Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

The design and development of wearable biosensor systems for health monitoring has garnered lots of attention in the scientific community and the industry during the last years. Mainly motivated by increasing healthcare costs and propelled by recent technological advances in miniature biosensing devices, smart textiles, microelectronics, and wireless communications, the continuous advance of wearable sensor-based systems will potentially transform the future of healthcare by enabling proactive personal health management and ubiquitous monitoring of a patient's health condition. These systems can comprise various types of small physiological sensors, transmission modules and processing capabilities, and can thus facilitate low-cost wearable unobtrusive solutions for continuous all-day and any-place health, mental and activity status monitoring. This paper attempts to comprehensively review the current research and development on wearable biosensor systems for health monitoring. A variety of system implementations are compared in an approach to identify the technological shortcomings of the current state-of-the-art in wearable biosensor solutions. An emphasis is given to multiparameter physiological sensing system designs, providing reliable vital signs measurements and incorporating real-time decision support for early detection of symptoms or context awareness. In order to evaluate the maturity level of the top current achievements in wearable health-monitoring systems, a set of significant features, that best describe the functionality and the characteristics of the systems, has been selected to derive a thorough study. The aim of this survey is not to criticize, but to serve as a reference for researchers and developers in this scientific area and to provide direction for future research improvements.

/pdf/a-survey-on-wearable-sensor-based-systems-for-health-2d9ei1xsz2.pdf

A Survey on Wearable Sensor-Based Systems for Health Monitoring and Prognosis

A number of current face recognition algorithms use face representations found by unsupervised statistical methods. Typically these methods find a set of basis images and represent faces as a linear combination of those images. Principal component analysis (PCA) is a popular example of such methods. The basis images found by PCA depend only on pairwise relationships between pixels in the image database. In a task such as face recognition, in which important information may be contained in the high-order relationships among pixels, it seems reasonable to expect that better basis images may be found by methods sensitive to these high-order statistics. Independent component analysis (ICA), a generalization of PCA, is one such method. We used a version of ICA derived from the principle of optimal information transfer through sigmoidal neurons. ICA was performed on face images in the FERET database under two different architectures, one which treated the images as random variables and the pixels as outcomes, and a second which treated the pixels as random variables and the images as outcomes. The first architecture found spatially local basis images for the faces. The second architecture produced a factorial face code. Both ICA representations were superior to representations based on PCA for recognizing faces across days and changes in expression. A classifier that combined the two ICA representations gave the best performance.

/pdf/face-recognition-by-independent-component-analysis-18jze5s1bq.pdf

Face recognition by independent component analysis

An extension of the infomax algorithm of Bell and Sejnowski (1995) is presented that is able blindly to separate mixed signals with sub- and supergaussian source distributions. This was achieved by using a simple type of learning rule first derived by Girolami (1997) by choosing negentropy as a projection pursuit index. Parameterized probability distributions that have sub- and supergaussian regimes were used to derive a general learning rule that preserves the simple architecture proposed by Bell and Sejnowski (1995), is optimized using the natural gradient by Amari (1998), and uses the stability analysis of Cardoso and Laheld (1996) to switch between sub- and supergaussian regimes. We demonstrate that the extended infomax algorithm is able to separate 20 sources with a variety of source distributions easily. Applied to high-dimensional data from electroencephalographic recordings, it is effective at separating artifacts such as eye blinks and line noise from weaker electrical signals that arise from sou...

/pdf/independent-component-analysis-using-an-extended-infomax-1atqgwzowt.pdf

Independent component analysis using an extended infomax algorithm for mixed subgaussian and supergaussian sources

In this paper, we developed and evaluated a robust single-lead electrocardiogram (ECG) delineation system based on the wavelet transform (WT). In a first step, QRS complexes are detected. Then, each QRS is delineated by detecting and identifying the peaks of the individual waves, as well as the complex onset and end. Finally, the determination of P and T wave peaks, onsets and ends is performed. We evaluated the algorithm on several manually annotated databases, such as MIT-BIH Arrhythmia, QT, European ST-T and CSE databases, developed for validation purposes. The QRS detector obtained a sensitivity of Se=99.66% and a positive predictivity of P+=99.56% over the first lead of the validation databases (more than 980,000 beats), while for the well-known MIT-BIH Arrhythmia Database, Se and P+ over 99.8% were attained. As for the delineation of the ECG waves, the mean and standard deviation of the differences between the automatic and manual annotations were computed. The mean error obtained with the WT approach was found not to exceed one sampling interval, while the standard deviations were around the accepted tolerances between expert physicians, outperforming the results of other well known algorithms, especially in determining the end of T wave.

A wavelet-based ECG delineator: evaluation on standard databases

Transthoracic electrical bioimpedance (TEB) measurement for acquiring of hemodynamic parameters e.g. stroke volume (SV) and cardiac output (CO) becomes commonly used. For precise and reliable measurement, a good understanding of the measurement system is needed to provide robust electrode placement, reproducible results, and large signal amplitudes. We propose an evaluation by MRI based human phantom with FEM based quasi-static solver for electrode placement in bioimpedance measurement. Placements according to Osypka, Cheetah and Bernstein et al. were simulated and compared with measurements taken from a (real) human subject. As a parameter for evaluation of signal quality, the percentage of current passing through the descending aorta from the overall injected current is measured. The placement according to Osypka results in 1.46 % of current flow through the descending aorta (Cheetah placement 1.12 %, Bernstein et al. placement 0.877 %) for the male human phantom DUKE. The simulation was compared with a real human subject (with comparable age, height, weight) by calculating the baseline impedance (Z 0 ). The simulation seems to fit at best with Osypka placement, where the deviation between simulated baseline impedance and real human subject is 19.9 % (deviation for Cheetah 34.1 %, deviation for Bernstein 62.5 %). The simulation with MRI based human phantoms seems to be a very good basis for further investigations of current injection and bioimpedance measurement comprehension.

Evaluation of Electrode Setups by MRI Based Human Phantom with FEM Based Quasi-Static Solver for Bioimpedance Measurement *

Es wird ein analytisches Verfahren entwickelt, mit dem die Wirbelstromverluste in einer Synchronmaschine bei stationarem Betrieb mittels der Feldtheorie berechnet werden konnen. Als Ergebnis einer Fourieranalyse des erregenden Standerstrombelages liegen dessen Oberwellen in den mitbewegten Lauferkoordinaten mit paarweise gleichen Zeitabhangigkeiten vor. Fur die verschiedenen Zeitabhangigkeiten sowie die zugehorigen Quer- und Langsfelder werden separate Randwertprobleme definiert, denen die abgewickelte ebene form der Maschine zugrunde gelegt wird. Die Losung erfolgt mit Hilfe einer Randbedingung dritter Art, welche auf den Oberflachen der hochpermeablen leitenden Pole naherungsweise den Skineffekt erfast. Die Verlustverteilungen auf den Poloberflachen sowie die Gesamtverluste werden fur verschiedene Oberwellenpaare angegeben. In Teil 2 wird die Naherungsmethode durch die Feldberechnung bei voller Berucksichtigung des Skineffektes uberpruft, eine einfache Moglichkeit zur Abschatzung der Ergebnisse angegeben und auf die Lastabhangigkeit der Verluste eingegangen.

Wirbelstromverluste in Massivpolen von Synchronmaschinen bei stationärem Betrieb

Recent technological advances are enabling the sector of wearables to rapidly expand. An increasingly big emphasis is placed on introducing continuous monitoring of biosignals such as heart rate, SpO2 and ECG. Extrapolating from the current trend of making clinical signals available to the general public, arterial blood pressure is a realistic and useful next step. Its robust, non-obtrusive, continuous acquisition is of great interest and importance, especially in clinical settings. Using a body sensor network, we acquire photoplethysmography, bioimpedance and pulse applanation tonometry signals synchronously at the periphery. The pulse waves are decomposed into spectral amplitude and phase. The values thereof are then used as features for blood pressure estimation. We apply a polynomial regression method and evaluate performance employing leave one out cross validation. A single initial parameter calibration is applied for the entire measurement. Our results with noisy datasets exhibit acceptable tracking of mean arterial pressure, with a baseline error of 0.62 mmHg, absolute error of 4.6 mmHg and standard deviation of 5.3 mmHg. Spectral phase information mostly outperforms amplitude information. Optically, electrically and electro-mechanically derived signals perform similarly, but best compliance and quality is achieved for electrical bioimpedance. We conclude that using a single, peripherally acquired plethysmographic waveform for blood pressure tracking is feasible, offering increased placement flexibility and compliance at the cost of reduced accuracy.

Spectral Parametrization of PPG, IPG and pAT Pulse Waves for Continuous Noninvasive Blood Pressure Estimation

Data Reduction in High-quality Audio Signals

Decomposition of evoked magnetoencephalography (MEG) data into their underlying neuronal signals is an important step in the interpretation of these measurements. Often, independent component analysis (ICA) is employed for this purpose. However, ICA can fail as for evoked MEG data the neuronal signals may not be statistically independent. We therefore consider an alternative approach based on the recently proposed shifted factor analysis model, which does not assume statistical independence of the neuronal signals. We suggest the application of this model in the time domain and present an estimation procedure based on a Taylor series expansion. We show in terms of synthetic evoked MEG data that the proposed procedure can successfully separate evoked dependent neuronal signals while standard ICA fails. Latency estimation of neuronal signals is an inherent part of the proposed procedure and we demonstrate that resulting latency estimates are superior to those obtained by a maximum likelihood method.

Reinhold Orglmeister

Papers

Evaluation of Electrode Setups by MRI Based Human Phantom with FEM Based Quasi-Static Solver for Bioimpedance Measurement *

Wirbelstromverluste in Massivpolen von Synchronmaschinen bei stationärem Betrieb

Spectral Parametrization of PPG, IPG and pAT Pulse Waves for Continuous Noninvasive Blood Pressure Estimation

Data Reduction in High-quality Audio Signals

Shifted factor analysis for the separation of evoked dependent MEG signals.