Frequently Asked Questions

The ‘Classifier’ is a DFX Points category that indicates the output data point is significantly processed or produced by a NuraLogix developed CLASSIFIER.

The ‘Model’ refers to a DFX Points category that indicates the output data point is significantly processed or produced by a NuraLogix developed machine trained MODEL.

The ‘Feature’ refers to a DFX Points category that indicates the output data point results from a NuraLogix generated FEATURE.

The ‘Algorithm’ refers to a DFX Points category that indicates the output data point results from a NuraLogix developed ALGORITHM.

A DFX ‘Source’ is a DFX Points category that indicates the output data point originates from a “raw” or minimally processed data stream.

Typically ‘SOURCE’ category reflects the INPUT data that is directly injected from the client application, is passed through as output points, and not created by the DeepAffex™ engine.

Over the last century, medical researchers have observed and documented in the literature modulations of blood-flow that appear to carry information that is relevant and correlated with the subject’s physiology.

In this way human ‘biosignals’ that are encoded within the circulatory system may expose the subject’s underlying physiological state.

One way that DeepAffex™ makes information available to the user (researcher) is through the recovery of FBF envelopes.

By applying demodulating techniques to the (recovered) ‘biosignal’ envelopes, the source (information) may be decoded.

Examples of low-frequency (lf) modulations that are routinely recovered from FBF data include the Mayer Wave, Traube-Hering Wave and the Thermal Wave.

Examples of cyclical or periodic waveforms that are routinely recovered from FBF data include the Breathing (respiration cycle) envelope and the Heart Beat envelope.

The ROI is a programmable pixel-level geometric outline that becomes ‘virtually’ superimposed onto the face of the subject, whose purpose is to define a coverage area for where the facial blood-flow (FBF) extraction will be targeted.

The DeepAffex™ SDK extraction library performs simultaneous extractions from multiple facial ROI masks while maintaining synchronization with the video frame timing.

The facial blood-flow (FBF) extracted data is provided to the DeepAffex™ user under the ‘Regions Of Interest’ (ROI) group of DFX points.

The DeepAffex™ engine currently supports 3x categories of ROI masks:

• A ‘default’ set of ROI masks that are generated according to anatomical landmarks of the face
• User defined or ‘configurable’ ROI masks such as combinations of existing ROIs or defining entirely new geometric shapes
• An ‘experimental’ set of ROI masks that are systematically generated according to simple patterns such as a facial grids

A growing listing of available DFX Points can be reviewed on https://www.deepaffex.ai.

Nuralogix provides a Software Development Kit (SDK) for the software integration of client-side TOI extraction of facial blood flow.

• Contactless – No sensors to touch. Measurements are taken just like a selfie!
• Quick – A measurement only takes 30 seconds.
• Measure your resting heart rate and stress level. Our Anura Stress Index (ASI) can help you quantify and monitor your stress over time.
• A history of your measurements is saved, and you can share your results (should you choose) with your family, friends or doctor.
• Private – No images or video are saved on your device or transmitted as part of a measurement. Your blood flow information is encrypted when sent to our cloud servers for analysis and your measurement history is stored using the same encryption techniques that are used by banks.
• Use Anura™ often to increase your awareness of stress and general health.
• Completely free!

• Human skin is translucent.
• Light and its respective wavelengths are reflected at different layers below the skin and can be used to reveal blood flow information in the human face.
• This information is contained in conventional videos.
• A technique known as Transdermal Optical Imaging (TOI™) is used to extract blood flow information from the video. This technique was developed by our team of neuroscience, psychology, physiology, signal processing and deep learning experts.
• Anura™ uses your phone’s video camera to extract and send encrypted blood flow information (no video or photo) to the cloud to be processed by our DeepAffex™ Affective Intelligence engine.
• DeepAffex™ applies advanced signal processing and AI models to predict your resting heart rate and stress, then sends these results back to Anura.

‘SNR’ is a generalized 3 letter acronym that refers to a “Signal-to-Noise-Ratio” value expressed in logarithmic units of dB (decibels).

The Anura™ app indicates the SNR as a “quality metric” that applies to the purity of the extracted Facial Blood-Flow (FBF) data.

Anura™ calculates the SNR by comparing the extracted FBF signal levels against the background “out-of-band” level (noise floor) and representing the quality as the log-ratio of these two levels.

The Anura Enterprise STAGE 1 (centralized) collection requires designatedOPERATOR users to apply their licensed Anura™ (devices) for collecting measurements (data and results) on behalf of subjects according to the prescribed STUDY protocol.

The OPERATOR must enter a unique identifier in Anura™ to track each individual participant/subject/member/student (e.g. a mobile cell #) measurements.

Each OPERATOR may measure an unlimited number of individuals.

There is no limit on the number of measurements collected per individual.

The Anura Enterprise STAGE 2 (decentralized) collection of the STUDY expands the collection activities to all individuals that are selected to enroll in the STUDY.

Each participant/subject/member/student will use their personally licensed Anura (app) to collect measurement(s) on themselves according to the prescribed STUDY protocol.

Every individual must supply unique identification in Anura (e.g. their mobile cell #).

There is no limit on the number of measurements an individual may collect.

The Anura™ Enterprise STAGE 2 license enables the Enterprise to conduct fully decentralized collection of measurements (data and results) for an (unlimited) number of individual PARTICIPANTS (subjects).

The +E2 option option refers to Anura™ Enterprise mode using a STAGE 2 license – whereby any user that is ‘targeted’ by the Enterprise (i.e. such as a student or an employee) is able to use Anura app to conduct personal measurements but with all results stored under the Enterprise assigned OPERATOR and STUDY values.

The STAGE 2 license is distributed as a QR-CODE (matrix barcode) and is used to enable Anura Enterprise mode when (a) user performs a mobile QR-CODE scan of the Enterprise license (key) to activate the Anura app followed by (b) the user then performs an ‘Sign-Up’ or account registration.

Note that STAGE 2 users do not require preassigned DFX accounts prior to activating the Anura™ app on their device (using the QR-CODE license key).

The Anura™ Enterprise STAGE 1 license enables the Enterprise to maintain centralized control over the collection of measurements (data and results) by pre-assigning OPERATORS for measurement collections.

The +E1 option refers to Anura™ Enterprise mode using a STAGE 1 license – whereby designated DFX OPERATOR users (accounts) are preassigned by the Enterprise to conduct Anura app measurements on behalf of an unlimited number of PARTICIPANTS (subjects).

The STAGE 1 license is distributed as a QR-CODE (matrix barcode) and is used to enable Anura ‘Enterprise mode’ when (a) user that has been designated by the Enterprise as an OPERATOR performs a mobile QR-CODE scan of the Enterprise license (key) to activate the Anura app followed by (b) the user then performs a ‘login’ using a preassigned username (account).

Once both steps have been completed the Anura™ app is then enabled for measurements (collection) using the Enterprise assigned ORGANIZATION and STUDY parameters.

Note that STAGE 1 users must have preassigned DFX accounts prior to activating the Anura app on their device (using the QR-CODE license key).

The +E option refers to the B2B licensed Anura™ when operating in the Enterprise mode.

This is a licensed operating mode for Anura™ which provides assignment of both the ORGANIZATION and STUDY configuration to the Enterprise (licensed) values.

This mode allows individually licensed Anura™ apps to produce measurements (data and results) which will be linked on DFX under the (Enterprise) ORGANIZATION and (Enterprise) STUDY configuration.

Note that this operating mode provides any Enterprise (i.e. business) with the means to generate a secure and private collection of measurements (data and results) that may be subdivided into as many STUDIES as required according to its business needs.

The +C option refers to unlicensed Anura™ operating in the ‘Consumer mode’.

This is the default (non-licensed) operating mode for the Anura™ app whenever it is downloaded/installed from a mass distribution point (e.g. iOS = Apple app store, Android = anywhere).

In this operating mode Anura supports both ‘anonymous’ users (those users who chose NOT to create a DFX login account) as well as ‘registered’ users (those users who DO create DFX login accounts).

Regardless of the user login status all measurements taken by all users are collected/stored/processed under a common global ORGANIZATION label (defaults to NURALOGIX) and with a common global STUDY label (defaults to ANURA).

This is true for both our China DeepAffex™ cloud as well as our International DeepAffex™ cloud engines and services.

Note that this assigns all measurements (data and results) produced by Anura™ users running in +C mode will be directly linked to the ORGANIZATION labelled as ‘NURALOGIX’.

We began by taking continuous heart rate measurements of a large sample using traditional, contact-based methods (e.g. EKG). We then used this data to generate 3-D models of stress to derive a ground truth measure of stress using three dimensions.

Using TOI™ signals, we obtain heart rate variability (HRV) measures based on time domain, frequency domain, statistical, dynamic systems, and nonlinear methods extracting over 20 features.

By applying advanced machine learning techniques and other statistical methods, we maximized the correspondence between the more than 20 features extracted from the 30 second TOI™ signal and the ground truth.

We then computed the statistical means and standard deviations of the predicted stress values to create a norm based on a Gaussian distribution.

Finally, we converted the predicted stress values into a stress index with 5 levels where 1 is the lowest and 5 is the highest. The difference between each level represents one standard deviation.

The Anura™ heart-rate measurements is accurate to within +/- 1 bpm when compared against electrocardiogram (ECG) measurements.

Anura is only intended for informational and educational purposes to promote general wellness.

The measurements results provided to you should not be treated as medical advice or to diagnose, treat, cure or prevent and disease, ailment or injury.

Should you have any health-related questions, please call or see your physician or other medical professional.

Anura™ is a general wellness app that uses your mobile phone’s video camera to measure, track and maintain a record of your resting heart rate and stress level based on the blood flow in your face.

It uses Affective Intelligence (developed by NuraLogix) to transform your mobile phone into your personal health AI™.

Anura’s primary goal is to increase people’s awareness of stress and general health.

Anura™ is available for download as both an iOS and an Anroid app.

The Anura™ heart rate result is calibrated to align with the medically indicated ranges for the human resting heart rate.

The medically indicated range for a “normal” resting heart rate lies between 60 bpm and 100 bpm.

When Anura™ measures a heart rate that falls lower than 60 bpm, the result will be indicated on the “below normal” range (Bradycardia).

When Anura™ measures a heart rate that is above 100 bpm, the result will be indicated on the “above normal” range (Tachycardia).

If you are a B2B application developer, we are committed to make it easy for you to build solutions that take advantage of the power of DeepAffex™.

To get started you must (a) sign up for a DeepAffex™ developer account (b) register your app to get an app key and secret then (3) download our SDK, Cloud API documentation and examples to begin developing your DFX enabled solution.

Yes – the use of the DFX network API is restricted to customers holding a valid DeepAffex™ license.

The DFX Cloud API enables customers to programmatically access and manage their organization’s DeepAffex™account data and results.

The DFX Cloud API is published as reference specification (documentation) describing the programmatic interfaces (or endpoints) that are provided by the DeepAffex™ cloud engine.

These network endpoints allow a remote client (program) to invoke and access all DeepAffex™ micro-services and features.

The DFX Cloud API utilizes a combination of REST for synchronous (blocking) transactions as well as Web-socket protocols for asynchronous (non-blocking) communications.

One possibility is that you have exhausted your count (quantity) of licensed devices.

Every physical device (handset, desktop, laptop) that requires client communication with DFX must first register with the API to receive a unique device ‘token’. The maximum number of devices that are permitted to connect with DFX is controlled by your license-key.

You can make requests to the API to ‘unregister’ any/all currently registered device ‘tokens’ and reclaim device balance.

The DFX Dashboard enforces Role-Based-Access (RBA) authorization to all user privileges.

The DFX Dashboard is available to all users with a licensed DFX account.

The DFX ‘Org-Admin’ (or OA) user is a Dashboard assigned user role that is responsible for managing the ORGANIZATION on DeepAffex™.

The OA user is an DFX administrator who manages the ORG configuration on behalf of the Enterprise (i.e. business).

The DFX ‘Lead researcher’ (or LR) user is a Dashboard user role that is similar to a Principal Investigator (PI) role in a scientific study.

The LR user has DFX access authority over multiple DR users.

The DFX ‘Researcher’ (or DR) user is a Dashboard user role that is similar to a Lab Researcher’s role in coordinating a scientific study.

The DR user has DFX access authority over multiple OPERATOR users.

The DFX ‘Operator’ (or OP) user is a Dashboard user role that is similar to a Field Researcher’s role in conducting a scientific study.

The DR user is the account holder, parent and owner of multiple subject / member / participant (profile) measurements.

The DFX Dashboard is a web-based browser tool that is utilized for all access, configuration, management and administration of DeepAffex™ account data and results.

The DFX Dashboard is available to users with a registered DFX account.

All DeepAffex™ services and functionality may be fully integrated into your software application by using the DFX SDK library and the Cloud API references both available from NuraLogix.

The STUDY exists at the second-highest level (i.e. under the ORGANIZATION) of the data hierarchy and partitioning schema for all measurements that reside within DFX.

The STUDY is used to create measurement ‘subsets’ within an ORGANIZATION and each ORGANIZATION must provide at least one (1) STUDY to contain its measurement data/results.

The STUDY is a secondary means of segregating measurement data/results under DFX.

Note that DeepAffex™ supports an unlimited number of non-overlapping STUDIES under an ORGANIZATION each of which may contain an unlimited number of measurement data/results.

The template is the association that binds DFX Points with a STUDY.

A template consists of a pre-defined set or collection of DFX Points that are assigned (by the SA) to an ORGANIZATION.

A template defines the collection/group of DFX Points that are assigned (by the SA) to an ORGANIZATION.

The ORGANIZATION exists at the highest level (i.e. at the top) of the data hierarchy and partitioning schema for all measurements that reside within DFX.

For this reason, the ORGANIZATION is designated as the true ‘owner’ of any/all measurement data/results that are taken/stored with this label.

The ORGANIZATION is therefore the primary means of segregating and isolating utilized by the DeepAffex™ cloud.

Note that DeepAffex™ supports an unlimited number of independent ORGANIZATION(s) each of which may contain unlimited measurement data/results.

In the B2B cloud services model, the ORGANIZATION represents the DeepAffex™ partitioning ‘silo’ that gets created and licensed by a business.

All DeepAffex™ measurement configurations, data/results are securely and privately isolated according to the ORGANIZATION (silos).

There is no possibility for data visibility or information leakage across/between ORGANIZATION(s).

Each ORGANIZATION (short form ORG) is completely independent and private from every other ORG.

Only the SUPER ADMIN (SA) user (i.e. NuraLogix employee) has the role-based-access privileges required for creating the DeepAffex™ ORG(s) ‘silos’.

By default all worldwide TOI measurement data/results made by (unlicensed) users/devices are lumped together on DeepAffex™ under a single, shared ORGANIZATION (container) with access rights assigned to NuraLogix.

However if a business requires access rights, ownership and partitioning of measurement data/results from users pertaining to their particular business interests, then the business would acquire a DeepAffex™ ORGANIZATION license for this purpose.

A DeepAffex™ ORGANIZATION license will provide the business with the designated rights to access a private, isolated ORGANIZATION with ownership of all measurement data/results granted to the business.

The DeepAffex™ ORGANIZATION license guarantees partitioning and segregation of all business configured measurement data/results.

Refers to the entire DeepAffex™ cloud-hosted infrastructure (including API, database, automation and TOI analysis components)

‘DFX’ is a generalized 3-letter acronym (short form) referring to the DeepAffex™ cloud service product from NuraLogix.

The (short form) ‘DFX’ acronym is often used within our documentation in place of the (longer form) DeepAffex™ name.

The term Affective AI refers to the ability of a computing device to detect, measure and analyze human affects that take the form of physiological and psychological states.

A different kind of “AI”.

The NuraLogix DeepAffex™ engine is cloud-based Affective Intelligence platform that utilizes innovative facial blood-flow imaging technology to assess human physiology and psychological affects.

You are invited to sign up for your personal DeepAffex “Trial” account at DeepAffex. Navigate to the upper corner to find the “My Account” link where you will see the ‘login’ dialog box. Choose the ‘Sign-up’ option and complete the simple form requesting a valid username, password, region etc. Next be sure to review our Software License Agreement (SLA) and indicate your acceptance using the checkbox. Once the form has been submitted, you will receive a “Verification” email followed by a “Welcome” email summarizing the details of your account and immediatly providing a “Trial” license-key for accessing DeepAffex.

This “Trial” account registration enables you to develop your own custom (client) application using the SDK libraries, the API docs and taking advatage of the provided functional sample/refernce code all available from DeepAffex. The FAQpage is also a valuable resource for the new integrator.

The “Trial” account license-key provides access to a restricted set of DFX Points for a 90 day (3 month) evaluation period.

The TOI extraction process relies on the face-tracker (FT) component to locate and dynamically track the subject’s facial landmarks and head contour. Note that the face-tracker is NOT used to recognize or otherwise identify the subject.

The face-tracker data is used anonymously to provide additional value to the TOI extraction process. For example the face position (3D coordinates) and simple facial expressions may also be provided to our user.

We have also implemented powerful machine-trained models which accurately predict demographics information such as AGE / GENDER / HEIGHT / WEIGHT

The term ‘FBF’ is a generalized 3-letter acronym (short form) meaning Facial Blood Flow and may be used interchangeably. This label refers to the cross sectional bloodflow data that is retrieved or extracted from the subject’s face using TOI.

The (short form) ‘FBF’ acronym is often used within our documentation in place of the (longer form) ‘Facial Blood Flow’.

The term ‘TOI’ is a generalized 3-letter acronym (short form) for Transdermal Optical Imaging™ and may be used interchangeably.

The (short form) ‘TOI’ acronym is often used within our documentation in place of the (longer form) ‘Transdermal Optical Imaging’.