@datagrok/bio-signals

Biosignals

BioSignals is a package for the Datagrok platfrom. The goal of the project is to offer an efficient and automated biosignal processing routine. The initial version is based on pyphysio - a python library developed by Andrea Bizzego.

The package reinforces the existing pyhton code with datagroks' visualization and data processing tools. The pipeline itself is designed with scientific community in mind, standartizing and thus facilitating the usual ECG, EEG, EDA, etc. signal processing workflows. The fusion of manual and automated steps is largely enabled by our interactive viewers , scripting capabilities, detector functions, data augmentation, and a curated collection of the scientific methods.

In particular, project's initial goals are:

Goal	Example
Automatically read various biosensor file formats	Integrate with the built-in file share browser
Provide efficient interactive visualizations for raw biosensor data, including domain-specific visualizations	"Head view" for EEG
Provide efficient ways for manipulating raw biosensor data	Marking regions, etc
Provide a collection of high-performance DSP algorithms	See DSP package
Detect type of signals, along with the metadata	Sampling rate, recorded events
Automatically suggest analyses and pipelines applicable to the current dataset, derive high-level features out of the raw biosensor signal	"Extract step count" for the accelerometry data
Visually define pipelines	Similar to Simulink block diagrams
Allow to build predictive models by integrating previously defined pipelines with theDatagrok's predictive modeling capabilities	Training a model to find "bad" quality segments based on the manually annotated data

Currently, the project is in its early stages, and we welcome you to contribute to this repository.

Pyphysio

Pyphysio is a library that contains the implementations of the most important algorithms for the analysis of physiological data. The latter methods are divided into the following categories:

Category	Input	Output	Examples
Filters	Signal	Filtered signal of the same type	Apply elliptic filter to raw ECG signal
Estimators	Signal	Signal of a different type	Get inter beat intervals (IBI) from ECG signal
Segmentators	Signal	Series of segments	Cut 24-hour Holter monitor record into 10-minute segments to compute how many arrhythmias occurred over different time intervals
Indicators	Signal	Value	Compute sequence of HRV indicators from segmented RR intervals
Tools	Signal	Arbitrary data	Detects outliers in the IBI signal, compute rising slope of R peaks, estimate the power spectral density (PSD) of the signal

How to add your script

1. Go to Functions | Scripts | Actions | New Script
2. Write your script, and test it on files
3. Set tag #filters, #estimators or #indicators

Now it is available in corresponding app section.

A unique approach to every signal

Signal type	Definition
ECG (Electrocardiogram)	Electrical activity of the heart
EDA (Electro-dermal activity)	Variation of the electrical conductance of the skin in response to sweat secretion
Accelerometer signal	Rate of change of body's velocity
EMG (Electromyography)	Electric potential generated by muscle cells when they are electrically or neurologically activated
EEG (Electroencephalogram)	Electrical activity of the brain
ABP (Arterial Blood Pressure signal)	Pressure of circulating blood against the walls of blood vessels
BVP / PPG (Blood Volume Pulse / Photoplethysmography)	Volumetric variations of blood circulation
Respiration

Since various signals require a different combination of filters and information extraction steps, a separate pipeline has to be designed for every input. We plan to first separately recreate the recommended workflows and then combine them into a complete package.

Automation

A substantial part of this project targets improvements to user-friendliness. Our goal is to create a smart environment, which streamlines the pre-processing steps by autonomously selecting and suggesting the most appropriate tools.

Detectors.js

A file containing functions for preliminary data analysis. Once any table is uploaded to Datagrok , this script decides whether the BioSignals package should be added to the 'Algorithms' list. Currently, the proposed mechanism relies on column labels, however its functionality can be extended to draw insights from actual data.

Types of auto-input

By design all the inputs could be split into three sub-types:

• User input: completely manual (for areas where no inferences from data can be made);
• Auto suggest: completely automated (input suggestions inferred from data);
• Auto Limit: automatically limit the options, given previous choices;

The pipeline can then be viewed as a branching decision tree, which offers and/or block certain paths depending on retrieved metadata and the sequence of inputs.