Search results

Clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. EACPR/AHA Joint Scientific Statement

From an evidence-based perspective, cardiopulmonary exercise testing (CPX) is a well-supported assessment technique in both the United States (US) and Europe. The combination of standard exercise testing (ET) [i.e. progressive exercise provocation in association with serial electrocardiograms (ECGs), haemodynamics, oxygen saturation, and subjective symptoms] and measurement of ventilatory gas exchange amounts to a superior method to: (i) accurately quantify cardiorespiratory fitness (CRF), (ii) delineate the physiologic system(s) underlying exercise responses, which can be applied as a means to identify the exercise-limiting pathophysiological mechanism(s) and/or performance differences, and (iii) formulate function-based prognostic stratification. Cardiopulmonary ET certainly carries an additional cost as well as competency requirements and is not an essential component of evaluation in all patient populations. However, there are several conditions of confirmed, suspected, or unknown aetiology where the data gained from this form of ET is highly valuable in terms of clinical decision making.1

The Data Analytic Capability Wheel: An Implementation Framework for Digitalization

Propositions on peak data

Data, the raw material from which information is derived, is stored, copied, moved and modified more easily than ever. This quantum leap reaches levels outside our imagination. Surrounded by sensors, recommendation systems, invisible algorithms, spreadsheets and blockchains, the ‘difference that makes a difference’ can no longer be identified. Big Data is a More Data ideology, driven by old school hypergrowth premisses. As Nathan Jurgenson once observed: “Big Data always stands in the shadow of the bigger data to come. The assumption is that there is more data today and there will necessarily be even more tomorrow, an expansion that will bring us ever closer to the inevitable pure ‘data totality.” (2) Nothing symbolizes the current hypergrowth obsession better than Big Data. Let’s investigate what happens when we apply degrowth to data and reserve datafication–as a decolonial project, a collective act of refusal, an ultimate sign of boredom. We’re done with you, data system, stand out of my light.

MIMU PDR with bias estimation using an optimization-based approach.

Traditional IMU based PDR systems suffer from rapidly growing drift effects due to the inherent bias of the inertial sensor. Many existing solutions to mitigate this problem use aiding sensors or information as heuristics or map data. We propose a new optimization framework to solve the PDR estimation problem where the sensors biases are explicitly included as state variables and therefore be used to correct for bias effects in the PDR. By using a smoothing approach and exploiting the rigid structure of a MIMU array one can solve for the slowly varying sensor biases. This paper presents the method and gives an exemplary result of a walking trial. Good agreements in the position and orientation with an optical reference system were found. Moreover, accelerometer and gyroscope biases could be estimated accordingly. Further research includes the performance of more experiments under various conditions such that a more quantitative evaluation can be obtained. In addition, an exploration of a (pseudo) realtime filter version would be valuable such that the system can be applied online.

Demand Forecasting

Here, we delve into Demand Forecasting via Machine Learning, dissecting how to predict future demand using time-sensitive data. Westveer highlights key forecasting models, from the basic Simple Exponential Smoothing to the advanced SARIMA, applied to an electricity production dataset. The session, encapsulating the essence of data-driven forecasting, culminates in a compelling three-year predictive outlook, illustrating the transformative potential of machine learning in strategic planning and decision-making.

Effects of data smoothing on the reconstruction of helical axis reconstruction in human joint kinematics

In biomechanical joint-motion analyses, the continuous motion to be studied is often approximated by a sequence of finite displacements, and the Finite Helical Axis(FHA) or "screw axis" for each displacement is estimated from position measurements on a number of anatomical or artificial landmarks. When FHA parameters are directly determined from raw (noisy) displacement data, both the position and the direction of the FHA are ill-determined, in particular when the sequential displacement steps are small. This implies, that under certain conditions, the continuous pathways of joint motions cannot be adequately described. The purpose of the present experimental study is to investigate the applicability of smoothing (or filtering)techniques, in those cases where FHA parameters are ill-determined. Two different quintic-spline smoothing methods were used to analyze the motion data obtained with Roentgenstereophotogrammetry in two experiments. One concerning carpal motions in a wrist-joint specimen, and one relative to a kinematic laboratory model, in which the axis positions are a priori known. The smoothed and nonsmoothed FHA parameter errors were compared. The influences of the number of samples and the size of the sampling interval (displacement step) were investigated, as were the effects of equidistant and nonequidistant sampling conditions and noise invariance