 25 bands scaled according to the Bark scale. Four speech tasks comprising isolated words, rapid repetition of the syllables /pa/-/ta/-/ka/, sentences, and read texts are evaluated. The method proves to be more accurate than classical approaches in the automatic classification of speech of people with PD and healthy controls. The accuracies range from 85% to 99% depending on the language and the speech task. Cross-language experiments are also performed confirming the robustness and generalization capability of the method, with accuracies ranging from 60% to 99%. This work comprises a step forward for the development of computer aided tools for the automatic assessment of dysarthric speech signals in multiple languages.

Influence of dimensionality and interface type on optical and electronic properties of CdS/ZnS core-shell nanocrystals--A first-principles study.

Semiconducting nanocrystals (NCs) have become one of the leading materials in a variety of applications, mainly due to their size tunable band gap and high intensity emission. Their photoluminescence (PL) properties can be notably improved by capping the nanocrystals with a shell of another semiconductor, making core-shell structures. We focus our study on the CdS/ZnS core-shell nanocrystals that are closely related to extensively studied CdSe/CdS NCs, albeit exhibiting rather different photoluminescence properties. We employ density functional theory to investigate the changes in the electronic and optical properties of these nanocrystals with size, core/shell ratio, and interface structure between the core and the shell. We have found that both the lowest unoccupied eigenstate (LUES) and the highest occupied eigenstate (HOES) wavefunction (WF) are localized in the core of the NCs, with the distribution of the LUES WF being more sensitive to the size and the core/shell ratio. We show that the radiative lifetimes are increasing, and the Coulomb interaction energies decrease with increasing NC size. Furthermore, we investigated the electronic and optical properties of the NCs with different interfaces between the core and the shell and different core types. We find that the different interfaces and core types have rather small influence on the band gaps and the absorption indexes, as well as on the confinement of the HOES and LUES WFs. Also the radiative lifetimes are found to be only slightly influenced by the different structural models. In addition, we compare these results with the previous results for CdSe/CdS NCs, reflecting the different PL properties of these two types of NCs. We argue that the difference in their Coulomb interaction energies is one of the main reasons for their distinct PL properties.

From soft to hard magnetic Fe-Co-B by spontaneous strain: a combined first principles and thin film study.

In order to convert the well-known Fe-Co-B alloy from a soft to a hard magnet, we propose tetragonal strain by interstitial boron. Density functional theory reveals that when B atoms occupy octahedral interstitial sites, the bcc Fe-Co lattice is strained spontaneously. Such highly distorted Fe-Co is predicted to reach a strong magnetocrystalline anisotropy which may compete with shape anisotropy. To probe this theoretical suggestion experimentally, epitaxial films are examined. A spontaneous strain up to 5% lattice distortion is obtained for B content up to 4 at%, which leads to uniaxial anisotropy constants exceeding 0.5 MJ m(-3). However, a further addition of B results in a partial amorphisation, which degrades both anisotropy and magnetisation.

First-principles study of the influence of different interfaces and core types on the properties of CdSe/CdS core-shell nanocrystals.

With the expanding field of nanoengineering and the production of nanocrystals (NCs) with higher quality and tunable size, having reliable theoretical calculations to complement the experimental results is very important. Here we present such a study of CdSe/CdS core-shell NCs using density functional theory, where we focus on dependence of the properties of these NCs on core types and interfaces between the core and the shell, as well as on the core/shell ratio. We show that the density of states and the absorption indices depend rather weakly on the type of interface and core type. We demonstrate that the HOMO wavefunction is mainly localised in the core of the nanocrystal, depending primarily on the core/shell ratio. On the other hand the LUMO wavefunction spreads more into the shell of the nanocrystal, where its confinement in the core is almost the same in each of the studied structural models. Furthermore, we show that the radiative lifetimes decrease with increasing core sizes due to changes in the dipolar overlap integral of the HOMO and LUMO wavefunctions. In addition, the electron-hole Coulomb interaction energies follow a similar pattern as the localisation of the wavefunctions, with the smaller NCs having higher Coulomb interaction energies.

Magnetic polarization of the americium J=0 ground state in AmFe(2).

Trivalent americium has a nonmagnetic (J=0) ground state arising from the cancellation of the orbital and spin moments. However, magnetism can be induced by a large molecular field if Am^{3+} is embedded in a ferromagnetic matrix. Using the technique of x-ray magnetic circular dichroism, we show that this is the case in AmFe_{2}. Since ⟨J_{z}⟩=0, the spin component is exactly twice as large as the orbital one, the total Am moment is opposite to that of Fe, and the magnetic dipole operator ⟨T_{z}⟩ can be determined directly; we discuss the progression of the latter across the actinide series.

Experimental and theoretical study of electronic structure of lutetium bi-phthalocyanine.

Using Near Edge X-Ray Absorption Fine Structure (NEXAFS) Spectroscopy, the thickness dependent formation of Lutetium Phthalocyanine (LuPc2) films on a stepped passivated Si(100)2×1 reconstructed surface was studied. Density functional theory (DFT) calculations were employed to gain detailed insights into the electronic structure. Photoelectron spectroscopy measurements have not revealed any noticeable interaction of LuPc2 with the H-passivated Si surface. The presented study can be considered to give a comprehensive description of the LuPc2 molecular electronic structure. The DFT calculations reveal the interaction of the two molecular rings with each other and with the metallic center forming new kinds of orbitals in between the phthalocyanine rings, which allows to better understand the experimentally obtained NEXAFS results.

Parameter-free extraction of EMCD from an energy-filtered diffraction datacube using multivariate curve resolution.

We present a parameter-free method of extraction of the electron magnetic circular dichroism spectra from energy-filtered diffraction patterns measured on a crystalline specimen. The method is based on a multivariate curve resolution technique. The main advantage of the proposed method is that it allows extraction of the magnetic signal regardless of the symmetry and orientation of the crystal, as long as there is a sufficiently strong magnetic component of the signal in the diffraction plane. This method essentially overcomes difficulties in extraction of the EMCD signal caused by complexity of dynamical diffraction effects.

Formation and structure of graphene waves on Fe(110).

A very rich Fe-C phase diagram makes the formation of graphene on iron surfaces a challenging task. Here we demonstrate that the growth of graphene on epitaxial iron films can be realized by chemical vapor deposition at relatively low temperatures, and that the formation of carbides can be avoided in excess of the carbon-containing precursors. The resulting graphene monolayer creates a novel periodically corrugated pattern on Fe(110). Using low-energy electron microscopy and scanning tunneling microscopy, we show that it is modulated in one dimension forming long waves with a period of ∼4 nm parallel to the [001] direction of the substrate, with an additional height modulation along the wave crests. The observed topography of the graphene/Fe superstructure is well reproduced by density functional theory calculations, and found to result from a unique combination of the lattice mismatch and strong interfacial interaction, as probed by core-level photoemission and x-ray absorption spectroscopy.

Quantitative acoustic measurements for characterization of speech and voice disorders in early untreated Parkinson's disease.

An assessment of vocal impairment is presented for separating healthy people from persons with early untreated Parkinson's disease (PD). This study's main purpose was to (a) determine whether voice and speech disorder are present from early stages of PD before starting dopaminergic pharmacotherapy, (b) ascertain the specific characteristics of the PD-related vocal impairment, (c) identify PD-related acoustic signatures for the major part of traditional clinically used measurement methods with respect to their automatic assessment, and (d) design new automatic measurement methods of articulation. The varied speech data were collected from 46 Czech native speakers, 23 with PD. Subsequently, 19 representative measurements were pre-selected, and Wald sequential analysis was then applied to assess the efficiency of each measure and the extent of vocal impairment of each subject. It was found that measurement of the fundamental frequency variations applied to two selected tasks was the best method for separating healthy from PD subjects. On the basis of objective acoustic measures, statistical decision-making theory, and validation from practicing speech therapists, it has been demonstrated that 78% of early untreated PD subjects indicate some form of vocal impairment. The speech defects thus uncovered differ individually in various characteristics including phonation, articulation, and prosody.

Restoration of Guyton´s diagram for regulation of the circulation as a basis for quantitative physiological model development.

We present the current state of complex circulatory dynamics model development based on Guyton's famous diagram. The aim is to provide an open-source model that will allow the simulation of a number of pathological conditions on a virtual patient including cardiac, respiratory, and kidney failure. The model will also simulate the therapeutic influence of various drugs, infusions of electrolytes, blood transfusion, etc. As a current result of implementation, we describe a core model of human physiology targeting the systemic circulation, arterial pressure and body fluid regulation, including short- and long-term regulations. The model can be used for educational purposes and general reflection on physiological regulation in pathogenesis of various diseases.

Asymmetry of the two-beam geometry in EMCD experiments.

We analyse theoretically the influence of the asymmetry of the two-beam geometry on quantitative measurements of the energy-loss magnetic chiral dichroism. Our simulations indicate that this asymmetry is not very strong inside or close to the Thales circle, but in other regions of the diffraction plane it can hinder an accurate extraction of the orbital to spin moment ratio.

Quadratic X-ray magneto-optical effect upon reflection in a near-normal-incidence configuration at the M edges of 3d-transition metals.

We have observed a quadratic x-ray magneto-optical effect in near-normal-incidence reflection at the M edges of iron. The effect appears as the magnetically induced rotation of approximately 0.1 degrees of the polarization plane of linearly polarized x-ray radiation upon reflection. A comparison of the measured rotation spectrum with results from x-ray magnetic linear dichroism data demonstrates that this is the first observation of the Schäfer-Hubert effect in the x-ray regime. Ab initio density-functional theory calculations reveal that hybridization effects of the 3p core states necessarily need to be considered when interpreting experimental data. The discovered magneto-x-ray effect holds promise for future ultrafast and element-selective studies of ferromagnetic as well as antiferromagnetic materials.

Hidden order in URu2Si2 originates from Fermi surface gapping induced by dynamic symmetry breaking.

Spontaneous, collective ordering of electronic degrees of freedom leads to second-order phase transitions that are characterized by an order parameter driving the transition. The notion of a 'hidden order' has recently been used for a variety of materials where a clear phase transition occurs without a known order parameter. The prototype example is the heavy-fermion compound URu(2)Si(2), where a mysterious hidden-order transition occurs at 17.5 K. For more than twenty years this system has been studied theoretically and experimentally without a firm grasp of the underlying physics. Here, we provide a microscopic explanation of the hidden order using density-functional theory calculations. We identify the Fermi surface 'hot spots' where degeneracy induces a Fermi surface instability and quantify how symmetry breaking lifts the degeneracy, causing a surprisingly large Fermi surface gapping. As the mechanism for the hidden order, we deduce spontaneous symmetry breaking through a dynamic mode of antiferromagnetic moment excitations.

Optimal aperture sizes and positions for EMCD experiments.

The signal-to-noise ratio (SNR) in energy-loss magnetic chiral dichroism (EMCD)--the equivalent of X-ray magnetic circular dichroism (XMCD) in the electron microscope--is optimized with respect to the detector shape, size and position. We show that an important increase in SNR over previous experiments can be obtained when taking much larger detector sizes. We determine the ideal shape of the detector but also show that round apertures are a good compromise if placed in their optimal position. We develop the theory for a simple analytical description of the EMCD experiment and then apply it to dynamical multibeam Bloch wave calculations and to an experimental data set. In all cases it is shown that a significant and welcome improvement of the SNR is possible.

Magnetic circular dichroism in electron energy loss spectrometry.

The measurement of circular dichroism in the electron microscope is a new, emerging method and, as such, it is subject to constant refinement and improvement. Different ways can be envisaged to record the signal. We present an overview of the key steps in the energy-loss magnetic chiral dichroism (EMCD) experiment as well as a detailed review of the methods used in the intrinsic way where the specimen is used as a beam splitter. Lateral resolution up to 20-30 nm can be achieved, and the use of convergent beam techniques leads to an improved S/N ratio. Dichroic effects are shown for Ni and Co single crystal; as a counterexample, measurements were carried also for a non-magnetic (Ti) sample, where no dichroic effect was found.