1. An electron lens for an electron emitter, comprising:
a focusing lens layer; and
a polymer spacer layer between the focusing lens layer and the electron emitter wherein the polymer spacer layer defining at least one opening having an undercut of about 1 micron to about 2 microns per about 6.5 microns of depth.
2. The electron lens of claim 1 wherein the polymer spacer material is between about 2 microns and about 12 microns thick.
3. The electron lens of claim 1 wherein the polymer spacer layer has been cured to remove volatile content.
4. A focused electron emitter comprising the electron lens of claim 1.
5. An electronic device comprising at least one electron lens of claim 1.
6. A focused emitter, comprising:
a tunneling layer less than about 500 angstroms in thickness disposed on a semiconductor substrate;
a polymer spacer layer disposed on the semiconductor substrate and defining a first opening, havinq an undercut of about 1 micron to about 2 microns per about 6.5 microns of depth, disposed over the tunneling layer;
a focusing lens layer disposed on the polymer spacer layer and defining a second opening disposed over the tunneling layer; and
a cathode layer disposed on the tunneling layer.
7. The focused emitter of claim 6 wherein the polymer spacer layer is between about 2 microns and about 12 microns thick.
8. The focused emitter of claim 6 wherein the polymer spacer layer has been cured to remove volatile content.
9. An electronic device comprising at least one focused emitter of claim 6.
10. The electronic device of claim 9 wherein the electronic device is a display device.
11. The electronic device of claim 9 wherein the electronic device is a mass storage device.
12. A focused electron emitter, comprising:
a tunneling layer less than about 500 angstroms in thickness;
means for focusing electrons emitted from tunneling layer; and
polymer means for spacing the means for focusing electrons from the tunneling layer wherein the polymer means has been cured to remove volatile components and defines an opening having an undercut of about 1 micron to about 2 microns per about 6.5 microns of depth.
13. The focused electron emitter of claim 12 wherein the tunneling layer is about 100 Angstroms.
14. The focused electron emitter of claim 12 wherein the means for focusing electrons and the polymer means for spacing have substantially the same temperature expansion coefficient.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A system that learns to associate ventricle-atrial (VA) intervals with temporal patterns of at least one hemodynamic sensor using neural network processing, comprising:
an input layer temporal synchronizer performing a pre-processing stage of at least one hemodynamic sensor input;
a middle layer consisting of dynamic synapses, wherein an array of dynamic synapses are grouped in columns and
an output layer consisting of one integrate and fire neuron for each dynamic synapse column of the middle layer and wherein a hit count rate membership function is for each integrate and fire neuron is calculated and an average normalized summation of all integrate and fire neurons gives the association of a VA intervals with the input hemodynamic sensor temporal patterns in all heart conditions.
2. A system according to claim 1 wherein said spiking neural network synapses are trained applying a Hebbian learning rule and wherein a learning rate parameter is changed locally and selectively in each synapse module according to the value of the hit count rate membership function of the post synaptic integrate and fire neuron and according to the activation state of the synapse such that a synapse that was active and caused a saturation of the hit count rate membership function of the output layer integrate and fire neuron reduce its learning rate parameter.
3. A system according to claims 1 and 2 wherein in each synapse a local and selective learning rate parameter is adjustable separately in order to improve stability and preserve plasticity of the artificial neural network.
4. An adaptive cardiac resynchronization therapy system, delivering biventricular stimulation to the heart with dynamic AV delay and VV interval, wherein said stimulation is modified continuously in correlation with the hemodynamic performance of the heart, and wherein said system uses a spiking neural network as in claim 1 that learns to associate the VA interval based on a hemodynamic sensor temporal patterns wherein the associated VA intervals replace the sensed atrial event signal during atrial fibrillation episodes.
5. An adaptive dual chamber pacemaker or ICD system, delivering ventricular stimulation to the heart with dynamic AV delay, wherein said stimulation is modified continuously in correlation with the hemodynamic performance of the heart, and wherein said system uses a spiking neural network as in claim 1 that learns to associate the VA interval based on a hemodynamic sensor temporal patterns wherein the associated VA interval replace the sensed atrial event signal during atrial fibrillation episodes.
6. A system according to claim 1 wherein said spiking neuron network architecture is implemented in a processor operating with low clock frequency at the range 1-10 KHz and with synaptic weight adaptation deriving reference from a hemodynamic sensor input and implementing a Hebbian learning rule.
7. A system for providing adaptive cardiac resynchronization therapy or dual chamber pacing or intracardiac cardioverter defibrillation wherein said system switches back an forth between a sensed atrial operation mode and an associated VA interval operation mode based upon temporal pattern recognition of the spiking neural network as in claim 1 wherein the associated VA interval replace the sensed atrial event signal as long as the spiking neural network associated VA interval is valid and where a switch back to working with the sensed event occurs when the VA interval association is poor or any system failure occurs.
8. A system as in claims 1 to 7 wherein said VA interval is obtained online using any other neural network architecture or learning module.
9. A method for regulating a controlled delivery of a physiologically active agent to a patient comprising the steps of:
obtaining continuous signal from at least one sensor monitoring at least one physiological parameter of said patient;
processing said at least one continuous signal using an algorithmic processing module and a learning module, and wherein said learning modules carries out adaptive learning in connection with said at least one sensor wherein said learning is controlled and supervised by said algorithmic module, and
delivering a physiological signal by a delivery module in response to said processed signal, wherein said delivery module is controlled by said algorithmic module.
10. A method as in claim 9 for delivering biventricular pacing to congestive heart failure patients that provides biventricular pacing also during atrial fibrillation episodes comprising the steps of:
programming initial atrioventricular (AV) delay parameter VV (interventricular delay) interval parameter of said algorithmic module;
learning to associate the VA interval with temporal patterns of hemodynamic sensor during normal sinus atrial rate and
switching on according to the associated VA interval when atrial fibrillation is detected;
switching back on in correlation with the sensed atrial event when normal sinus rhythm is recovered;
providing pacing in a non-adaptive CRT mode whereby pacing is provided according to said pre-programmed AV delay and VV interval parameters,
switching to an adaptive CRT mode wherein said AV delay and VV interval change dynamically in order to achieve optimal hemodynamic performance, and wherein said adaptive mode is limited to perform above a low limit of hemodynamic performance, and
switching back to the non adaptive CRT mode whenever the hemodynamic performance drops below a low limit of hemodynamic performance or a sensor failure or any other system failure is detected.
11. A method for regulating a controlled delivery of a physiologically active agent to a patient as in claim 9 as used for delivering ventricular pacing in dual chamber pacemakers or intracardiac cardioverter defibrillators (ICDs) that manages ventricular pacing also during atrial fibrillation episodes comprising the steps of:
programming initial AV (atrioventricular) delay parameter of said algorithmic module;
learning to associate the VA interval with temporal patterns of hemodynamic sensor during normal sinus atrial rate;
switching to working in correlation with the associated VA interval when atrial fibrillation is detected;
switching back to working in correlation with the sensed atrial event when normal sinus rhythm is recovered;
providing pacing in a non-adaptive mode whereby pacing is provided in correlation with said pre-programmed AV delay;
switching to an adaptive mode wherein said AV delay changes dynamically in order to achieve optimal hemodynamic performance, and wherein said adaptive mode performs only above a low limit of hemodynamic performance, and
switching back to the non adaptive mode whenever the hemodynamic performance drops below a low limit of hemodynamic performance or a sensor failure occurs or any other system failure is detected.
12. A method for regulating a controlled delivery of a physiologically active agent to a patient as in claim 9 whereby biventricular pacing is delivered that uses an associated VA interval replacing the sensed atrial event in the controller timing cycles, comprising the steps of:
programming initial AV delay parameter interventricular (VV) delay interval parameter of said algorithmic module;
learning to associate the VA interval with temporal patterns of hemodynamic sensor in correlation with the sensed atrial event;
switching to working in correlation with the associated VA interval when a convergence of learning occurs;
switching back to working in correlation with the sensed atrial event when failure occurs;
providing pacing in a non-adaptive CRT mode whereby pacing is provided in correlation with said pre-programmed AV delay and VV interval parameters;
switching to an adaptive CRT mode wherein said AV delay and VV interval change dynamically in order to achieve optimal hemodynamic performance, and wherein said adaptive mode is limited to perform above a low limit of hemodynamic performance, and
switching back to the non adaptive CRT mode whenever the hemodynamic performance drops below a low limit of hemodynamic performance or a sensor failure or any other system failure is detected.
13. A method regulating a controlled delivery of a physiologically active agent to a patient as in claim 9 for delivering ventricular pacing in dual chamber pacemakers or intracardiac cardioverter defibrillators, wherein an associated VA interval can replace the sensed atrial event in the controller timing cycles, comprising the steps of:
programming initial AV (atrioventricular) delay parameter of said algorithmic module;
learning to associate the VA interval with temporal patterns of hemodynamic sensor in correlation to the sensed atrial event;
switching to working according to the associated VA interval when convergence of learning occurs;
switching back to working according to the sensed atrial to event when failure occurs;
providing pacing in a non-adaptive mode whereby pacing is provided according to said pre-programmed AV delay;
switching to an adaptive mode wherein said AV delay changes dynamically in order to achieve optimal hemodynamic performance, and wherein said adaptive mode is limited to perform above a low limit of hemodynamic performance, and
switching back to the non adaptive mode whenever the hemodynamic performance drops below a low limit of hemodynamic performance or a sensor failure or any other system failure is detected.