CONTROL-CORE: A Modular Simulation Environment for Design and Prototyping of Closed-loop Peripheral Neuromodulation Control Systems using the O2S2PARC Platform

Investigators

Babak Mahmoudi, Mayuresh Kothare, Rajanikanth Vadigepalli, Gautam Kumar, Charles Horn

Contact info (email)

b.mahmoudi@emory.edu

1. Define context(s)

aid in clinical trial design

identify/explore new therapies

Current Conformance Level / Target Conformance Level

Comprehensive

Primary goal of the model/tool/database

We will develop a virtual simulation environment called CONTOL-CORE for in silico experimentation with the design, analysis, prototyping, deployment, and execution of feedback control architectures for closed-loop stimulation of the peripheral nervous system. The modular algorithm libraries will be designed based on modularity, reusability, flexibility, and integration with O2S2PARC, with building blocks that could accommodate different languages (PYTHON, C, MATLAB). The applicability of the developed simulation platform will be demonstrated for the cardiac and gastrointestinal (GI) systems in simulation by leveraging existing data, models, and the O²S²PARC platform. The two chosen systems cover the two extremes of SPARC-relevant organs in two ways: (1) very slow (GI system) to very fast (cardiac) time scales; and (2) fully data-driven (GI system) to very detailed mechanistic (cardiac) models due to the different levels of basic understanding of these organs. The developed tools will likely to be extensible to other SPARC-relevant organs/functions that lie between these two extremes.

Biological domain of the model

Gastrointestinal system, Cardiac System, potentially other SPARC-relevant organs such as colon and bladder

Structure(s) of interest in the model

Organ-specific functional activity

Spatial scales included in the model

Few centimetres to the full organ

Time scales included in the model

Few seconds (cardiac system) to several minutes (GI system)

2. Data for building and validating the model

Data for building the model	Published?	Private?	How is credibility checked?	Current Conformance Level / Target Conformance Level
in vitro (primary cells cell, lines, etc.)
ex vivo (excised tissues)
in vivo pre-clinical (lower-level organism or small animal)	Yes. Data are available on cardiac VNS from rats. (Source -- literature)	Yes, data on gastric EMG is available from Ferrets (Dr. Horn’s Lab, U. of Pittsburgh).	Data are recorded from the same animal in multiple experimental trials.	Adequate
in vivo pre-clinical (large animal)	Yes, data on cardiac VNS is available from dogs and pigs (Dr. Ardell's Lab, UCLA)		Data are recorded from the same animal in multiple experimental trials and o	Adequate
Human subjects/clinical
Other: ________________________	Yes. The synthetic data will be generated using a published multi-scale cardiac model.		The published model was validated against the reported data from human subjects under healthy and heart failure physiological conditions.	Adequate

Data for validating the model	Published?	Private?	How is credibility checked?	Current Conformance Level / Target Conformance Level
in vitro (primary cells cell, lines, etc.)
ex vivo (excised tissues)
in vivo pre-clinical (lower-level organism or small animal)	Yes	Yes	Data are recorded from the same animal in multiple experimental trials.	Adequate
in vivo pre-clinical (large animal)	Yes
Human subjects/clinical
Other: ________________________	Yes. The synthetic data will be generated using a published multi-scale cardiac model.		The published model was validated against the reported data from human subjects under healthy and heart failure physiological conditions.	Adequate

3. Validate within context(s)

	Who does it?	When does it happen?	How is it done?	Current Conformance Level / Target Conformance Level
Verification
Validation	We have validated our cardiac and GI models, both physiological and data-driven, using published experimental data as well as simulated data.	During the development of physiological and data-driven modeling	Models have been validated either using the physiological data (e.g. heart rate) reported in the literature in response to VNS or simulated data (synthetic data generated from physiological models to develop data-driven models).	Comprehensive
Uncertainty quantification	We will do.	Throughout the development of the modular algorithm libraries for dynamic modeling and analysis and closed-loop control.	Since no model is ever perfect, we will perform rigorous robustness analysis to assess the impact of model uncertainty on the closed-loop performance. We will develop information-theoretic metric to quantify the uncertainty in our developed generative models.	Adequate
Sensitivity analysis
Other:__________
Additional Comments

4. Limitations

Disclaimer statement (explain key limitations)	Who needs to know about this disclaimer?	How is this disclaimer shared with that audience?	Current Conformance Level / Target Conformance Level
Since we will develop low-dimensional dynamical models that can be used in designing closed-loop control strategies, our central assumption is the existence of low-dimensional latent dynamical states of organ functions from which we can reconstruct the observed spatiotemporal dynamics subject to vagus nerve stimulations.	Users who are going to use our CONTROL-CORE libraries in designing vagus nerve stimulation parameters.	A list of limitations will be provided with each report/publication from each model/system run and stated in the accompanying documentation.	Extensive

5. Version control

Current Conformance Level / Target Conformance Level
Extensive

	Repository?	Code Review?
individual modeler	GitHub, O2S2PARC
within the lab
collaborators		Reimplementation of the code to reproduce the results.

6. Documentation

	Current Conformance Level / Target Conformance Level
Code commented?	Extensive
Scope and intended use described?	Extensive
User’s guide?	Extensive
Developer’s guide?	Extensive

7. Dissemination

Current Conformance Level / Target Conformance Level
Adequate

Target Audience(s):	“Inner circle”	Scientific community	Public
Simulations	Reports	Publications, Github, SPARC portal	Github, SPARC portal
Models	Reports	Publications, Github, SPARC portal	Github, SPARC portal
Software	O2S2PARC	O2S2PARC
Results	Reports	Publications	Publications
Implications of results	Reports	Publications	Publications

8. Independent reviews

Current Conformance Level / Target Conformance Level
Adequate

Reviewer(s) name & affiliation:
When was review performed?
How was review performed and outcomes of the review?

9. Test competing implementations

Current Conformance Level / Target Conformance Level
Adequate

	Yes or No (briefly summarize)
Were competing implementations tested?	No, we are in the phase of testing our closed-loop designs on ControlCore platform.
Did this lead to model refinement or improvement?

10. Conform to standards

Current Conformance Level / Target Conformance Level
Extensive

	Yes or No (briefly summarize)
Are there operating procedures, guidelines, or standards for this type of multiscale modeling?	Yes, our CONTROL-CORE simulation platform will be tested within the O2S2PARC.
How do your modeling efforts conform?	Yes, we will meet the requirements of O2S2PARC platform for integrating our CONTROL-CORE simulation platform within O2S2PARC.

11. (optional) Additional information to support items 1-10

We will use the containerization platform Docker to host algorithm libraries and develop a set of microservices (available as containers on O2S2PARC) to provide standard interfaces to pull data and/or existing models from repositories and make them accessible for simulating the closed-loop peripheral stimulation control systems. To facilitate “easy-to-use” CONTROL-CORE platform, we will use the Flowchart format in the O2S2PARC platform to develop both graphic-user-interface (GUI) and a coding interface that will enable users from a variety of backgrounds and from multiple sites to effectively utilize the platform in a collaborative fashion.