Model-driven stimulation for targeted functional restoration in chronic spinal cord injury

Outline of structure
	-Review intial proposal documents
	-Review lab book & folder
	-Review Evernote notes
		-> put together draft structure & projects to focus on

Targeted functional restoration requires:
	Simulation of electric currents in patient-specific imaging
	Biophysical models of nerve activation
	Numerical optimisation of stimulation parameters
	Validation of model
	Applications

Intro / lit review - functional restoration / nerve stimulation methods / neural interfacing / noninvasive methods / use of computational techniques -> potential for use of advanced computational methods to develop novel treatment methods for nervous system disease using targeted stimulation
	Technical intro: electrical properties of tissue, biophysical models, FEM, combined simulations, etc.
	-> Split intro: applied intro / motivation -> intro to relevant techniques & fields

Electroanatomy - ?section on electrical properties of tissue & anatomic considerations relevant to stimulation
	Skin electrical properties
	Average skin thickness in relevant regions, etc.
	?electrical properties of other tissues if required
	Could also model effects of changes, anisotropy, etc.
	Depth of sacrum, vertebral column, etc. in tissue (on average) -> implications for targeting
	Anatomic windows for targeting & positions
	Specific positions of sacral nerves, etc. for targeting
		?Reviews vs. original work possible here
		Imaging, anatomic specimens, etc.

Technology development: production of toolset to allow patient-specific modelling of nervous system activation from applied currents
	Production of FEM from clinical imaging
		STL vs. NURBS vs. volume
		Full workflow -> automation & comparison to existing methods
		?.oni format, parsers for method, adaptation for HPC, etc.
	Mapping of arbitrary geometries into scan
		Rescaling & geodesic measurements
	Biophysical measures of nerve activation
		Activating functions vs. axon-cable models
	Mapping of biophysical models into anatomic space
		Model of spinal cord microanatomy, etc.
	Optimisation methods
		Derivative free optimisation with biophysical objective functions
		Adaptive mesh refinement
	Validation
		3D printing, proof of principle testing
		-> NRRDosurgery toolbox for automated meshing from segmented images & numerical analysis of electric fields in patient-specific imaging +/- optimisation
		3D printing of anatomic structures
	Production of conductive arrays
		Dual-material printing with conductive filaments & interfacing with existing electronics (?combine with production of optimised electrode geometries)
		?simple validation e.g. for neurophysiology - ulnar nerve, etc. - could d/w Prof. Jones

Development of clinical tools: use of computational tools to develop new treatment methods
	Optimisation of stimulation parameters
	Optimisation & testing of electrode geometries
	Interferential stimulation methods
	Network DBS

	Optimisation of stimulation parameters: for electrode in a given location, derive optimal stimulation parameters & derive optimal positioning
		tES: place in location & stimulate, ?accuracy of optimisation; & 3D models, etc. - pipeline for scan & position -> output parameters
		SCS: require array geometry -> map from fluoroscopy into CT -> optimise parameters -> compare to existing / trial optimal
	Optimisation of electrode geometries: test and optimise annular ring method - fields generated / activation / directionality (/focality) of stimulation
		Also printing of optimised electrode geometries: lit review / 3D printing methods / printing of conductive arrays based on simulations - printing methods, electrical performance, interfacing with systems, validation
	Interferential stimulation: use simulations to compare with interferential stimulation with high frequency stimulation
		Lit review re. interfering fields; model of interference vs. standard, ?validation
	Network DBS: lit review; map biophysical models to tracts; patient specific models

Applications: use of developed tools & theory to specific anatomic regions & pathologies
	Sacral: S3 nerve root - ?pain, etc.
	Lumbar: ?SCI
	Thoracic: ?pain
	Cervical: ?SCI
	Peripheral: stellate ganglion, DRG, ?sublingual, etc.
	Cranial: ?tES, DBS

	Sacral: relatively straightforward, superficial target - review of applications & current literature re. stimulation etc.; ?use as starting point for later applications
		Lit review / compare simplified vs. patient-specific simulation / 3D printed validation of fields / ?compare electrode geometries
	Lumbar & cervical: some literature re. SCI - review this & model existing montages, compare to models of epidural stimulation; ?try to do early to use as basis for doing pilot study if working / collaborate with other groups e.g. Edgerton
		Lit review / model existing montages
		?validate by targeting dorsal columns, reflexes, etc.
	Thoracic: lit review & model & ?pilot with paraesthesias; ?possibility of pilot with pain
	Peripheral: not an immediate focus; DRG may be an application of methods at various levels
	Cranial: could compare to existing methods (SimNIBS, ROAST, etc.)
		Lit review re. tES simulation; assess level of patient-specific imaging; trial model of brain - ?NRRDosurgery for cranial tES
		Network DBS: could integrate with tractography to do targeted stimulation of networks - Ashley willing to contribute re. tractography & application to clinical datasets
			Lit review re. patient-specific modelling
			Meshing of brain model
			Mapping of DBS electrodes into brain
			Mapping of biophysical models to tracts
			?optimisation of stimulation parameters / electrode position, etc.

Conclusion: computational methods allow simulation of electric fields in patient-specific models. Tools developed on this premise allow novel treatment methods, with possibilities incl. noninvasive SCS trialing, functional restoration and optimisation of electrode geometries and stimulation parameters
