This paper proposes a new approach for drawing mode detection in online handwriting. The system classifies groups of ink traces into several categories. The main contributions of this work are as follows. First, we improve and optimize several state-of-the-art recognizers by adding new features and applying feature selections. Second, we use several classifiers for the recognition. Third, we perform multiple classifier combination strategies for combining the outputs. Finally, a large experimental evaluation on two data sets is performed: the publicly available Touch&Write database which has been acquired on a pen-enabled multi-touch surface, and the publicly available IAMonDo-database which serves as a benchmark. In our experiments on the IAM-OnDo-database we achieved a recognition rate of 97%, which is much higher than other results reported in the literature. On the more balanced multi-touch surface data set we achieved a recognition rate of close to 98%.
In this paper a method for Early Recognition (ER) of Motion Templates (MTs) is presented. We define ER as an algorithm to provide recognition results before a motion sequence is completed. In our experiments we apply Long Short-Term Memory (LSTM) and optimize the training for the task of recognizing the motion template as early as possible. The evaluation has shown that the recognition accuracy for a frame-by-frame classification the LSTM achieves a recognition accuracy of 88% if no training data of the person him/herself is included, and 92% if the training data also contains motion sequences of the person. Furthermore, the average earliness - the number of time frames it takes before the LSTM correctly classifies a motion pattern - is around 24.77 frames, which is less than a second with the used tracking technology, i.e., the Microsoft Kinect.
Heart rate turbulence (HRT) is a biphasic reaction to a ventricular premature contraction (VPC) mainly mediated by the baroreflex. It can be used for risk stratification in different disease patterns. Despite existing standards there is a lot of variation in terms of measuring and calculating HRT, which complicates research and application. Objective: This systematic review outlines and evaluates the methodological spectrum of HRT research, especially filtering criteria, parameter calculation and thresholds. Approach: The analysis includes all research papers written in English that have been published before 12.10.2018, are listed on PubMed and involve calculation of HRT parameter values. Main results: HRT assessment is still being performed in various ways and important specifications of the methodology are not given in many articles. Nevertheless, some suggestions regarding HRT methodology can be made: A normalised turbulence slope should be used to uncouple the parameter from heart rate and frequency of extrasystoles. Filtering criteria as formerly reviewed in the guidelines should be met and mentioned. The minimal number of VPCSs as well as new cut-off values for different risks need to be further evaluated. Most importantly, the exact and complete methodology must be described to ensure reproducibility and comparability. Significance: Methodical variation hinders comparability of research and medical application. Our continuing questions help to further standardise the measurement and calculation of HRT and increase its value for medical risk stratification.
3 June 2021 in npj Systems Biology and Applications
Reuse of mathematical models becomes increasingly important in systems biology as research moves toward large, multi-scale models composed of heterogeneous subcomponents. Currently, many models are not easily reusable due to inflexible or confusing code, inappropriate languages, or insufficient documentation. Best practice suggestions rarely cover such low-level design aspects. This gap could be filled by software engineering, which addresses those same issues for software reuse. We show that languages can facilitate reusability by being modular, human-readable, hybrid (i.e., supporting multiple formalisms), open, declarative, and by supporting the graphical representation of models. Modelers should not only use such a language, but be aware of the features that make it desirable and know how to apply them effectively. For this reason, we compare existing suitable languages in detail and demonstrate their benefits for a modular model of the human cardiac conduction system written in Modelica.
The Hodgkin-Huxley model of the squid giant axon has been used for decades as the basis of many action potential models. These models are usually communicated using just a list of equations or a circuit diagram, which makes them unnecessarily complicated both for novices and for experts. We present a modular version of the Hodgkin-Huxley model that is more understandable than the usual monolithic implementations and that can be easily reused and extended. Our model is written in Modelica using software engineering concepts, such as object orientation and inheritance. It retains the electrical analogy, but names and explains individual components in biological terms. We use cognitive load theory to measure understandability as the amount of items that have to be kept in working memory simultaneously. The model is broken down into small self-contained components in human-readable code with extensive documentation. Additionally, it features a hybrid diagram that uses biological symbols in an electrical circuit and that is directly tied to the model code. The new model design avoids many redundancies and reduces the cognitive load associated with understanding the model by a factor of 6. Extensions can be easily applied due to an unifying interface and inheritance from shared base classes. The model can be used in an educational context as a more approachable introduction to mathematical modeling in electrophysiology. Additionally the modeling approach and the base components can be used to make complex Hodgkin-Huxley-type models more understandable and reusable.
The Modelica language is becoming increasingly popular among scientists and engineers as platform for modelling physical or biological systems. Although Modelica is maintained as non-proprietary language by the Modelica Association, a considerable number of commercial implementations and development environments is complemented by a surprisingly small number of open source tools. In this paper, we present the communication service Mo|E that connects any text editor as front-end with a Modelica compiler as back-end. Based on the simple HTTP communication protocol, editor plugins for a software developer's favourite text editor can be developed easily, hence turning any editor into a lightweight Modelica development tool. We also present a first implementation of a plugin for the text editor Atom that exhibits features necessary for efficient software development, such as display of compile errors, code completion, go to declaration or view of context-sensitive documentation. In addition, Modelicaspecific checking of the number of equations in a model is supported.