Tracking the level of anesthesia-mediated unconsciousness

Abstract

Systems and methods for controlling anesthetic-mediated unconsciousness in a patient are disclosed. An electroencephalogram (EEG) of a patient under a GABAergic-anesthetic may be recorded using EEG electrodes. Unconsciousness maintained by GABAergic anesthetics, such as propofol, propanidid, and / or sevoflurane, is characterized by slow-delta oscillations (0.3-4 Hz) and alpha oscillations (8-14 Hz) that may be viewed in an electroencephalogram (EEG). An alpha modulation index and a slow modulation index may be computed to characterize the level of unconsciousness of the patient and set a desired level of unconsciousness. The system may automatically or manually adjust the infusion rate of the anesthetic to achieve the desired level of unconsciousness.

Description

Attorney Docket No. MIT-25121WO01 Tracking the Level of Anesthesia-Mediated Unconsciousness CROSS REFERENCE TO RELATED APPLICATION(S)

[0001] The present application claims the priority benefit, under 35 U.S.C. 119(e), of U.S. Application No.63 / 509,896, filed June 23, 2023, which is incorporated herein by reference in its entirety for all purposes. GOVERNMENT SUPPORT

[0002] This invention was made with government support under NS123120 and GM118269 awarded by the National Institutes of Health. The government has certain rights in the invention. BACKGROUND

[0003] Each anesthetic agent has a distinct neurophysiological signature that is readily visible in the electroencephalogram (EEG) and local field potential and relates directly to the anesthetic’s mechanism of action. For this reason, anesthetic EEG signatures can be used to monitor the level of unconsciousness in patients receiving general anesthesia for surgery. In adults, unconsciousness maintained by GABAergic anesthetics such as propofol, propanidid, and sevoflurane is primarily characterized by slow-delta oscillations (e.g., about 0.3–4 Hz) and alpha oscillations (e.g., about 8–14 Hz). A GABAergic or GABAnergic anesthetic is an anesthetic that interacts with the receptors of the neurotransmitter Gamma-Aminobutyric Acid (GABA). If the anesthetic dose is increased sufficiently, these oscillations may transform into burst-suppression, a state of profound brain inactivation during which quiescent or isoelectric (flatline) periods are interspersed between bursts of activity.

[0004] Experimental and modeling studies have shown that the slow-delta-alpha oscillatory state (SDA) is primarily a neurophysiological process. The alpha oscillations are a highly coherent thalamus-prefrontal cortex rhythm, whereas the slow-delta oscillations represent up-down states during which neural spiking activity is strongly down-regulated across large parts of the cortex. The presence of SDA oscillations is considered a marker of an adequate level of unconsciousness for surgery. Modeling studies and the clinical use of burst suppression suggest that this state is a neurometabolic phenomenon. The quiescent periods during burst suppression disrupt the SDAAttorney Docket No. MIT-25121WO01 oscillations when diminished levels of ATP make it difficult for neurons to maintain their membrane potentials. The presence of burst suppression is a more profound state of unconsciousness that has been associated with post-operative cognitive disorders, particularly in the elderly. This is why for anesthetic management of unconsciousness during surgery it is recommended to dose the anesthetics so as to avoid burst suppression. On the other hand, the state of burst suppression is often maintained intentionally for several days or more in intensive care unit patients placed in a medical coma to treat refractory status epilepticus or intracranial hypertension. SUMMARY

[0005] During unconsciousness maintained by GABAergic anesthetic agents (e.g., propofol, propanidid, and / or sevoflurane), the typical electroencephalogram (EEG) signatures are slow-delta oscillations (e.g., about 0.3–4 Hz) and alpha oscillations (e.g., about 8–14 Hz) that at higher doses devolve into burst suppression, a marker of profound brain inactivation. Alpha-wave-amplitude and slow-wave-frequency modulation processes continuously track the transition between these states in humans anesthetized with GABAergic anesthetic agents. Inventive systems and methods compute time-varying signals (termed Modulation Indices or MIs) from these modulation processes that track the level of unconsciousness of patients undergoing general anesthesia by processing information from the subject’s electroencephalograms (EEG) in real-time.

[0006] These systems and methods of monitoring and tracking the level of unconsciousness include a closed-loop anesthetic delivery (CLAD) system that automatically adjusts anesthetic administration rate to automatically control the level of unconsciousness under general anesthesia in the operating room (OR), an emergency room (ER), a hospital (e.g., for a general procedure and / or treatment), an out-patient setting, and / or level of sedation in the intensive care unit.

[0007] In some aspects, the techniques described herein relate to a method including sedating a subject with a GABAergic anesthetic while the subject is sedated with the GABAergic anesthetic, measuring an electroencephalogram (EEG) of the subject, filtering oscillatory signals out of the EEG from desired frequency bands, determining a state of anesthetic-mediated unconsciousness of the subject based on the oscillatory signals, and adjusting a dosage of the GABAergic anesthetic based on the state of anesthetic-mediated unconsciousness of the subject and a desired level ofAttorney Docket No. MIT-25121WO01 unconsciousness of the subject. If desired, the dosage of the GABAergic anesthetic can be adjusted manually or automatically.

[0008] In some aspects, the techniques described herein relate to a method wherein filtering oscillatory signals out of the EEG from desired frequency bands includes band-pass filtering the EEG between 0.3–4 Hz to obtain slow-delta oscillations, between 8–14 Hz to obtain alpha waves, and / or between 20–30 Hz to obtain beta waves. (Other frequency bands are also suitable, including, e.g., 0.1–2.0 Hz for slow-delta oscillations and 30–40 Hz for beta-gamma waves, or more generally between 0 Hz and 200 Hz.)

[0009] In some aspects, the techniques described herein relate to a method wherein filtering oscillatory signals out of the EEG from desired frequency bands includes extracting at least one signal from the EEG between X Hz and Y Hz, where X and Y are numbers between 0 and 200 with X < Y.

[0010] In some aspects, determining the state of anesthetic-mediated unconsciousness of the subject can include determining amplitude and / or frequency modulation for each of the oscillatory signals. Determining the state of anesthetic-mediated unconsciousness of the subject can also include computing a modulation index (MI) for each of the oscillatory signals.

[0011] In some aspects, the techniques described herein relate to a method wherein the adjusting a dosage of the GABAergic anesthetic includes manually adjusting the dosage of the GABAergic anesthetic by an anesthesia care giver.

[0012] In some aspects, the techniques described herein relate to a method wherein the adjusting a dosage of the GABAergic anesthetic includes automatically adjusting the dosage of the GABAergic anesthetic.

[0013] In some aspects, the techniques described herein relate to a method further including alerting an anesthesia care giver when the state of anesthetic-mediated unconsciousness of the subject is moving to or has moved beyond specified limits and providing an indication of a possible intervention to correct movement of the state of anesthetic-mediated unconsciousness of the subject.

[0014] In some aspects, the techniques described herein relate to a method wherein the GABAergic anesthetic is at least one of propofol, propanidid, or sevoflurane.Attorney Docket No. MIT-25121WO01

[0015] In some aspects, the techniques described herein relate to a method of monitoring an anesthetic state of a subject sedated with a GABAergic anesthetic, the method including while the subject is sedated with the GABAergic anesthetic, obtaining electroencephalography (EEG) data of the subject with electrodes in electrical communication with the subject, extracting slow-delta oscillations within a passband of 0.3–4 Hz from the EEG data, detecting a frequency modulation of the slow-delta oscillations, extracting alpha oscillations within a passband of 8–14 Hz from the EEG data, detecting an amplitude modulation of the alpha oscillations, extracting beta oscillations within a passband of 20–30 Hz from the EEG data, detecting an amplitude modulation of the beta oscillations, determining the anesthetic state of the subject based on the frequency modulation of the slow-delta oscillations, the amplitude modulation of the alpha oscillations, and the amplitude modulation of the beta oscillations, and adjusting a dosage of the GABAergic anesthetic based on the anesthetic state of the subject and a desired level of unconsciousness of the subject.

[0016] In some aspects, the techniques described herein relate to a method of monitoring an anesthetic state of a subject sedated with a GABAergic anesthetic, the method including while the subject is sedated with the GABAergic anesthetic, obtaining EEG data of the subject with electrodes in electrical communication with the subject, extracting at least one signal from the EEG data between X Hz and Y Hz, where X and Y are numbers between 0 and 200 with X < Y, detecting a modulation of the at the least one signal, determining the anesthetic state of the subject based on the modulation of the at least one signal, and adjusting a dosage of the GABAergic anesthetic based on the anesthetic state of the subject and a desired level of unconsciousness of the subject.

[0017] In some aspects, the techniques described herein relate to a system including an electroencephalogram (EEG) recording device to measure an EEG of a subject sedated with a GABAergic anesthetic, an infusion line connected to the subject to provide the GABAergic anesthetic to the subject, a monitor, operably connected to the electroencephalogram (EEG) recording device, to filter oscillatory signals out of the EEG from desired frequency bands, a controller, operably connected to the monitor to determine a state of anesthetic-mediated unconsciousness of the subject based on the oscillatory signals and to adjust a dosage of the GABAergic anesthetic based on the state of anesthetic-mediated unconsciousness of the subject and a desired level of unconsciousness of the subject, a graphical user interface (GUI), operably connected to the monitor and controller, to set the desired level of unconsciousness of the subject,Attorney Docket No. MIT-25121WO01 and an infusion pump, operably connected to the monitor, controller, and GUI, to pump the dosage of the GABAergic anesthetic into the subject through the infusion line.

[0018] In some aspects, the techniques described herein relate to a system wherein the electroencephalogram (EEG) recording device includes one or more electrodes in electrical communication with the subject.

[0019] In some aspects, the techniques described herein relate to a system wherein the desired level of unconsciousness of the subject may be set automatically.

[0020] In some aspects, the techniques described herein relate to a system wherein the desired level of unconsciousness of the subject may be set by an anesthesia care giver.

[0021] All combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are part of the inventive subject matter disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are part of the inventive subject matter disclosed herein. The terminology used herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein. BRIEF DESCRIPTIONS OF THE DRAWINGS

[0022] The skilled artisan will understand that the drawings primarily are for illustrative purposes and are not intended to limit the scope of the inventive subject matter described herein. The drawings are not necessarily to scale; in some instances, various aspects of the inventive subject matter disclosed herein may be shown exaggerated or enlarged in the drawings to facilitate an understanding of different features. In the drawings, like reference characters generally refer to like features (e.g., functionally and / or structurally similar elements).

[0023] FIG.1 illustrates a closed-loop anesthetic delivery (CLAD) system.

[0024] FIG. 2A illustrates an example session of the CLAD system of FIG. 1 for controlling anesthetic-mediated unconsciousness in a non-human primate. The top panel shows the slow-wave modulation index (SMI) of the subject (curved trace) being controlled to track precisely the target (horizontal lines, L1, L2, and L3) as the target changes levels, and the bottom panel shows the anesthetic infusion rate administered by the infusion pump.

[0025] FIG. 2B illustrates an example session of the CLAD system of FIG. 1 for controllingAttorney Docket No. MIT-25121WO01 anesthetic-mediated unconsciousness in a non-human primate. The top panel shows the alpha modulation index (AMI) of the subject (curved trace) being controlled to track precisely the target (horizontal lines, L1, L2, and L3) as the target changes levels, and the bottom panel shows the anesthetic infusion rate administered by the infusion pump.

[0026] FIG. 3A shows the evolution of EEG dynamics during propofol-mediated unconsciousness. The top panel of FIG. 3A is a graph showing the target propofol effect-site concentration for a human volunteer subject receiving a computer-controlled propofol infusion. The middle panel of FIG. 3A shows how the propofol infusion rate is increased in a stepwise manner to achieve five increasing target effect-site concentrations then the infusion is turned off. The bottom panel of FIG. 3A shows the temporal traces of the correct response probability for verbal cues and sound clicks to infer loss (LoC) and recovery (RoC) of consciousness.

[0027] FIG.3B shows EEG spectra for four-second raw EEG segments recorded at the timepoints (a, b, c, d, e, f, and g) indicated in the bottom panel of FIG. 3A showing amplitude modulation. The high-frequency variations are predominantly alpha oscillations (e.g., 8–14Hz), whereas the low-frequency variations are slow-delta oscillations (e.g., 0.3–4Hz). The timepoint a–g progression shows the transition during unconsciousness from slow-delta and alpha oscillations into burst suppression (at timepoint d) and back.

[0028] FIG.4A shows a raw EEG signal (left – bottom trace) recorded during propofol-mediated unconsciousness shown along with its alpha oscillation component (middle), its slow-delta oscillation component (right) and the sum of the two (left – top trace).

[0029] FIG. 4B shows EEG traces showing the modulation of the filtered alpha oscillations extracted at the timepoints (a, b, c, d, e, f, and g) indicated in FIG.3A. The alpha wave modulation continuously tracks the transition from SDA to burst suppression, and back to SDA during propofol-mediated unconsciousness. The first 4 seconds correspond to the traces in FIG.3B.

[0030] FIG.4C shows a trace illustrating alpha wave up-state durations (upper dashed line) and down-state durations (lower dashed line), and their computation through thresholding.

[0031] FIG.4D shows EEG traces similar to those in FIG.4B but applied to the filtered slow-delta oscillations.

[0032] FIG.4E is a trace illustrating slow wave cycle durations and down-state durations and theirAttorney Docket No. MIT-25121WO01 computation through crossings (lower dashed line) and thresholding (upper dashed line), respectively.

[0033] FIG.5A shows, in the top panel, the propofol target effect-site concentration (stair-stepped trace) and correct response probability curves for cues (other trace) as in FIG. 3A. The arrows indicate loss (LoC) and recovery (RoC) of consciousness. The bottom panel shows an EEG spectrogram.

[0034] FIG.5B shows time series plots showing the alpha up-state and down-state durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0035] FIG.5C shows a time course of the alpha modulation index (AMI).

[0036] FIG.5D shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0037] FIG.5E shows a time course of the slow modulation index (SMI).

[0038] FIG.5F shows a time course of the burst suppression probability (BSP).

[0039] FIG. 6A is a graph showing the propofol administration scheme (top panel), combining boluses and continuous infusion and an EEG spectrogram (bottom panel). The dark band is a period of missing EEG data.

[0040] FIG.6B shows time series plots showing the alpha up-state and down-state durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0041] FIG.6C shows a time course of the AMI.

[0042] FIG.6D shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0043] FIG.6E shows a time course of the SMI.

[0044] FIG.6F shows a time course of the BSP.

[0045] FIG.7A is a graph showing the manual propofol infusion scheme (top panel), with boluses and continuous infusion and an EEG spectrogram showing a loss of alpha power (arrow)Attorney Docket No. MIT-25121WO01 approximately 30 minutes after the start of the recording (bottom panel).

[0046] FIG.7B shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0047] FIG.7C is a time course of the AMI.

[0048] FIG.7D shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0049] FIG.7E is a time course of the SMI.

[0050] FIG.7F is a time course of the BSP.

[0051] FIG.8A is a graph showing the time course of the end-tidal sevoflurane concentration (top panel), following a propofol bolus administered for induction and a spectrogram of the raw EEG signal (bottom panel).

[0052] FIG.8B shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0053] FIG.8C is a time course of the AMI.

[0054] FIG.8D shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0055] FIG.8E is a time course of the SMI.

[0056] FIG.8F is a time course of the BSP.

[0057] FIG. 9 illustrates a biophysical model characterizing the progression from SDA to burst suppression. EEG traces band passed filtered in the alpha oscillation range and in the slow-delta oscillation range obtained by simulating the network model disclosed herein at increasing propofol target effect-site concentrations and neural metabolic effects. Darker bars above the traces on the left indicate alpha wave up-states and light bars indicate alpha wave down-states. A square above a trace on the right indicates the onset of a slow oscillation cycle, and a bar above a trace on theAttorney Docket No. MIT-25121WO01 right indicates a down-state. The changes in parameters to go from a to g are given in Table 2.

[0058] FIG.10A shows an EEG spectrogram.

[0059] FIG.10B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0060] FIG.10C shows time series plots showing the alpha up-state (bottom trace) and down- state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0061] FIG.10D is a time course of the AMI.

[0062] FIG.10E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0063] FIG.10F shows a time course of the SMI.

[0064] FIG.10G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.10A.

[0065] FIG.11A shows an EEG spectrogram.

[0066] FIG.11B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0067] FIG.11C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0068] FIG.11D shows a time course of the AMI.

[0069] FIG.11E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0070] FIG.11F shows a time course of the SMI.Attorney Docket No. MIT-25121WO01

[0071] FIG.11G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.11A.

[0072] FIG.12A shows an EEG spectrogram.

[0073] FIG.12B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0074] FIG.12C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0075] FIG.12D shows a time course of the AMI.

[0076] FIG.12E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0077] FIG.12F shows a time course of the SMI.

[0078] FIG.12G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.12A.

[0079] FIG.13A shows an EEG spectrogram.

[0080] FIG.13B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0081] FIG.13C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0082] FIG.13D shows a time course of the AMI.

[0083] FIG.13E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.Attorney Docket No. MIT-25121WO01

[0084] FIG.13F shows a time course of the SMI.

[0085] FIG.13G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.13A.

[0086] FIG.14A shows an EEG spectrogram.

[0087] FIG.14B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0088] FIG.14C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0089] FIG.14D shows a time course of the AMI.

[0090] FIG.14E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0091] FIG.14F shows a time course of the SMI.

[0092] FIG.14G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.14A.

[0093] FIG.15A shows an EEG spectrogram.

[0094] FIG.15B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0095] FIG.15C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0096] FIG.15D shows a time course of the AMI.

[0097] FIG.15E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are KalmanAttorney Docket No. MIT-25121WO01 filter estimates of the corresponding mean state durations.

[0098] FIG.15F shows a time course of the SMI.

[0099] FIG.15G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.15A.

[0100] FIG.16A shows an EEG spectrogram.

[0101] FIG.16B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0102] FIG.16C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0103] FIG.16D shows a time course of the AMI.

[0104] FIG.16E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0105] FIG.16F shows a time course of the SMI.

[0106] FIG.16G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.16A.

[0107] FIG.17A shows an EEG spectrogram.

[0108] FIG.17B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0109] FIG.17C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0110] FIG.17D shows a time course of the AMI.

[0111] FIG.17E shows time series plots showing the slow-oscillation cycle durationsAttorney Docket No. MIT-25121WO01 (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0112] FIG.17F shows a time course of the SMI.

[0113] FIG.17G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.17A.

[0114] FIG.18A shows an EEG spectrogram.

[0115] FIG.18B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0116] FIG.18C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0117] FIG.18D shows a time course of the AMI.

[0118] FIG.18E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0119] FIG.18F shows a time course of the SMI.

[0120] FIG.18G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.18A.

[0121] FIG.19A shows an EEG spectrogram.

[0122] FIG.19B shows propofol target effect-site concentration (middle step wise trace) and probability of correct response to verbal cues (inner vertical trace) and sound clicks (outer vertical trace).

[0123] FIG.19C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0124] FIG.19D shows a time course of the AMI.Attorney Docket No. MIT-25121WO01

[0125] FIG.19E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0126] FIG.19F shows a time course of the SMI.

[0127] FIG.19G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.19A.

[0128] FIG.20A show an EEG spectrogram.

[0129] FIG.20B is a graph showing the propofol administration scheme, combining boluses (narrow vertical bars) and continuous infusion (horizontal box).

[0130] FIG.20C show time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0131] FIG.20D shows a time course of the AMI.

[0132] FIG.20E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0133] FIG.20F shows a time course of the SMI.

[0134] FIG.20G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.20A.

[0135] FIG.21A shows an EEG spectrogram.

[0136] FIG.21B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0137] FIG.21C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0138] FIG.21D shows a time course of the AMI.

[0139] FIG.21E shows time series plots showing the slow-oscillation cycle durationsAttorney Docket No. MIT-25121WO01 (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0140] FIG.21F shows a time course of the SMI.

[0141] FIG.21G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.21A.

[0142] FIG.22A shows an EEG spectrogram.

[0143] FIG.22B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0144] FIG.22C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0145] FIG.22D shows a time course of the AMI.

[0146] FIG.22E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0147] FIG.22F shows a time course of the SMI.

[0148] FIG.22G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.22A.

[0149] FIG.23A shows an EEG spectrogram.

[0150] FIG.23B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0151] FIG.23C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0152] FIG.23D shows a time course of the AMI.

[0153] FIG.23E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces areAttorney Docket No. MIT-25121WO01 Kalman filter estimates of the corresponding mean state durations.

[0154] FIG.23F shows a time course of the SMI.

[0155] FIG.23G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.23A.

[0156] FIG.24A shows an EEG spectrogram.

[0157] FIG.24B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0158] FIG.24C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0159] FIG.24D shows a time course of the AMI.

[0160] FIG.24E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0161] FIG.24F shows a time course of the SMI.

[0162] FIG.24G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.24A.

[0163] FIG.25A shows an EEG spectrogram.

[0164] FIG.25B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0165] FIG.25C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0166] FIG.25D shows a time course of the AMI.

[0167] FIG.25E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.Attorney Docket No. MIT-25121WO01

[0168] FIG.25F shows a time course of the SMI.

[0169] FIG.25G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.25A.

[0170] FIG.26A shows an EEG spectrogram.

[0171] FIG.26B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0172] FIG.26C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0173] FIG.26D shows a time course of the AMI.

[0174] FIG.26E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0175] FIG.26F shows a time course of the SMI.

[0176] FIG.26G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.26A.

[0177] FIG.27A shows an EEG spectrogram.

[0178] FIG.27B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0179] FIG.27C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0180] FIG.27D shows a time course of the AMI.

[0181] FIG.27E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0182] FIG.27F shows a time course of the SMI.Attorney Docket No. MIT-25121WO01

[0183] FIG.27G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.27A.

[0184] FIG.28A shows an EEG spectrogram.

[0185] FIG.28B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0186] FIG.28C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0187] FIG.28D shows a time course of the AMI.

[0188] FIG.28E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0189] FIG.28F shows a time course of the SMI.

[0190] FIG.28G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.28A.

[0191] FIG.29A shows an EEG spectrogram.

[0192] FIG.29B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0193] FIG.29C show time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0194] FIG.29D shows a time course of the AMI.

[0195] FIG.29E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0196] FIG.29F shows a time course of the SMI.

[0197] FIG.29G shows a time course of the BSP. The time window used for calibrationAttorney Docket No. MIT-25121WO01 in AMI, SMI and BSP is provided above FIG.29A.

[0198] FIG.30A shows an EEG spectrogram.

[0199] FIG.30B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0200] FIG.30C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0201] FIG.30D shows a time course of the AMI.

[0202] FIG.30E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0203] FIG.30F shows a time course of the SMI.

[0204] FIG.30G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.30A.

[0205] FIG.31A shows an EEG spectrogram.

[0206] FIG.31B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0207] FIG.31C show time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0208] FIG.31D shows a time course of the AMI.

[0209] FIG.31E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0210] FIG.31F shows a time course of the SMI.

[0211] FIG.31G shows a time course of the BSP.

[0212] FIG.32A show an EEG spectrogram.Attorney Docket No. MIT-25121WO01

[0213] FIG.32B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0214] FIG.32C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0215] FIG.32D show a time course of the AMI.

[0216] FIG.32E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0217] FIG.32F shows a time course of the SMI.

[0218] FIG.32G shows a time course of the BSP.

[0219] FIG.33A shows an EEG spectrogram.

[0220] FIG.33B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0221] FIG.33C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0222] FIG.33D shows a time course of the AMI.

[0223] FIG.33E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0224] FIG.33F shows a time course of the SMI.

[0225] FIG.33G shows a time course of the BSP.

[0226] FIG.34A shows an EEG spectrogram.

[0227] FIG.34B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0228] FIG.34C shows time series plots showing the alpha up-state (lower trace) andAttorney Docket No. MIT-25121WO01 down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0229] FIG.34D shows a time course of the AMI.

[0230] FIG.34E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0231] FIG.34F shows a time course of the SMI.

[0232] FIG.34G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.34A.

[0233] FIG.35A show an EEG spectrogram.

[0234] FIG.35B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0235] FIG.35C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0236] FIG.35D shows a time course of the AMI.

[0237] FIG.35E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0238] FIG.35F shows a time course of the SMI.

[0239] FIG.35G shows a time course of the BSP.

[0240] FIG.36A shows an EEG spectrogram.

[0241] FIG.36B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0242] FIG.36C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.Attorney Docket No. MIT-25121WO01

[0243] FIG.36D shows a time course of the AMI.

[0244] FIG.36E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0245] FIG.36F shows a time course of the SMI.

[0246] FIG.36G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.36A.

[0247] FIG.37A shows an EEG spectrogram.

[0248] FIG.37B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0249] FIG.37C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0250] FIG.37D shows a time course of the AMI.

[0251] FIG.37E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0252] FIG.37F shows a time course of the SMI.

[0253] FIG.37G shows a time course of the BSP.

[0254] FIG.38A shows an EEG spectrogram.

[0255] FIG.38B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0256] FIG.38C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0257] FIG.38D shows a time course of the AMI.

[0258] FIG.38E shows time series plots showing the slow-oscillation cycle durationsAttorney Docket No. MIT-25121WO01 (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0259] FIG.38F shows a time course of the SMI.

[0260] FIG.38G shows a time course of the BSP.

[0261] FIG.39A shows an EEG spectrogram.

[0262] FIG.39B is a graph showing the propofol administration scheme, combining boluses (vertical bars) and continuous infusion (horizontal box).

[0263] FIG.39C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0264] FIG.39D shows a time course of the AMI.

[0265] FIG.39E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0266] FIG.39F shows a time course of the SMI.

[0267] FIG.39G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.39A.

[0268] FIG.40A shows an EEG spectrogram.

[0269] FIG.40B is a graph shows the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bars) administered for induction.

[0270] FIG.40C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0271] FIG.40D shows a time course of the AMI.

[0272] FIG.40E shows time series plots showing the slow-oscillation cycle durations (lower trace) and the slow-oscillation down-state durations (upper trace). The smooth traces areAttorney Docket No. MIT-25121WO01 Kalman filter estimates of the corresponding mean state durations.

[0273] FIG.40F shows a time course of the SMI.

[0274] FIG.40G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.40A.

[0275] FIG.41A shows an EEG spectrogram.

[0276] FIG.41B is a graph shows the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bars) administered for induction.

[0277] FIG.41C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0278] FIG.41D shows a time course of the AMI.

[0279] FIG.41E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0280] FIG.41F shows a time course of the SMI.

[0281] FIG.41G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.41A.

[0282] FIG.42A shows an EEG spectrogram.

[0283] FIG.42B is a graph shows the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bars) administered for induction.

[0284] FIG.42C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0285] FIG.42D shows a time course of the AMI.

[0286] FIG.42E shows time series plots showing the slow-oscillation cycle durationsAttorney Docket No. MIT-25121WO01 (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0287] FIG.42F shows a time course of the SMI.

[0288] FIG.42G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.42A.

[0289] FIG.43A shows an EEG spectrogram.

[0290] FIG.43B is a graph shows the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bars) administered for induction.

[0291] FIG.43C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0292] FIG.43D shows a time course of the AMI.

[0293] FIG.43E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0294] FIG.43F shows a time course of the SMI.

[0295] FIG.43G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.43A.

[0296] FIG.44A shows an EEG spectrogram.

[0297] FIG.44B is a graph shows the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bars) administered for induction.

[0298] FIG.44C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0299] FIG.44D shows a time course of the AMI.Attorney Docket No. MIT-25121WO01

[0300] FIG.44E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0301] FIG.44F shows time course of the SMI.

[0302] FIG.44G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.44A.

[0303] FIG.45A shows an EEG spectrogram.

[0304] FIG.45B is a graph shows the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (horizontal line) administered for induction.

[0305] FIG.45C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0306] FIG.45D shows a time course of the AMI.

[0307] FIG.45E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0308] FIG.45F shows a time course of the SMI.

[0309] FIG.45G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.45A.

[0310] FIG.46A shows an EEG spectrogram.

[0311] FIG.46B is a graph showing the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bar) administered for induction.

[0312] FIG.46C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.Attorney Docket No. MIT-25121WO01

[0313] FIG.46D shows a time course of the AMI.

[0314] FIG.46E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0315] FIG.46F shows a time course of the SMI.

[0316] FIG.46G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.46A.

[0317] FIG.47A shows an EEG spectrogram.

[0318] FIG.47B is a graph showing the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bar) administered for induction.

[0319] FIG.47C shows time series plots showing the alpha up-state (upper trace) and down-state (lower trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0320] FIG.47D shows a time course of the AMI.

[0321] FIG.47E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0322] FIG.47F shows a time course of the SMI.

[0323] FIG.47G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.47A.

[0324] FIG.48A shows an EEG spectrogram.

[0325] FIG.48B is a graph showing the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bar) administered for induction.

[0326] FIG.48C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of theAttorney Docket No. MIT-25121WO01 corresponding mean state durations.

[0327] FIG.48D shows a time course of the AMI.

[0328] FIG.48E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0329] FIG.48F shows a time course of the SMI.

[0330] FIG.48G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.48A.

[0331] FIG.49A shows an EEG spectrogram.

[0332] FIG.49B is a graph showing the time course of the end-tidal sevoflurane concentration (shaded region), following a propofol bolus (vertical bars) administered for induction.

[0333] FIG.49C shows time series plots showing the alpha up-state (lower trace) and down-state (upper trace) durations. The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0334] FIG.49D shows a time course of the AMI.

[0335] FIG.49E shows time series plots showing the slow-oscillation cycle durations (upper trace) and the slow-oscillation down-state durations (lower trace). The smooth traces are Kalman filter estimates of the corresponding mean state durations.

[0336] FIG.49F shows a time course of the SMI.

[0337] FIG.49G shows a time course of the BSP. The time window used for calibration in AMI, SMI and BSP is provided above FIG.49A.

[0338] FIG.50A shows box plots showing the statistics for AMI, SMI and BSP for each volunteer cases. The statistics are derived over the time interval where the marker is computed during the session, and each time step provides on data point. The horizontal bars correspond to the median.

[0339] FIG.50B shows box plots showing the statistics for AMI, SMI and BSP forAttorney Docket No. MIT-25121WO01 surgical cases maintained by propofol. The horizontal bars correspond to the median.

[0340] FIG.50C shows box plots showing the statistics for AMI, SMI and BSP for surgical cases maintained by propofol and showing weak alpha waves. The horizontal bars correspond to the median.

[0341] FIG.50D shows box plots showing the statistics for AMI, SMI and BSPfor surgical cases maintained by sevoflurane. The horizontal bars correspond to the median.

[0342] FIG.51 illustrates a schematic of the biophysical network model. The cortical network consists of excitatory pyramidal (PYR) neurons and inhibitory fast-spiking (FS) interneurons modeled with hodgkin-huxley-type dynamics. PYR neurons may synapse onto each other and onto FS neurons through AMPA-ergic projections. FS neurons may synapse onto each other and onto PYR neurons through GABAa-ergic projections. A global inhibitory self-loop represents the corticothalamic interaction and may mediate the neurophysiologic effect triggered by inhibitory facilitation as a function of GABAergic anesthetic effect site concentration. ATP- dependent potassium channels (KATP) are expressed on PYR neurons, and mediate the metabolic effect triggered by impaired ATP production as a function of GABAergic anesthetic effect site concentration.

[0343] FIG.52A in the top panel shows the propofol target effect-site concentration (inner step wise trace) and correct response probability curves for cues (outer trace) as in FIG. 3A. The arrows indicate loss (LoC) and recovery (RoC) of consciousness. The bottom panel shows an EEG spectrogram. This case is also presented in FIGS.5A–5F.

[0344] FIG.52B shows a time course of the alpha oscillation modulation index (AMI).

[0345] FIG.52C shows a time course of the SMI.

[0346] FIG.52D shows a time course of the BSP.

[0347] FIG.52E is a modulogram showing the coupling between the amplitude of the alpha wave (8–14Hz) and the phase of the slow wave (0.1–1Hz). The graph shows that the co- localization of high amplitude of the alpha wave with phase 0 (peak) of the slow wave increases as propofol effect site concentration increases. DETAILED DESCRIPTIONAttorney Docket No. MIT-25121WO01

[0348] Unconsciousness maintained by GABAergic anesthetics, such as propofol, propanidid, and / or sevoflurane, is characterized by slow-delta oscillations (0.3–4 Hz) and alpha oscillations (8–14 Hz) that are readily visible in an electroencephalogram (EEG). At higher doses, these slow-delta-alpha (SDA) oscillations transition into burst suppression. This is a marker of a state of profound brain inactivation during which isoelectric (flatline) periods alternate with periods of the SDA patterns present at lower doses. Brain activity evolves dynamically from SDA to burst suppression and back during unconsciousness maintained with propofol or sevoflurane in volunteer subjects and surgical patients. Two dynamic processes continuously modulate the SDA oscillations: alpha-wave amplitude modulation and slow-wave frequency modulation. An alpha modulation index and a slow modulation index characterize how these processes track the transition from SDA oscillations to burst suppression and back to SDA oscillations as a function of increasing and decreasing anesthetic doses, respectively. A biophysical model reveals that these dynamics track the combined evolution of the neurophysiological and metabolic effects of a GABAergic anesthetic on brain circuits. The modulatory dynamics mediated by GABAergic anesthetics can be used to monitor and precisely control the level of unconsciousness in patients under general anesthesia.

[0349] Modulation of the alpha oscillation amplitude by the phase of the slow oscillation can be a marker of anesthetic state. Moreover, alpha suppression periods and decreased alpha amplitude may predict the subsequent appearance of burst suppression. The lengths of the suppression periods of burst suppression can be characterized as a function of anesthetic dose and level of hypothermia, another process that can produce burst suppression. While SDA oscillations and burst suppression are produced by GABAergic anesthetics, the transition from the former to the latter has not been studied.

[0350] We found a modulatory process of brain activity that tracks continuously the transition from the SDA oscillations to burst suppression, and back to SDA oscillations. With this modulation process, the transition from SDA to burst suppression and back to SDA can be characterized as a function respectively of increasing and decreasing doses of a GABAergic anesthetic, such as propofol or sevoflurane, in healthy volunteers and surgical patients. The modulation may occur in parallel for the alpha and slow oscillations. Modulation indices (MIs) can be applied to the alpha frequency band (alpha modulation index: AMI) and the slow frequency band (slow modulation index: SMI) to track these transitions and can be used to determineAttorney Docket No. MIT-25121WO01 adjustments to the dosage level of the GABAergic anesthetic. These MIs can also be applied to other frequency bands, such as beta (e.g., about 20–30 Hz) and gamma (e.g., about >30 Hz) bands. Closed-Loop Anesthetic Delivery System

[0351] FIG.1 shows a closed-loop anesthetic delivery (CLAD) system 100 that determines and uses MIs for automatic and / or manual control of anesthetic delivery to a patient 101 during unconsciousness. The system 100 may include an EEG recording system 110, a monitor 120, a controller 130, an infusion pump 140, and a user interface, shown in FIG. 1 as a graphical user interface (GUI) 150. The EEG recording system 110, monitor 120, controller 130, infusion pump 140, and GUI 150 may be implemented in separate (purpose-built) devices, different combinations of devices, or a single device, such as an appropriately programmed computer. If implemented as separate devices or combinations of devices, the EEG recording system 110, monitor 120, controller 130, infusion pump 140, and / or GUI 150 may be connected through one or more wires and / or wireless connections (e.g., through the use of Wi-Fi connections, Bluetooth connections, cellular connections, satellite links, and / or a local area network). The monitor 120, controller 130, and GUI 150 may be combined into a single physical device (e.g., a computing device such as a computer). Alternatively, the monitor 120, controller 130, and GUI 150 may be contained in one or more separate devices (e.g., multiple computing devices, such as multiple computers).

[0352] The EEG recording system 110 may include one or more EEG electrodes 111 that are placed on the scalp of the patient 101 to measure EEG activity while the patient is under GABAergic-anesthetic mediated unconsciousness. The EEG electrodes 111 may be conductive electrodes in the form of single electrodes (e.g., a reference electrode, a ground electrode, and a measurement electrode), strips (e.g., six electrodes), and / or a cap (e.g., a whole head cap including at least 32 electrodes with a plurality of channels). Preferably there are at least three EEG electrodes 111 including a reference electrode, a ground electrode, and a measurement electrode, which is connected to the brain region of interest. The position of the EEG electrodes 111 on the patient 101 may be selected depending on the desired signal. For example, EEG electrodes 111 on the frontal area of the scalp of the patient 101 may record alpha and slow wave oscillations. Alternatively, the electrodes 111 on the rest of the head of the patient 101 may record slow-delta oscillations. The EEG electrodes 111 may be operably connected to the EEG recording system 110 through one or more wires 112. The EEG recording system 110 may also include a preamplifier 113 to amplify the EEG measured from the patient 101. If desired, one or more filtersAttorney Docket No. MIT-25121WO01 (not shown) can filter noise and out-of-band signals from the recorded EEG signals in the analog domain before pre-amplification. The EEG recording system 110 may also include a digitizer 114 (an analog to digital converter) to convert the analog EEG signals into digital EEG signals. The EEG recording system 110 may also include an EEG digital data transmitter 115 to transmit the digital EEG signals to another component in the system 100 (e.g., the EEG digital data receiver 121).

[0353] The EEG recording system 110 may be operably connected to the monitor 120 and controller 130 through a wired or wireless connection. For example, the EEG recording system 110 may be operably connected to a computing device (e.g., a computer) where the monitor 120 is deployed. The computing device may also deploy the controller 130 and GUI. The monitor 120 may include an EEG digital data receiver 121 to receive the digital EEG signals from the EEG digital data transmitter 115 of the EEG recording system 110. The EEG digital data receiver 121 and EEG digital data transmitter 115 may be connected through a wired and / or wireless medium (e.g., a wireless network). The monitor 120 may also include an EEG data processor 122 that may filter oscillatory signals out of the received digital EEG values from desired frequency bands. The monitor 120 may also include a modulation feature processor 123 that may compute features from the filtered oscillatory signals signal that quantify the modulation. For example, the modulation feature processor 123 may compute at least one of cycle durations (or frequency), up-state durations, and / or down-state durations for one or more oscillatory bands (e.g., slow-delta oscillations, alpha waves, and / or beta waves). The monitor 120 may also include a modulation index processor 124 that integrates the modulation features into modulation indices (MIs) (as described below) to indicate the level of anesthetic-state of unconsciousness of the patient 101.

[0354] The monitor 120 and controller 130 may be operably connected such that controller 130 may use output of the monitor 120. The controller 130 may include an estimation component 131 that estimates the responsiveness of the MIs (and the patient 101) to the anesthetic. The controller 130 may also include a prediction component 132 that predicts the value of the MIs into the future. The prediction component 132 may be able to predict the value of the MIs from about 5 seconds to about 30 minutes into the future. For example, the prediction component 132 may be able to predict the value of the MIs about 5 seconds, about 10 seconds, about 20 seconds, about 30 seconds, about 40 seconds, about 50 seconds, about 1 minute, about 5 minutes, about 10 minutes, about 15 minutes, about 20 minutes, about 25 minutes, or about 30 minutes into the future,Attorney Docket No. MIT-25121WO01 including all values in between. The controller 130 may also include a decision component 133 that decides on the infusion rate to bring the MI values (e.g., the value reflecting the patient’s 101 actual level of unconsciousness) as close as possible to the target MI value (e.g., the value reflecting the patient’s 101 desired level of unconsciousness) set using the GUI 150. These components can be implemented in the controller 130 as software stored in the controller’s memory and executed by the controller’s processor, which is coupled to the memory. The controller 130 may also include an infusion data transmitter 134 to transmit the infusion rate to an infusion data receiver 141 contained in the pump 140 through a wireless or wired connection.

[0355] The monitor 120 and controller 130 may be operably connected to the pump 140 through a wired or wireless connection. The pump 140 may include the infusion data receiver 141 that receives the infusion rate from the controller 130. The pump 140 may also include an infusion pump 142 and an infusion line 143. The infusion pump 142 pumps the anesthetic into the infusion line 143. The infusion line 143 is tubing through which the anesthetic is pumped through a catheter 145 into a vein of the patient 101. The vein may be in the patient’s arm 102 or in the back of the patient’s hand, for example. Alternatively, the vein may be in the patient’s foot or neck (e.g., the jugular vein). The pump 140 may also include a pump information transmitter 144 that may transmit information to the pump information receiver 156 about the pump 140 and the infusion rate. The monitor unit GUI display 154 may also display information received by the pump information receiver 156 (e.g., infusion rate, operational status of the pump 140, version of the pump 140, serial number of the pump 140, the anesthetic being administered by the pump 140, a time series of the infusion rate, and / or the volume of anesthetic administered by the pump 140). The pump information receiver 156 may also be operably connected to the controller 130 such that the controller 130 may use the information received by the pump information receiver 156.

[0356] The monitor 120 and controller 130 may also include a GUI / storage communication component 135 that bidirectionally communicates information, parameters, and user input with the GUI 150 from the monitor 120 and controller 130. The GUI / storage communication component 135 may also send information from the monitor 120 and controller 130 to be stored in the GUI 150.

[0357] The GUI 150 can be implemented in a display with keyboard, mouse, or other input device(s) or in a touchscreen or other suitable device. The hardware in which the GUI 150 is implemented may include a processor and memory for implementing different softwareAttorney Docket No. MIT-25121WO01 components as well as specialized chips and modules, including transmitters, receivers, and network interfaces. For instance, the specialized chips may include a monitor / control communication component 151 that bidirectionally communicates information, parameters, and user input with the monitor 120 and the controller 130.

[0358] The GUI 150 may also include a monitor user GUI input 152 that allows user to change monitor parameters and customize monitoring the level of unconsciousness of the patient 101. The monitor parameters may be customized to display any of the features computed from the data, including the raw EEG data. For example, a user may be able to change the monitor parameters for the thresholds for MI feature computations, one or more signal filtering variables, unprocessed EEG data (e.g., raw EEG data), filtered EEG data, the EEG spectrogram, the AMI, the SMI, and / or the data filtered in specific bands. The user may also be able to zoom in and / or out on a graph (e.g., a spectrogram).

[0359] The GUI 150 may also include a control user GUI input 153 that allows a user (e.g., an anesthesiologist) to change control parameters and customize controlling the level of unconsciousness of the patient 101. During operation of the system 100, the user may set the target level of unconsciousness of the patient 101 using the control user GUI input 153. The GUI 150 may also include a monitor unit GUI display 154 that may display information to the user for monitoring, including but not limited to EEG signals, filtered EEG signals, and / or modulation indices of the patient 101.

[0360] The GUI 150 may also include a control unit GUI display 155 that may display information for control including, but not limited to, infusion rates, infusion history, and / or infusion pump information. The infusion pump information may include, but is not limited to, the operational status of the pump 140, the version of the pump 140, a serial number of the pump 140, the anesthetic being administered by the pump 140, a time series of the infusion rate, and / or the volume of anesthetic that has been administered by the pump 140, for example. The GUI 150 may also include a pump information receiver 156 that may receive information about the pump 140 and the infusion rate of the pump 140.

[0361] The system 100 may also include monitor information storage 157 and control information storage 158. The monitor information storage 157 may store information from the monitor 120 including, but not limited to, EEG signals, filtered EEG signals, modulation features, and / or MI information. The control information storage 158 may store information from theAttorney Docket No. MIT-25121WO01 controller 130 including, but not limited to, estimation parameters, prediction values, control variables, and / or infusion rates. The control variables may include, but are not limited to, parameters from the responsiveness model, intermediate values of the computation for the estimation and prediction, the mode of control (e.g., manual or automatic), AMI, SMI, functions of both AMI and SMI, and / or power in specific frequency bands, for example.

[0362] The estimation component 131 may use the rate of change in the MI(s) and the past amount of anesthetic given to the patient 101 to derive how responsive the patient 101 may be to the anesthetic. The rate of change may be computed using an estimation procedure (such as the Kalman filter described below that may take as observations the first differences between MI measurements every second). This rate may decrease as the anesthetic clears the patient’s 101 system and may increase as the anesthetic accumulates in the patient’s 101 system. The estimation component 131 may approximate this relationship with a dynamical system where the past amount of anesthetic infused is the input and the rate of change in SMI is the output. The parameters of this dynamical system may specify the responsiveness of the patient 101 to the anesthetic.

[0363] To estimate these parameters, the estimation component 131 may apply estimation techniques on past data (e.g., on the order of seconds’ to minutes’ worth of past data) by combining the rates of changes in MI and the amounts of anesthetic infused per second (e.g., the infusion rate). The past data may include patient 101 EEG data obtained through the modulation index and / or past infusion data. The estimation component 131 may rely on about 10 seconds of past data to about 10 minutes of past data to determine how responsive the patient 101 may be to the anesthetic. For example, the estimation component 131 may rely on about 10 seconds, about 20 seconds, about 30 seconds, about 40 seconds, about 50 seconds, about 1 minute, about 2 minutes, about 3 minutes, about 4 minutes, about 5 minutes, about 6 minutes, about 7 minutes, about 8 minutes, about 9 minutes, or about 10 minutes of past data, including all values in between, Preferably the estimation component 131 relies on about 30 seconds to about 3 minutes of past data, more preferably about 1 minute of past data to determine how responsive the patient 101 may be to the anesthetic. These estimates for the parameters may reduce or minimize the error between the rate of change in MI observed directly and the rate of change in MI computed from anesthetic infusion information as model-output. With this approach, the estimation component 131 may obtain a responsiveness model that may be continually updated (e.g., on the order of seconds to minutes) to reflect the responsiveness of the MI(s) to anesthetic infusion.Attorney Docket No. MIT-25121WO01

[0364] The prediction component 132 may use the responsiveness model, the current MI(s), and the amount of anesthetic that will be administered to the patient 101 in the future to predict what the MI(s) may be in the future. By fixing a constant dose of anesthetic and using it as input to the responsiveness model, the prediction component 132 may compute the rate of change in MI(s) in the next few minutes (e.g., in about 1–10 minutes). By then integrating this rate of change starting from the current value of the MI(s), the prediction component 132 may compute the value of the MI(s) in the next few minutes (e.g., in about 1–5 minutes). As a result, by keeping the dose of anesthetic unchanged from its current level, the prediction component 132 can predict whether or not the MI will deviate from its target level. If the predicted MI in the few minutes (e.g., about 1–5 minutes) is lower (e.g., deeper) than the target level, then the anesthetic may have been over-dosed. Alternatively, if the predicted MI in the few minutes (e.g., about 1–5 minutes) is higher (e.g., lighter), then the anesthetic may have been under-dosed.

[0365] The decision component 133 component may use the prediction component 132 to decide on the amount of anesthetic to administer to the patient 101. Using another estimation procedure, decision component 133 may compute the dose of anesthetic that drives the MI as close as possible to the target MI in the next few minutes (e.g., about 1–5 minutes). This may be computed by finding the dose that reduces or minimizes the error over the next few minutes (e.g., about 1–5 minutes) between the target level and the predicted MI from the responsiveness model using the dose as input. This optimal dose may be provided on the monitor unit GUI display 154 as a recommendation for the anesthesiologist (e.g., in the case of manual control of anesthetic delivery) and / or provided to the infusion data transmitter 134 to be sent to pump 140 (e.g., in the case of automatic control of anesthetic delivery). For example, if the CLAD system 100 is designed as both a recommendation system and an automatic controller, the optimal dose may be provided to both the monitor unit GUI display 154 as a recommendation for the anesthesiologist and the infusion data transmitter 134 to be automatically sent to pump 140.

[0366] A new optimal dose of anesthetic, based on an updated responsiveness-model, may be continually recomputed (e.g., about every few seconds to about every few minutes) and provided either to the monitor unit GUI display 154 as a recommendation for the anesthesiologist or to the infusion data transmitter 134 to be sent to the pump 140.

[0367] In operation, the CLAD system 100 may be used for automatic or manual control of anesthetic delivery during unconsciousness. The CLAD system 100 may automatically controlAttorney Docket No. MIT-25121WO01 the rate of anesthetic delivery to maintain a desired level of unconsciousness in a patient 101 under anesthetically induced unconsciousness. Instead of, or in addition to, automatic control, the CLAD system 100 may allow for the manual control of the rate of anesthetic delivery to maintain a desired level of unconsciousness in a patient 101 under anesthetically induced unconsciousness, for example by providing recommendation(s) on the rate of anesthetic delivery to an anesthesiologist. The CLAD system 100 may allow for up to two hours of full control of anesthetic delivery during unconsciousness of a patient 101. For example, CLAD system 100 may allow for 30 minutes of control, 1 hour of control, 1.5 hours of control, and / or 2 hours of control, including all values in between.

[0368] The EEG of a patient 101 under GABAergic-anesthetic mediated unconsciousness may be recorded using EEG electrodes 111. The EEG data may be collected by the EEG recording system 110, which filters, amplifies, and digitizes the collected EEG data. The processed EEG data may then be sent to the monitor 120 and controller 130. The monitor 120 and controller 130 may then filter the EEG data in the desired frequency bands, compute the markers of unconsciousness (e.g., modulation indices, including AMI and SMI), and characterize the level of unconsciousness of the patient 101. The monitor 120 and controller 130 may also determine the difference between the target marker value (e.g., the desired level of unconsciousness) and the value of the actual marker (e.g., the actual level of unconsciousness of the patient 101). Additionally, based on the difference between the target marker value and the value of the actual marker, the monitor 120 and controller 130 may output to the infusion pump 140 instructions on how much to change the infusion rate of the anesthetic to maintain the value of the actual marker as close as possible to the target marker value. The monitor 120 and / or controller 130 may also send data and information to the GUI 150 to be displayed.

[0369] For example, the characterized level of unconsciousness may then be sent to the GUI 150, which shows a representation of the level of unconsciousness to an anesthesiologist. The anesthesiologist may assess the level of unconsciousness. In a system configured for manual control of anesthetic delivery, the anesthesiologist may manually adjust the administration rate of the infusion pump 140. In a system configured for automatic control of anesthetic delivery, the anesthesiologist may update the parameters of the controller 130 using the monitor unit GUI display 154 and set the desired level of unconsciousness. The controller 130 may then compute the amount of drug (e.g., anesthetic) needed to reach the desired level of unconsciousness and mayAttorney Docket No. MIT-25121WO01 adjust the administration rate of the infusion pump 140 automatically (e.g., to reach the desired level of unconsciousness). The infusion pump 140 administers the anesthetic drug to the patient 101. The system 100 can also alert an anesthesia care giver when the state of anesthetic-mediated unconsciousness of the subject is moving to or has moved beyond specified limits and provide an indication of a possible intervention to correct movement of the state of anesthetic-mediated unconsciousness of the patient 101.

[0370] For example, the CLAD system 100 may issue a warning if the patient 101 is becoming too unconscious (e.g., heading towards burst suppression) or at risk of waking up. The warning may be in the form of a sound or visual alert. For example, the warning may appear on the GUI display 154 for an anesthesiologist. The warning may also include a recommendation from the controller 130. For example, if the patient 101 is heading towards burst suppression the controller 130 may recommend decreasing the infusion rate of the pump 140 and / or halting the infusion of the anesthetic for a period of time to move the patient’s 101 brain state away from burst suppression. Alternatively, if the patient 101 is at risk of waking up the controller 130 may recommend increasing the infusion rate of the pump 140. An anesthesiologist may then adjust the desired level of unconsciousness using the monitor unit GUI display 154 so that the patient 101 is brought back to a desired level of unconsciousness.

[0371] FIGS. 2A and 2B show an example session of the CLAD system 100 of FIG. 1 demonstrating the automatic control of anesthetic-mediated unconsciousness in a non-human primate. FIG. 2A demonstrates the automatic control of anesthetic-mediated unconsciousness using the slow-wave modulation index (SMI) and FIG.2B demonstrates the automatic control of anesthetic-mediated unconsciousness using the alpha modulation index (AMI).

[0372] In the top panel of FIG.2A, a rhesus macaque (macaca mulatta) was administered the anesthetic propofol to induce and maintain unconsciousness. The target level of unconsciousness (horizontal lines, L1, L2, and L3) was set by selecting a target value for the SMI. An SMI of 0 is unconscious and an SMI of 100 is fully awake. Unconsciousness was induced by administering propofol intravenously at a constant rate (0.6 mL / min) for about 10 minutes. Unconsciousness was maintained using the CLAD system 100 of FIG. 1 (shaded boxes; SMI Adaptive Control) to change the propofol infusion rate every 4 seconds to keep the SMI (curved trace) as close as possible to the target levels (L1, L2 and L3). The target levels were input on the GUI 150 and changed approximately every 40 minutes. The SMI is shown from when the animalAttorney Docket No. MIT-25121WO01 closed its eyes (eyes closed) until when it opened them (eyes opened). The animal was allowed to recover consciousness by stopping infusion of propofol (end of shaded area). The bottom panel of FIG.2A shows the propofol infusion rate administered by the infusion pump 140. The propofol infusion rate may be automatically adjusted to achieve the desired level of unconsciousness (e.g., to match the desired SMI).

[0373] In the top panel of FIG.2B, a rhesus macaque (macaca mulatta) was administered the anesthetic propofol to induce and maintain unconsciousness. The target level of unconsciousness (horizontal lines, L1, L2, and L3) was set by selecting a target value for the AMI. An AMI of 0 is unconscious and an AMI of 100 is fully awake. Unconsciousness was induced by administering propofol intravenously at a constant rate (0.6 mL / min) for about 10 minutes. Unconsciousness was maintained using the CLAD system 100 of FIG. 1 (shaded boxes; AMI Adaptive Control) to change the propofol infusion rate every 4 seconds to keep the AMI (curved trace) as close as possible to the target levels (L1, L2 and L3). The target levels were input on the GUI 150 and changed approximately every 40 minutes. The AMI is shown from when the animal closed its eyes (eyes closed) until when it opened them (eyes opened). The animal was allowed to recover consciousness by stopping infusion of propofol (end of shaded area). The bottom panel of FIG.2B shows the propofol infusion rate administered by the infusion pump 140. The propofol infusion rate may be automatically adjusted to achieve the desired level of unconsciousness (e.g., to match the desired AMI). Results

[0374] The following sections provide examples of how the CLAD system 100 may be used to control anesthetic-mediated unconsciousness in a human.

[0375] Experimental data. EEG data from four different groups of patients illustrate the relationship between SDA and burst suppression was analyzed. Each group of patients received either propofol or sevoflurane as the primary anesthetic to maintain unconsciousness. All data were recorded under human studies protocols approved by the Massachusetts General Hospital Human Research Committee. These patients were: 10 young (ages 18–35 years old) healthy volunteers who received increasing followed by decreasing doses of propofol through computer-controlled infusions; 10 surgical patients (ages 24–82 years old) who received propofol by manual titration; 10 surgical patients who received propofol by manual titration whose EEG showed diminished or absent alpha waves (ages 50–90 years old); and 10 surgical patients (ages 53–72 years old) whoAttorney Docket No. MIT-25121WO01 received sevoflurane by manual titration. The EEG data were analyzed using standard multi-taper spectral analysis methods, bandpass filtering and state-space methods as described below. The results for all 40 subjects are given in FIGS. 10A–49G, and FIGS. 50A–50D show a statistical summary of the results.

[0376] FIGS. 4A–4E show alpha and slow-delta wave modulation continuously tracking the transition from SDA to burst suppression, and back to SDA during propofol-mediated unconsciousness.

[0377] FIGS.5A–5F show EEG dynamics of a volunteer subject under propofol-mediated unconsciousness with modulation progression from SDA to burst suppression and back to SDA.

[0378] FIGS. 6A–6F show EEG dynamics of a 44-year-old woman who underwent a laparoscopic cholecystectomy under propofol-mediated unconsciousness with modulation progression from SDA to burst suppression and back to SDA.

[0379] FIGS.7A–7F show EEG dynamics of a 70-year-old man who underwent a robotic laparoscopic prostatectomy under propofol-mediated unconsciousness show dissipation of alpha oscillations.

[0380] FIGS. 8A–8F show EEG dynamics of a 54-year-old man who underwent a photoselective vaporization of the prostate under sevoflurane-mediated unconsciousness.

[0381] FIGS.10A–10F show case 1: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0382] FIGS.11A–11F show case 2: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0383] FIG.12A–12F show case 3: EEG dynamics of a volunteer subject with propofol- mediated unconsciousness.

[0384] FIGS.13A–13F show case 4: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0385] FIG.14A–14F show case 5: EEG dynamics of a volunteer subject with propofol- mediated unconsciousness.

[0386] FIGS.15A–15F show case 6: EEG dynamics of a volunteer subject withAttorney Docket No. MIT-25121WO01 propofol-mediated unconsciousness.

[0387] FIGS.16A–16F show case 7: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0388] FIGS.17A–17F show case 8: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0389] FIGS.18A–18F show case 9: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0390] FIGS.19A–19F show case 10: EEG dynamics of a volunteer subject with propofol-mediated unconsciousness.

[0391] FIGS.20A–20F show case 11: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 44- year-old woman who underwent a laparoscopic cholecystectomy.

[0392] FIGS.21A–21F show case 12: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 24- year-old woman who underwent examination under anesthesia.

[0393] FIGS.22A–22F show case 13: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 37- year-old woman who underwent a left breast wire-localized lumpectomy with sentinel lymph node biopsy.

[0394] FIGS.23A–23F show case 14: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 34- year-old woman who underwent a living-donor kidney transplant and laparoscopic nephrectomy.

[0395] FIGS.24A–24F show case 15: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 47- year-old woman who underwent laparotomy and a complete cytoreduction for ovarian cancer.

[0396] FIGS.25A–25F show case 16: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 49- year-old woman who underwent sentinel lymph node mapping and biopsy, mastectomy nippleAttorney Docket No. MIT-25121WO01 sparing and placement of breast implant.

[0397] FIGS.26A–26F show case 17: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. He is a 66- year-old man who underwent a cystoscopy bladder biopsy.

[0398] FIGS.27A–27F show case 18: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 60- year-old woman who underwent Left breast lumpectomy and left axillary sentinel lymph node biopsy.

[0399] FIGS.28A–28F show case 19: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. He is a 69- year-old man who underwent ventral hernia repair with mesh, cholecystectomy, abdominoplasty and excision of heterotropic ossification.

[0400] FIGS.29A–29F show case 20: EEG dynamics with strong alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. He is a 82- year-old man who underwent laser enucleation of the prostate and urethral dilation.

[0401] FIGS.30A–30F show case 21: EEG dynamics with dissipation of alpha power in a surgical patient under general anesthesia with propofol- mediated unconsciousness. He is a 70- year-old man who underwent a pelvic lymph node dissection and a robot-assisted radical laparoscopic prostatectomy.

[0402] FIGS.31A–31F show case 22: EEG dynamics with weak alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. He is a 52- year-old man who underwent a photoselective vaporization of the prostate.

[0403] FIGS.32A–32F show case 23: EEG dynamics with weak alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. He is a 66- year-old man who underwent a prostate brachytherapy.

[0404] FIGS.33A–33F show case 24: EEG dynamics with weak alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. He is a 70- year-old man who underwent abdominal component separation, incisional hernia repair with mesh, and bilateral cholecystectomy.Attorney Docket No. MIT-25121WO01

[0405] FIGS.34A–34F show case 25: EEG dynamics with dissipation of alpha power in a surgical patient under general anesthesia with propofol- mediated unconsciousness. She is a 50- year-old woman who underwent a laparoscopic low anterior resection and ileostomy.

[0406] FIGS.35A–35F show case 26: EEG dynamics with weak alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 65- year-old woman who underwent a wire-localized lumpectomy in the right breast and right axillary tag-localized sentinel lymph node mapping and biopsy.

[0407] FIGS.36A–36F show case 27 EEG dynamics with dissipation of alpha power in a surgical patient under general anesthesia with propofol- mediated unconsciousness. He is a 70- year-old man who underwent cystoscopy and transurethral resection of a bladder tumor.

[0408] FIGS.37A–37F show case 28: EEG dynamics with weak alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 66- year-old woman who underwent a pelvic exam under anesthesia, robot-assisted laparoscopic sacrocolpopexy with synthetic mesh and cystoscopy.

[0409] FIGS.38A–38F show case 29: EEG dynamics with weak alpha power in a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is a 62- year-old woman who underwent a living-donor kidney transplant and a peritoneal dialysis catheter removal.

[0410] FIGS.39A–39F show case 30: EEG dynamics with absence of alpha power of a surgical patient under general anesthesia with propofol-mediated unconsciousness. She is an 89+ year-old woman who underwent exploratory laparotomy, colostomy takedown, colonoscopy, complex incisional hernia repair and peristomal hernia repair.

[0411] FIGS.40A–40F show case 31: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. He is a 58-year-old man who underwent a laparoscopic para-esophageal hernia repair.

[0412] FIGS.41A–41F show case 32: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. She is a 53-year-old woman who underwent a right breast lumpectomy.

[0413] FIGS.42A–42F show case 33: EEG dynamics of a surgical patient under generalAttorney Docket No. MIT-25121WO01 anesthesia with sevoflurane-mediated unconsciousness. She is a 56-year-old woman who underwent a transanal tumor excision.

[0414] FIGS.43A–43F show case 34: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. He is a 57-year-old man who underwent left laser lithotripsy, ureteroscopy, retrograde, stent and cystoscopy.

[0415] FIGS.44A–44F show case 35: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. She is a 65-year-old woman who underwent an open reduction internal fixation (ORIF) of left tibial plateau fracture.

[0416] FIGS.45A–45F show case 36: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. He is a 72-year-old man who underwent an open reduction internal fixation (ORIF) of right tibial plateau fracture.

[0417] FIGS.46A–46F show case 37: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. She is a 63-year-old woman who underwent a kidney transplant.

[0418] FIGS.47A–47F show case 38: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. She is a 67-year-old woman who underwent a total laparoscopic hysterectomy – bilateral salpingo-oophorectomy and pelvic lymph node dissection.

[0419] FIGS.48A–48F show case 39: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. He is a 54-year-old man who underwent a photoselective vaporization of the prostate.

[0420] FIGS.49A–49F show case 40: EEG dynamics of a surgical patient under general anesthesia with sevoflurane-mediated unconsciousness. He is a 66-year-old man who underwent a photoselective vaporization of the prostate.

[0421] FIGS.50A–50D show a summary of AMI, SMI and BSP performance.

[0422] FIGS.52A–52E show the progression of phase-amplitude coupling in a volunteer subject under propofol-mediated unconsciousness.

[0423] The analysis of the modulations during the transition from SDA to burstAttorney Docket No. MIT-25121WO01 suppression and back was restricted to the periods where the alpha and slow oscillations were present. For the volunteer subjects, this period began following loss of consciousness when the alpha oscillations appeared and lasted until the start of emergence when the alpha oscillations transitioned to beta oscillations. For the surgical patients, this period began following induction when the alpha or slow oscillations appeared and lasted until either anesthetic administration ended or the alpha oscillations transitioned into beta oscillations and / or the slow-delta oscillations dissipated, whichever event occurred first. The periods not analyzed at the start and end of a session are marked in gray in these figures.

[0424] Transition from alpha oscillations to burst suppression and back in a healthy volunteer administered propofol. With changes in the propofol target effect-site concentration (FIG. 3A, upper panel), it may be possible to track changes in behavioral response (FIG. 3A, middle panel) and corresponding changes in the spectral content of the EEG (FIG. 3A, lower panel). These data show a modulation of the alpha oscillation amplitude in which the raw EEG transitions from having high-amplitude alpha oscillations to low-amplitude alpha oscillations. This modulation is present as soon as the alpha oscillations appear shortly after loss of consciousness (FIG. 3B at timepoints a and b). With increasing propofol target effect-site concentration, the periods of high-amplitude alpha oscillations decrease, and the periods of low-amplitude alpha oscillations increase (FIG.3B at timepoint c) until the raw EEG signal becomes burst suppression (FIG.3B at timepoint d). As the target propofol effect-site concentration declines (FIG.3A), the periods of low-amplitude alpha oscillations shorten, and the periods of high-amplitude alpha oscillations lengthen (FIG. 3B at timepoints e–g). Hence, in these young healthy volunteers receiving propofol, the alpha amplitude modulation tracks the transition from the appearance of alpha oscillations to burst suppression and back to alpha oscillations as a function of increasing and decreasing propofol doses.

[0425] The alpha amplitude modulation is more apparent when the EEG is filtered to isolate the alpha band (FIG.4A). Defining the periods of high-amplitude alpha oscillations as up- states and the periods of low-amplitude alpha oscillations as down-states enhances the presentation of the modulation in FIGS.4B and 4C (See below for a complete definition of the alpha up-states and the alpha down-states). Filtering the data in FIGS.3A and 3B to isolate the alpha band reveals that as the propofol target effect-site concentration increases, the up-state durations decrease, and the down-state durations increase (FIG. 4B at timepoints a–d). As the propofol target effect-siteAttorney Docket No. MIT-25121WO01 concentration decreases, the dynamics reverse (FIG.4B at timepoints e–f). As illustrated with the raw EEG signal, burst suppression (FIG. 4B at timepoint d) is a point on the alpha amplitude modulation continuum during which alpha down-states are appreciably longer than the up-states. The alpha down-states correspond to the suppression periods.

[0426] Without being bound to any particular theory, our biophysical model, described below, may explain the changes in these dynamics of the alpha oscillations.

[0427] Transition from slow-delta oscillations to burst suppression and back in a healthy volunteer administered propofol. In addition to alpha-wave modulation, the data shows the presence of a slow-delta oscillation (e.g., about 0.3–4 Hz) frequency modulation which is readily visible in the raw EEG recording (FIG.3B) and in the slow-delta band extracted from the raw EEG signal by bandpass filtering (FIGS.4A, 4D, and 4E). The modulation is apparent when focusing on the duration of the slow cycles and the duration of their down periods (FIG.4E). (See below for definitions of the slow cycle and slow down-state durations.). At low propofol target effect-site concentrations, isolated large amplitude (up and down) cycles begin to appear in the slow waves (FIG.4D). Their durations are distinctly larger than the initial slow wave fluctuations. These cycles initially appear sparsely in the slow-wave fluctuations. As the propofol target effect- site concentration increases, they occur more frequently (FIGS. 4D at timepoints a–d). The occurrence of these large cycles increases until they coalesce: their up-states transition to the burst periods of burst suppression, whereas their troughs or down-states transition to the suppression periods of burst suppression (FIGS.4D and 4E). The progression of these dynamics also reverses (FIG.4D at timepoints e–g), as the propofol target effect-site concentration decreases.

[0428] As with alpha wave modulation, our biophysical model may explain the dynamics of slow-wave modulation.

[0429] Alpha- and slow-wave modulation track propofol effect site concentration during the transition from SDA to burst suppression, and back. An alpha modulation index (AMI) quantitatively tracks the EEG dynamics in the alpha band during the transition from SDA to burst suppression and back to SDA. AMI is a real-time measure derived from a ratio between the duration of the alpha up-states and down-states (FIGS.4B and 4C). We tested the AMI on the EEG recordings from all 10 volunteer subjects who underwent propofol-mediated loss and recovery of consciousness. The example in FIGS.5A–5F shows that the duration of an alpha up-Attorney Docket No. MIT-25121WO01 state decreases as the propofol target effect-site concentration increases, up to a point, after which the duration of an alpha down-state begins to increase (FIGS.5A and 5B). FIGS.5A and 5C and Table 1 (below) show that AMI mirrored the propofol target effect-site concentration.

[0430] Table 1: Pearson correlation between AMI / SMI and the target effect site concentration for each volunteer subject. Case 1 2 3 4 5 6 7 8 9 10

[0431] A slow modulation index (SMI) tracks the dynamics of the slow-wave frequency modulation during the transition from SDA to burst suppression and back to SDA. SMI is a real- time measure derived from a ratio between the frequency of slow wave cycles and duration of the slow wave down-state (FIGS.5B and 5C). FIG.5D shows that the number of slow-wave cycles per unit time decreased as the propofol targets effect-site concentration increased, up to a point where the slow wave down-states became more prominent and began to lengthen. Like AMI (FIGS. 5A and 5C), SMI also mirrored the propofol effect-site concentration (FIGS.5A and 5E and Table 1).

[0432] Computing the burst suppression probability (BSP), which tracks across time the instantaneous probability that a patient is in burst suppression, validates the AMI and SMI. FIGS. 5C, 5E, and 5F show that both AMI and SMI reflect the information present in BSP which is designed to track high effect-site concentrations. That is primarily because the suppressions that the BSP detects coincide with the down-states in the alpha waves and down-states in the slow waves. AMI and SMI track modulatory dynamics that are present at lower propofol target effect- site concentrations. Here, BSP is uninformative (FIGS.5C, 5E, and 5F).

[0433] Five of the 10 subjects in our volunteer cohort (Cases 1, 3, 6, 7 and 9; see FIGS. 10A–10G, 12A–12G, 15A–15G, 16A–16G, and 18A–18G) underwent a transition from SDA to burst suppression and back to SDA. AMI and SMI tracked these transitions. The remaining five volunteer subjects (Cases 2, 4, 5, 8 and 10; see FIGS.11A–11G, 13A–13G, 14A–14G, 17A–17G, and 19A–19G) stayed in SDA. This demonstrates that AMI and SMI may track not only theAttorney Docket No. MIT-25121WO01 transitions between SDA and burst suppression, but also the dynamics within the SDA stages of propofol-mediated unconsciousness. AMI and SMI track different oscillatory bands and are computed differently (FIGS. 4C and 4E). However, because both indices track the transitions between SDA and burst suppression (FIG.50A), this may suggest that the alpha- and slow-wave modulations may be driven by a common process. Our biophysical model illustrates that this common process may be due to the combined neurophysiological and metabolic effects of propofol.

[0434] Alpha- and slow-wave modulation are present in surgical patients under propofol-mediated unconsciousness. Anesthetic administration to the volunteer subjects used only propofol targeted to achieve specific effect-site concentrations in order to track loss and recovery of consciousness. In surgical cases, the situation is different as multiple agents (e.g., hypnotics, analgesics, and muscle relaxants) are administered simultaneously to achieve the state of general anesthesia and not simply unconsciousness (FIGS. 6A–6F). Nevertheless, the EEG dynamics are governed predominantly by the primary hypnotic agent (e.g., the GABAergic anesthetic). FIGS. 6A–6E show that the alpha- and slow-wave modulation dynamics are also present in surgical cases during which propofol is the primary hypnotic agent. In these cases, AMI and SMI tracked the transition from SDA to burst suppression and back (FIGS.6B–6E). AMI and SMI track the modulatory dynamics identified by the BSP when burst suppression is present as well as the alpha-wave amplitude and slow-wave frequency modulation present at propofol doses lower than those at which burst suppression occurs (FIG.6F).

[0435] Four of the 10 patients in our surgical cohort (Cases 11, 15, 18 and 20; see FIGS. 20A–20G, 24A–24G, 27A–27G, and 29A–29G) underwent the transition from SDA to burst suppression and back to SDA during surgery. Both AMI and SMI tracked these transitions. The other 6 patients (Cases 12–14, 16, 17 and 19; see FIGS.21A–23G, 25A–25G, 26A–26G, and 28A– 28G) remained in SDA and both the AMI and SMI tracked their dynamics as well.

[0436] Slow-wave modulation is present in surgical patients with weakened, absent, or dissipating alpha waves under propofol-mediated unconsciousness. While a signature of propofol- and sevoflurane-mediated unconsciousness is the presence of strong alpha waves, these oscillations are often weak, dissipating or absent in elderly patients. For the elderly, this limits the utility of an index that tracks the transition from SDA to burst suppression by monitoring alphaAttorney Docket No. MIT-25121WO01 wave modulation. In this case, the SMI remains a reliable metric for tracking modulatory dynamics during propofol-mediated unconsciousness.

[0437] FIGS. 7A–7F shows the EEG recorded from a 70-year-old man undergoing a robotic laparoscopic prostatectomy. At approximately 40 minutes after induction of general anesthesia, his alpha oscillations dissipate (FIG. 7A). This led to detection of very long alpha down-states (FIG. 7B) and a drastic decrease in the AMI (FIG. 7C). These changes in the AMI suggest an early increase in the level of unconsciousness. However, the progression of the slow- wave activity is preserved (FIG. 7D) and SMI tracked the progression from SDA to burst suppression and back as well (FIG. 7E). If AMI is recalibrated after the loss of alpha power— which is possible by this implementation—the alpha waves should still offer a suitable tracking index, afterwards, at their weakened level. BSP, that detects broadband suppression, indicates that there is no burst suppression in the first half of the surgical session despite the weakening of alpha oscillations (FIG.7F).

[0438] The EEG recordings of 10 surgical patients in whom unconsciousness during general anesthesia was maintained using propofol who showed either weak, dissipating, or absent alpha oscillations was analyzed. Five of these patients (Cases 22–24, 27 and 29; see FIGS.31A– 33G, 36A–36G, and 38A–38G) remained in SDA without going into burst suppression, as indicated by their raw EEG traces and their near zero BSP values. In these cases, a sudden weakening or dissipation of the alpha oscillations with concomitant AMI drops at those timepoints was often observed suggesting a more profound level of unconsciousness consistent with burst suppression, while none is present. However, SMI remains at levels comparable to those prior to alpha dissipation. The remaining 5 patients (Cases 21, 25, 26, 28 and 30; see FIGS. 30A–30G, 34A–34G, 35A–35G, 37A–37G, and 39A–39G) show light periods of burst suppression, preceded by alpha oscillation weakening and an early decrease of AMI relative to SMI. An overall decrease in AMI compared to SMI when alpha waves are weak may be observed when comparing FIG.50C to FIG.50B.

[0439] Alpha- and slow-wave modulation are present in surgical patients under sevoflurane-mediated unconsciousness. The modulatory dynamics observed for propofol should be present for other GABAergic anesthetics, as all the GABAergic anesthetics have similar spectral patterns. For confirmation, EEG recordings of 10 surgical patients receiving generalAttorney Docket No. MIT-25121WO01 anesthesia in whom unconsciousness was maintained with sevoflurane, another GABAergic anesthetic, was analyzed (FIGS.8A–8F). Six of these patients (Cases 31–33, 37, 39 and 40; see FIGS. 40A–42G, 46A–46G, 48A–48G, and 49A–49G) transitioned from SDA into burst suppression and back to SDA. AMI and SMI tracked these transitions. In two of these six cases (Cases 31 and 40; see FIGS. 40A–40G and 49A–49G), the appearance of burst suppression coincides with the administration of propofol boluses indicating a sudden increase of GABAergic anesthetic effect-site concentration. In the other cases (Cases 32, 33, 37 and 39; see FIGS.41A– 41G, 42A–42G, 46A–46G, and 48A–48G), the transition into burst suppression occurred on a high sevoflurane dose, and hence, a high effect-site concentration. This shows that this transition may also occur when sevoflurane is the primary hypnotic. The remaining 4 patients (Cases 34-36 and 38; see FIGS. 43A–45G and 47A–47G) remained in SDA, and AMI and SMI tracked these dynamics.

[0440] The progression of alpha- and slow-wave modulation signals a transition from a neurophysiological to a metabolic anesthetic effect in the brain. Building on existing biophysical models for alpha oscillations, slow-delta oscillations, and burst suppression, a neural circuit description of how alpha amplitude and slow-wave frequency modulation that may define the transition from SDA to burst suppression as a function of increasing effect site concentration of a GABAergic anesthetic. This neural circuit model is a cortical network that includes excitatory pyramidal neurons and inhibitory interneurons (FIG.51). Although directly combining the existing biophysical models may produce SDA and burst suppression, this configuration alone may be insufficient to produce alpha-wave amplitude and slow-wave frequency modulation since the existing biophysical models are not equipped with slow-wave generation mechanisms that are able to generate frequency modulation of a slow wave. The duration of a slow wave cycle changes throughout the transition from SDA to burst suppression. If the slow oscillation is dynamically constructed by an interaction of different GABAergic-anesthetic-dependent mechanisms, particularly during SDA, then a modulation process may emerge from the augmented model.

[0441] For the purpose of presentation, the formation of the modulation processes occurs in three stages. In the first stage, the effect-site concentration of the GABAergic anesthetic may reach a level sufficient to produce alpha and slow-delta oscillations (FIG. 9 at timepoint a). Its widespread actions may also facilitate global inhibition of cortical activity, triggered by a momentary increase in cortical activity and enabled potentially through cortical-thalamicAttorney Docket No. MIT-25121WO01 interactions. The global inhibition may hyperpolarize the membranes of the excitatory neurons which may manifest as increasingly larger troughs in the slow oscillations and interruption of alpha oscillations by interrupting neuronal spiking activity. The first signs of alpha and slow-wave modulation may appear in the EEG (FIG.9 at timepoint a).

[0442] In the second stage, as the effect-site concentration of the GABAergic anesthetic continues to increase, global inhibition is further facilitated. As it is triggered more frequently, large slow-wave cycles and alpha oscillation disruption may become more frequent. These effects may appear as a more profound modulation with increasing slow-wave cycle duration and shorter alpha oscillation up-states (FIG.9 at timepoints b–d). At the same time, ATP production may start to decline as the GABAergic anesthetic begins to disrupt oxidative phosphorylation in the mitochondrial membranes of the neurons (FIG.9 at timepoints c and d).

[0443] In the third stage, ATP production may be severely impaired so that neuronal spiking activity readily depletes ATP levels thereby forcing inward rectifier ATP-dependent potassium channels to open (FIG. 51). The opening of these channels may hyperpolarize the neurons and enhance the suppression periods observed in both the slow and alpha oscillations. As the GABAergic anesthetic effect-site concentration increases further, the recovery of adequate ATP levels may slow, leading to longer suppression periods, which may manifest as longer alpha and slow-wave down-states (FIG.9 at timepoints d–g), and eventually, burst-suppression (FIG.9 at timepoints f–g). The biophysical model output (FIG.9) may capture the dynamics present in the EEG signals (FIGS.4A–4E).

[0444] The data and modeling analyses shows that burst suppression may be the extreme of alpha amplitude and slow-frequency modulation (FIG.3B at timepoint d, FIG.3D at timepoint d and FIG.9 at timepoints f–g). However, impaired mitochondrial function (the metabolic effect) may not begin with large doses of the GABAergic anesthetic that are often associated with burst suppression. The modeling analysis disclosed herein suggests that the metabolic effect may be present when the alpha-wave down-states and the slow-wave troughs begin to lengthen (FIG.9 at timepoints d–e). These events may occur well in advance of the appearance of burst suppression in the EEG. As ATP production becomes more impaired, the metabolic effect may manifest prominently as the transition into burst suppression.

[0445] Our cortical network may produce SDA when there is sufficient binding of theAttorney Docket No. MIT-25121WO01 GABAergic anesthetic to the GABA receptors. FIG.9 shows that as the effect-site concentration increases, the slow-delta and alpha oscillations may undergo frequency and amplitude modulation, respectively, which may coalesce into burst suppression due to the combined neurophysiologic and metabolic effects of the anesthetic. Discussion

[0446] Alpha-wave and slow-wave modulation may continuously track the transition from SDA oscillations to burst suppression and back to SDA oscillations. EEG SDA oscillations and burst suppression are established markers of unconsciousness mediated by GABAergic anesthetics. Burst suppression may indicate a more profound state of brain inactivation. We discovered that alpha-wave amplitude modulation and slow-wave frequency modulation may track the transition from SDA oscillations to burst suppression and the transition from burst suppression to SDA oscillations as a function of an increasing and decreasing GABAergic anesthetic (e.g., propofol, propanidid, and / or sevoflurane) effect site concentrations respectively.

[0447] In our up-down state analysis of the alpha and slow modulations, we discovered two components for both processes. The transition from SDA to burst suppression may be marked by shortening of the alpha up-states followed by lengthening of the alpha down-states. During the same transition, the wide slow cycles may first become more frequent. Next the troughs of these cycles may lengthen and flatten into long suppression intervals. Our biophysical model suggests that the first component of the two modulatory processes may be a neurophysiological effect whereas the second component may be a metabolic effect.

[0448] Two indices, AMI and SMI, track the alpha and slow modulations, respectively, during the SDA-burst suppression-SDA transitions. The indices tracked these dynamics in human volunteers receiving propofol and in actual surgical patients receiving either propofol or sevoflurane as the GABAergic hypnotic agent during general anesthesia. The SMI tracked the changes in dynamics even in the absence or loss of the alpha oscillations, states that commonly occur in elderly patients. Because both indices track the SDA-burst suppression-SDA transition we inferred that both track changes in level of unconsciousness. Our biophysical model supports this inference as it shows that the modulatory processes which these indices track may be generated by the combined neurophysiological and metabolic effects of GABAergic anesthetics on neuralAttorney Docket No. MIT-25121WO01 circuits. A caveat is that none of the volunteers or patients received ketamine, a non-GABAergic anesthetic known to alter the EEG signatures of GABAergic anesthetics.

[0449] Signatures of alpha- and slow-waves during propofol-mediated unconsciousness. Alpha suppression during the early stages of propofol-mediated unconsciousness may be a predictor of the subsequent appearance of burst suppression. Our data analysis and biophysical modeling suggest that low amplitude alpha oscillations and early evidence of alpha suppression may be due to an earlier onset of the metabolic effects of the GABAergic anesthetic.

[0450] Modulation of the alpha oscillation by the slow oscillation has been reported in two forms for propofol: trough-max and peak-max. In the former, identified as a marker of both loss and recovery of consciousness, the amplitude of the alpha oscillation is maximal at the trough of the slow oscillation. In the latter, identified as a marker of unconsciousness, the amplitude of the alpha oscillation is maximal at the peak of the slow oscillation. That is, during peak-max the alpha up-states are reported to co-localize with the up-cycle in the slow oscillation. FIGS.5A and 52B shows this co-localization of propofol during the immediate pre-burst suppression and burst suppression states. However, in the transition to burst suppression, the slow oscillation troughs changed with increasing anesthetic dose and did not always signify a down-state for the alpha oscillations (FIG.9).

[0451] Our biophysical modeling shows that the slow oscillation shape and period may be formed dynamically as a function of increasing GABAergic anesthetic dose, due to its combined neurophysiological and metabolic effects (FIG. 9). The metabolic effects may lengthen both the alpha and the slow oscillation down-states. The two down-states may eventually overlap and devolve into suppression periods. Our model suggests that suppression of spiking activity may accompany the suppression periods. Indeed, in non-human primate studies of propofol-mediated unconsciousness, spiking activity was suppressed during the suppression periods in the local field potentials.

[0452] Mechanisms for the combined neurophysiological and metabolic effects of GABAergic anesthetics. Our data analyses and modeling link EEG alpha- and slow-wave modulation to neurophysiological and metabolic effects of GABAergic anesthetics. The presence of the slow, delta, and alpha oscillations may represent principally a neurophysiological effect,Attorney Docket No. MIT-25121WO01 whereas progression into the alpha and slow modulations may represent the neurophysiological effect and the progression of the metabolic effect. Burst suppression is the extreme presence of the modulatory effect. The modulations suggest that the metabolic effects of the GABAergic anesthetics may begin well in advance of burst suppression.

[0453] There is substantial experimental evidence to support possible mechanisms for the metabolic effects. Positron emission tomography studies in humans show that administering propofol decreases CMRO2 and cerebral blood flow. Findings from in vitro studies show that GABAergic anesthetics slow ATP production by abolishing mitochondrial membrane potentials, and thereby, blocking the conversion of ADP to ATP. In addition, experiments have shown that GABAergic anesthetics act on mitochondrial respiratory enzymes leading to a decrease in oxygen consumption and inducing a switch to glycolysis.

[0454] Implications of AMI and SMI for monitoring unconsciousness and brain metabolic state. AMI and SMI may be computed and displayed in real time as measures of unconsciousness for a patient in whom a GABAergic anesthetic is the primary agent maintaining unconsciousness. These indices contain the information in the BSP as a special case. Hence, this suggests that the AMI and the SMI may be used clinically to monitor unconscious during general anesthesia. As real time markers of unconsciousness, they may also be used to implement the system 100 for closed-loop control of anesthetic state in any situation where a patient is under general anesthesia or sedation. For example, in an operating room (OR), emergency room (ER), an intensive care unit, a hospital (e.g., for a general procedure and / or treatment), and / or an out- patient setting. These applications of AMI and SMI may be used to help maintain general anesthesia or sedation. Preferably, ketamine is not co-administered with the methods disclosed herein.

[0455] The metabolic hypothesis proposed by our biophysical model may be tested by measuring simultaneously neurophysiological responses and metabolic responses during controlled upward and downward titrations of propofol or another GABAergic anesthetic. The relationship between the indices, the neurophysiological, and the metabolic responses, may also be used to track the metabolic state of a patient in an operating room, an emergency room, an intensive care unit, a hospital (e.g., for a general procedure and / or treatment), and / or an out-patient setting. A more accurate characterization of brain state may lead to more judicious anestheticAttorney Docket No. MIT-25121WO01 dosing and a possible reduction in post-operative cognitive disorders. It may also enable more precise anesthetic titration to maintain a desired level of neuroprotection for patients placed in a medical coma to treat status epilepticus or to control intracranial hypertension. Our characterization of the modulatory dynamics mediated by GABAergic anesthetics during transitions between states of unconsciousness offers useful mechanistic insights that can be applied in clinical care. Materials and Methods Data Acquisition

[0456] All data collection and experimental protocols were approved by the Mass General Brigham Human Research Committee (Institutional Review Board). For the propofol volunteer study, all subjects provided informed consent. For the surgical patient studies, there was no data collection specific consent as the EEG recordings, physiological data and anesthetic administration data were collected as part of standard care and de-identified.

[0457] Volunteer subjects under propofol-mediated unconsciousness. The propofol volunteer dataset includes EEG recordings from 10 healthy volunteers between the ages 18–36, American Society of Anesthesiology Physical Status I, and with Mallampati Class I airway anatomy.

[0458] Propofol was administered via computer-controlled infusion to achieve target effect-site concentrations of 0, 1, 2, 3, 4, and 5 ^^g / mL, based on a three-compartment pharmacokinetic model of propofol. Each target concentration was held for 14 minutes. Clearance and volume parameters of the model were adjusted to each volunteer based on their age, sex, height, and weight. The dynamics of other compartments were assumed to be unaffected by the addition of an effect site compartment. In some cases, target effect-site concentrations were decreased in a stepwise fashion, so that both induction and emergence from unconsciousness were gradual. Unconsciousness was determined by the response of a subject to auditory stimuli (clicks or words), which were presented every four seconds. Unresponsiveness to the auditory stimuli was interpreted as unconsciousness.

[0459] Whole head EEG data were recorded using a 64-channel BrainVision MRI Plus system (Brain Products) with a sampling rate of 5,000 Hz, bandwidth 0.016–1000 Hz, andAttorney Docket No. MIT-25121WO01 resolution 0.5 ^^V least significant bit. The Fp1 channel was used for all further analysis. Subjects were instructed to close their eyes throughout the experiment to avoid eye-blink artifacts in the EEG.

[0460] Surgical patients under propofol-mediated unconsciousness, with strong alpha waves. This dataset was created from a database of real-time EEG recordings of 140 patients who underwent general anesthesia or monitored anesthesia care between August 1, 2020, and March 1, 2022. Frontal EEG data were recorded using a SedLine brain function monitor (Masimo Corporation, Irvine, CA, USA) with a sampling frequency of 178 Hz. The Fp2 channel was used for further analysis. A research assistant annotated accurate time points of all drug changes and all operating room (OR) events, including the patient’s arrival, EEG recording start and end times, surgical events, induction, intubation, and extubation.

[0461] Ten cases were selected to be included in this dataset if they met the following inclusion criteria: (i) the patient underwent general anesthesia; (ii) the recording did not exhibit excessive muscle artifacts or electrical impedance from electrocautery; (iii) the recording did not exhibit data quality issues characteristic of SedLine; (iv) the primary hypnotic agent was propofol; and (v) the EEG exhibited high alpha band power during unconsciousness. For criterion (iii), there tend to be three problems in EEG recordings from the SedLine monitors: (a) an undocumented change in the sample rate of the recorded EEG due to a change in display feed; (b) an undocumented change in the amplitude and quantization of the recorded EEG; and / or (c) clipping or stair-step-like distortion of the EEG signal that is dependent on the position of the signal on the screen. To address (a) and (b), we ensured that the display feed settings did not change during data collection to prevent undocumented changes in the sample rate, amplitude, and quantization. To address (c), we visually inspected the recording to ensure selection of cases without clipping or stair-step-like distortion of the EEG signal.

[0462] Surgical patients under propofol-mediated unconsciousness, with weakened, absent or dissipating alpha waves. This dataset was created in the same way as the dataset for surgical patients under propofol-mediated unconsciousness showing strong alpha waves, with the exception of the last criterion which was substituted with: (v) the EEG exhibited low alpha band power, dissipating alpha oscillations, and / or absence of alpha oscillations during unconsciousness.

[0463] Surgical patients under sevoflurane-mediated unconsciousness. This datasetAttorney Docket No. MIT-25121WO01 was created from a database of real-time EEG recordings of 247 patients who underwent general anesthesia or monitored anesthesia care between November 1, 2011, and August 20, 2015. Clinical information including approximate times of drug changes and events such as induction, intubation, and extubation were collected from the Epic electronic medical record system. Frontal EEG data were recorded using the SedLine brain function monitor (Masimo Corporation, Irvine, CA, USA) with a pre-amplifier bandwidth of about 0.5–92 Hz, sampling rate of about 250 Hz, and with 16- bit, 29 nV resolution. The Fp2 channel was used for further analysis.

[0464] Ten cases were selected to be included in this dataset if they met the following inclusion criteria: (i) the patient underwent general anesthesia; (ii) the recording did not exhibit excessive muscle artifacts or electrical impedance from electrocautery; (iii) the primary hypnotic agent was sevoflurane; and (iv) the EEG exhibited high alpha band power during unconsciousness. Filtering and Signal Processing

[0465] EEG polarity was adjusted so that the initial burst of activity during burst suppression corresponded to positive deflections. All EEG spectrograms for visualization were computed using the multi-tapered method. The alpha and slow oscillations were obtained from the raw EEG by band-pass filtering between about 8–14 Hz (alpha) and about 0.3–4 Hz (slow-delta) respectively, using a 2nd order Butterworth filter. In operation, the CLAD system 100 may use the following equations to compute the modulation indices. For example, the modulation feature processor 123 may estimate the modulation features from the EEG data acquired by the EEG recording system 110 using the below equations.

[0466] A state-space Gaussian Kalman filter was implemented to derive estimates of the durations utilized in AMI and SMI, in an online manner for real-time computation where ^^ and ^^ denote, respectively, the observation and its corresponding latent state that needs to be estimated. The state and observation equations were defined as provided in Equation 1 below: ^^^^+1^^^^+ ^^^^^^ ^^where ^^ ^^ is the EEG sampling period, and ^^^^∼ ^^(0, ^^^2^)and ^^^^∼ ^^(0, ^^^2^)are independently and identically drawn (iid) from^^denote the state atAttorney Docket No. MIT-25121WO01 time ^^ and its corresponding variance, respectively, estimated from the observations ^^0, ⋯ , ^^^^. The one-step Kalman prediction equations become Equation 2 below: ^^^^∨ ^^−1^^^^−1∨ ^^−1^^^2^∨ ^^−1^^^2^−1∨ ^^−1+ ^^ ^^2^^2^^. (2)

[0467] The^^^2 ^^^^=^∨ ^^−1^^^2^∨ ^^−1+ ^^^2^. (3) and the Kalman^^^^∨ ^^^^^^∨ ^^−1+ ^^^^( ^^^^− ^^^^∨ ^^−1) 22. (4) ^^^^∨ ^^− ^^^^^^^^∨ ^^−1

[0468] by running the Kalmanfilter on the reversed signal, namely ^^[ ^^ − ^^] where ^^ is the length of ^^. The initial conditions can instead be determined, with no changes to the resulting filtered traces, by computing the mean of ^^^^and the variance of ^^^^− ^^^^−1in a time window around the initial time.

[0469] In these equations, the Kalman gain ^^^^may quickly converge to ^^ satisfying (1 − ^^) ^^ ^^2^^^2^= ^^2^^^2^, and the interpretation of the Kalman filtering may coincide with applying a first-order low pass filter with cut-off frequency of ^^ ^^^^⁄ 2 ^^ where ^^^^denotes the sampling frequency. The cut-off frequency was adjusted to coincide with a period of 5 minutes ( ^^ = 300 ^^ ^^ ^^ ^^ ^^ ^^ ^^), to focus on the variations at that temporal scale. For AMI related computation, we set ^^^^and ^^^^to be any pair that satisfies(1 − ^^)^^ ^^2^^^2^= ^^2^^^2^with ^^ = 2 ^^⁄ (^^ ^^^^). In this setting, ^^ coincides with the steady state Kalman gain. For SMI related computations, we scale the observation noise variance by a factor of 10. In general, changing ^^ (via ^^) controls the smoothness of the filtered curves.

[0470] For the volunteer and sevoflurane subjects (1⁄^^ ^^ = 250Hz) we fixed ^^^^= 1 and ^^^^= 0.0209 for AMI, and ^^^^= 10 and ^^^^= 0.0209 for SMI. For the propofol surgical subjects (1⁄^^ ^^ = 178Hz) we fixed ^^^^= 1 and ^^^^= 0.0207 for AMI and ^^^^= 10 and ^^^^= 0.0207 for SMI. These parameters were fixed to be the same for all subjects in their respective category, depending on EEG sampling frequency.

[0471] An alternative approach would be to determine ^^^^and ^^^2^using an EM algorithm with prior distributions ^^^^and ^^^2^.Attorney Docket No. MIT-25121WO01

[0472] Both AMI and SMI use the same calibration 300-second window to determine the modulation thresholds. See the following subsections on AMI and SMI computation for more details. AMI computation

[0473] The alpha oscillation ^^ ^^ ^^^^ ^^ ^^ℎ ^^was obtained by band-pass filtering between about 8–14 Hz using a 2ndorder Butterworth filter. The goal was to fix a modulation threshold ^^ and determine when the upper envelope of ^^ ^^ ^^^^ ^^ ^^ℎ ^^is above or below ^^, indicating a period of up- state or down-state, respectively.

[0474] For a fixed modulation threshold, the up / down states were determined as follows. All the peaks of the absolute value signal | ^^ ^^ ^^^^ ^^ ^^ℎ ^^| above ^^ were determined, and any time point within 50 ms (half the period of a 10 Hz oscillation) of such a peak was considered to be in an alpha up-state. The remaining time points corresponded to an alpha down state. A peak in | ^^ ^^ ^^^^ ^^ ^^ℎ ^^| was detected at time point ^^ if | ^^ ^^ ^^^^ ^^ ^^ℎ ^^| at time points ^^ − 1 and ^^ + 1 had values lower than that at time point ^^.

[0475] For a calibration window of 300 seconds, the modulation threshold ^^ was determined as the 50th percentile of the amplitudes of all the peaks of ^^ ^^ ^^^^ ^^ ^^ℎ ^^within that calibration window. This choice was intended to roughly yield up- and down-states of similar durations during the calibration period. The peaks in ^^ ^^ ^^^^ ^^ ^^ℎ ^^in the calibration window were similarly detected as done for | ^^ ^^ ^^^^ ^^ ^^ℎ ^^|.

[0476] The duration of all up-states and down-states were computed. The computed durations appear in panel B, as scatter plots, of FIGS. 5–8 and 10–49. A signal for the up-state duration was formed by setting the value between the start of an up-state and the start of a next one (following a down-state) to be the duration of that first up-state. A similar down-state duration signal was derived, considering down- instead of up-states.

[0477] Real-time versions of the duration signals, for use as real-time observations in the AMI computation, that may not depend on future information were derived as follows. The observed duration of an alpha up-state at time ^^ may correspond to the maximum between the duration of the current up-state until time ^^ (if at time ^^, alpha oscillations are in an up-state) and the duration of the last alpha up-state. If at time ^^, alpha oscillations are in a down state, then theAttorney Docket No. MIT-25121WO01 observed duration is the duration of the last up-state. The signal was defined similarly for down- states. These signals were then passed through a Kalman filter for estimating the latent state duration, to yield two duration signals ^^^^ ^^and ^^^^ ^^ ^^ ^^, which also appear in panel B of the relevant figures.

[0478] An EEG signature was derived as ^^^^ ^^ ^^ℎ ^^= log( ^^^^ ^^⁄ ^^^^ ^^ ^^ ^^), which tracks the evolution of the duration of up- and down-states, and the by applying a logisticfunction on ^^^^ ^^ ^^ℎ ^^, specifically, 1⁄ [1 + exp(− ^^ ^^^^ ^^ ^^ℎ ^^)] , 5 below: ^^ ^^ ^^ =^^^^ ^^+ ^^^^ ^^ ^^ ^^. (5)

[0479] The how quickly AMI saturatesnear 0 or 1. We set ^^ = 0.5 for AMI throughout the study. This value can be modified as desired depending on the situation. Modifying ^^ may modify the dynamic range of the marker. The value of ^^ may be equal to 0.5 to 3, including all values in between. For example, ^^ may be equal to 0.5, 1, or 2. The value of ^^ may be increased (e.g., above 0.5) when ^^^^ ^^and ^^^^ ^^ ^^ ^^are expected to be close in value to each other. The Kalman filtering assumes that the noise term ^^^^= ^^^^− ^^^^may be independently and identically distributed. In filtering for alpha up-state durations, the observation ^^^^is fixed during an alpha down-state, yielding correlated observations. However, the state variable of the up-state duration can still vary. Indeed, anesthetic effect site concentration may continue to increase or decrease during a down state, altering the length of the next up-state which is not yet observable. Therefore, throughout the assumed process, ^^^^− ^^^^may not be correlated by design, and therefore the independent and identically distributed (iid) assumption may be satisfied.

[0480] While the evolution for alpha up- and down-states may be obtained separately through Kalman filtering, after which AMI is computed, it may also be possible to combine these observations into one estimation framework. Indeed, AMI may result from applying a logistic function onto the signature ^^^^= log( ^^^^ ^^⁄ ^^^^ ^^ ^^ ^^). It may then be possible to derive corresponding linear state and observation equations from: ^^^^= log( ^^^^ ^^)– ^^ ^^ ^^(^^^^ ^^ ^^ ^^), upon which to apply Kalman filtering, and directly estimate AMI.SMI computationAttorney Docket No. MIT-25121WO01

[0481] The slow oscillation ^^ ^^ ^^^^ ^^ ^^ ^^was obtained by band-pass filtering between about 0.3–4 Hz using a 2ndorder Butterworth filter. The goal was to fix two modulation thresholds ^^^^(crossing) and ^^^^(quiescence), with ^^^^< ^^^^and determine the duration of a slow oscillation cycle in ^^ ^^ ^^^^ ^^ ^^ ^^each detected by crossings at ^^^^and the duration of the down-states, being the periods of ^^ ^^ ^^^^ ^^ ^^ ^^below ^^^^.

[0482] To determine the duration of a slow oscillation cycles in ^^ ^^ ^^^^ ^^ ^^ ^^, the time points where ^^ ^^ ^^^^ ^^ ^^ ^^crosses ^^^^were determined and ^^ ^^ ^^^^ ^^ ^^ ^^was segmented into cycles delimited by three crossings: a first crossing from below ^^^^that indicated the start of a cycle, a mid crossing from above ^^^^, and a final crossing from below ^^^^that indicated the end of a cycle. Each cycle then consisted of a positive deflection followed by a negative deflection. The duration of all cycles was determined, and two signals, one for the cycle duration and one for the frequency of cycles, were derived by setting the value at a time point equal to the duration (or frequency) of the cycle it belongs to. The computed cycle durations are shown in a scatter plot in panel D of FIGS.5–8 and 10–49.

[0483] Real-time versions of the cycle duration and frequency signal, for use as observation signal in the SMI computation, that do not depend on future information were derived as follows. The observed duration of a slow cycle at time ^^ may correspond to the maximum between the duration of the current slow cycle until time ^^ and the duration of the last slow cycle. The signal may be defined similarly for the frequency of cycles observation. These signals for the cycle duration and cycle frequency were passed through a Kalman filter to yield two signals ^^^^ ^^ ^^ ^^and ^^^^ ^^ ^^ ^^. The signal ^^^^ ^^ ^^ ^^appears in panel D of FIGS.5–8 and panel E of FIGS.10–49.

[0484] To determine the duration of the down-states in ^^ ^^ ^^^^ ^^ ^^ ^^, time points with values below ^^^^may be considered to belong to a down-state. The different duration of the down-states were determined, and a down-state duration signal was formed by setting the value at the time point to be the duration of its corresponding down-state if it was during one, or the duration of the previous down-state if it does not belong to one. Real-time observation signal may be derived as performed for the cycle duration and frequency above, then Kalman filtered to yield a signal ^^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^, which appears in panel D of FIGS.5–8 and panel E of FIGS.10–49.

[0485] For a calibration window of 300 seconds, ^^^^was determined by a parameterAttorney Docket No. MIT-25121WO01 sweep, choosing the threshold that maximizes the number of crossings in that window. Note that ^^^^may have a value near 0 given the bandpass filtering between about 0.3–4 Hz to obtain the slow wave. To determine ^^^^, we found a threshold ^^′ > ^^^^in that calibration window such that the frequency of crossings is reduced by 30%, and then lowered that threshold slightly, e.g., by defining ^^^^= 0.5 ^^^^+ 0.5 ^^′. The suppression periods during burst suppression may fluctuate around ^^^^, and the definition of ^^^^may ensure that the suppression periods are detected by raising the threshold ^^^^to one that reduces crossing, and then lowering it again to ensure that it does not erroneously detect low amplitude non-suppression slow cycles. Other threshold calibration schemes, or detection methods, may also be possible.

[0486] An EEG signature was derived as given in Equation 6 below: ^^^^^^ ^^ ^^ ^^⁄ ^´^^^ ^^ ^^ ^^^^ ^^ ^^ ^^= log (^^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^⁄ ^´^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^), (6) where ^^^^ ^^ ^^ ^^of the crossing frequency and down-state duration, respectively. The signature tracks the evolution of the slow oscillation dynamics, and the marker may be derived by applying a logistic function on ^^^^ ^^ ^^ ^^, specifically, 1⁄ [1 + exp(− ^^ ^^^^ ^^ ^^ ^^)], which yields Equation 7: ^^ ^^ ^^ ^^ =( ^^^^ ^^ ^^ ^^⁄ ^´^^^ ^^ ^^ ^^)^^^^. (7)

[0487] range of SMI, and how quickly SMI saturates near 0 or 1. We set ^^ = 2 for SMI throughout the study. This value can be modified as needed depending on the situation. Modifying ^^ may modify the dynamic range of the marker. The value of ^^ may be equal to 2 to 5, including all values in between. For example, ^^ may be equal to 2, 3, 4, or 5. The value of ^^ may be increased (e.g., above 2) when the values of ^^ and ^^ are close to the baseline value. Similar to what was suggested for AMI computation, it may also be possible to combine the two observations for SMI into one estimation framework. Biophysical Modeling

[0488] The biophysical model consisted of interconnected 80 pyramidal neurons (PYR) and 20 fast spiking interneurons (FS). The neurons may be modeled using a single compartment with Hodgkin-Huxley-type dynamics. The voltage change ^^ ^^⁄ ^^ ^^ in each cell with membrane capacitance ^^^^is described by Equation 8:Attorney Docket No. MIT-25121WO01 ^^ ^^ ^^^^^^ ^^ = −∑ ^^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^− ∑ ^^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^+ ^^^^ ^^ ^^+ ^^^^ ^^ ^^ ^^ ^^. (8) The cells current ( ^^^^) for membrane^^ ^^ ^^ ^^ ^^ ^^ ^^. current that captures the metabolic effect and a global inhibitory current that captures inhibitory facilitation of the corticothalamic system. The synaptic currents ( ^^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^) depend on the connectivity.

[0489] Adenosine triophosphate (ATP)-gated potassium current. This current is adapted from and encapsulates the metabolic effect on neurons. It is defined as: 1 ^^KATP= ^^KATP^^(^^ − ^^^^)with ^^ = 1+10[ATP] (9) and is governed[N˙a]= ^^ ^^Na − 3 ^^ ^^[Na]3[ATP](10) [AT˙P] = ^^ATP([ATP]^^ ^^ ^^− [ATP]) − ^^^^[Na]3[ATP]. (11) The value of ^^^^ ^^ ^^may govern the rate of ATP production, and may be decreased as propofol concentration increased, as a proxy for metabolic impairment. Low levels of ATP may lead to opening up of the KATPchannels which hyperpolarizes the cell. The cell is then in a state of suppression until ATP levels replenish. That duration lengthens as the production rate is impaired.

[0490] Global thalamocortical inhibition current. This current encapsulates the neurophysiological effect on neurons during the modulation, and is defined as: ^^^^ ^^ ^^ ^^ ^^ ^^= ^^^^ ^^ ^^ ^^ ^^ ^^^^ (12) with ^^^^ ^^ ^^ ^^ ^^ ^^= 0.1 ^^A ⋅ ^^ ^^−2and 1 ^˙^ = −0.001 ^^ +^^− ^^^^ ^^ ^^ ^^ ^^. (13) Thefor the aggregate activity in the network, that will trigger the global inhibition. The threshold ^^^^ ^^ ^^ ^^ ^^ ^^was decreased as propofol concentration increased. Synchronous high activity in cortex triggers a surge of inhibition or dis- excitation through the corticothalamic system, that may lead to a momentary shutdown of activity. A full corticothalamic biophysical implementation of the global inhibition current can be achieved by incorporating a thalamic model of thalamocortical (TC) and reticular (RE) cells. In particular,Attorney Docket No. MIT-25121WO01 depolarization of TC cells can switch the thalamus out of bursting mode to yield a momentary down-state in thalamic activity. This depolarization can be triggered by a synchronous surge of cortical activity.

[0491] Aggregate activity, simulations and analysis. The aggregate population activity, from which spectral information was determined, included the sum of the membrane potentials of the PYR cells. Our simulations only altered the parameters ^^ATPand ^^^^ ^^ ^^ ^^ ^^ ^^to represent different effect site concentrations. The exact values are given below in Table 2.

[0492] Table 2: Simulation parameters for FIGS.8A–8F. EEG Trace a b c d e f g[ ] ur networ mo e was programme n ++ an comp e us ng gcc. The differential equations were integrated using a fourth-order Runge-Kutta algorithm, with integration time step of about 0.05 milliseconds. The model output was analyzed using Python 3. Conclusion

[0494] While various inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and / or structures for performing the function and / or obtaining the results and / or one or more of the advantages described herein, and each of such variations and / or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and / or configurations will depend upon the specific application or applications for which the inventive teachings is / are used. Those skilled in the art will recognize or be able to ascertain, using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and / orAttorney Docket No. MIT-25121WO01 method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and / or methods, if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

[0495] Also, various inventive concepts may be embodied as one or more methods, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[0496] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.

[0497] The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

[0498] The phrase “and / or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and / or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and / or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[0499] As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of” or “exactly one of,” or, whenAttorney Docket No. MIT-25121WO01 used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e., “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

[0500] As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and / or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[0501] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of” and “consisting essentially of” shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.

Claims

Attorney Docket No. MIT-25121WO01 CLAIMS 1. A method comprising: sedating a subject with a GABAergic anesthetic; while the subject is sedated with the GABAergic anesthetic, measuring an electroencephalogram (EEG) of the subject; filtering oscillatory signals out of the EEG from desired frequency bands; determining a state of anesthetic-mediated unconsciousness of the subject based on the oscillatory signals; and adjusting a dosage of the GABAergic anesthetic based on the state of anesthetic-mediated unconsciousness of the subject and a desired level of unconsciousness of the subject.

2. The method of claim 1, wherein filtering oscillatory signals out of the EEG from desired frequency bands comprises band-pass filtering the EEG between 0.3–4 Hz to obtain slow-delta oscillations, between 8–14 Hz to obtain alpha waves, and / or between 20–30 Hz to obtain beta waves.

3. The method of claim 1, wherein filtering oscillatory signals out of the EEG from desired frequency bands comprises extracting at least one signal from the EEG between X Hz and Y Hz, where X and Y are numbers between 0 and 200 with X < Y.

4. The method of claim 1, wherein determining the state of anesthetic-mediated unconsciousness of the subject comprises determining amplitude and / or frequency modulation for each of the oscillatory signals.

5. The method of claim 1, wherein determining the state of anesthetic-mediated unconsciousness of the subject comprises computing a modulation index (MI) for each of the oscillatory signals.

6. The method of claim 1, wherein the adjusting a dosage of the GABAergic anesthetic comprises manually adjusting the dosage of the GABAergic anesthetic by an anesthesia care giver.

7. The method of claim 1, wherein the adjusting a dosage of the GABAergic anestheticAttorney Docket No. MIT-25121WO01 comprises automatically adjusting the dosage of the GABAergic anesthetic.

8. The method of claim 1, further comprising: alerting an anesthesia care giver when the state of anesthetic-mediated unconsciousness of the subject is moving to or has moved beyond specified limits; and providing an indication of a possible intervention to correct movement of the state of anesthetic-mediated unconsciousness of the subject.

9. The method of claim 1, wherein the GABAergic anesthetic is at least one of propofol, propanidid, or sevoflurane.

10. A method of monitoring an anesthetic state of a subject sedated with a GABAergic anesthetic, the method comprising: while the subject is sedated with the GABAergic anesthetic, obtaining electroencephalography (EEG) data of the subject with electrodes in electrical communication with the subject; extracting slow-delta oscillations within a passband of 0.3–4 Hz from the EEG data; detecting a frequency modulation of the slow-delta oscillations; extracting alpha oscillations within a passband of 8–14 Hz from the EEG data; detecting an amplitude modulation of the alpha oscillations; extracting beta oscillations within a passband of 20–30 Hz from the EEG data; detecting an amplitude modulation of the beta oscillations; determining the anesthetic state of the subject based on the frequency modulation of the slow-delta oscillations, the amplitude modulation of the alpha oscillations, and the amplitude modulation of the beta oscillations; and adjusting a dosage of the GABAergic anesthetic based on the anesthetic state of the subject and a desired level of unconsciousness of the subject.

11. The method of claim 10, wherein the adjusting a dosage of the GABAergic anesthetic comprises manually adjusting the dosage of the GABAergic anesthetic by an anesthesia care giver.

12. The method of claim 10, wherein the adjusting a dosage of the GABAergic anestheticAttorney Docket No. MIT-25121WO01 comprises automatically adjusting the dosage of the GABAergic anesthetic.

13. The method of claim 10, wherein the GABAergic anesthetic is at least one of propofol, propanidid, or sevoflurane.

14. A method of monitoring an anesthetic state of a subject sedated with a GABAergic anesthetic, the method comprising: while the subject is sedated with the GABAergic anesthetic, obtaining electroencephalography (EEG) data of the subject with electrodes in electrical communication with the subject; extracting at least one signal from the EEG data between X Hz and Y Hz, where X and Y are numbers between 0 and 200 with X < Y; detecting a modulation of the at the least one signal; determining the anesthetic state of the subject based on the modulation of the at least one signal; and adjusting a dosage of the GABAergic anesthetic based on the anesthetic state of the subject and a desired level of unconsciousness of the subject.

15. The method of claim 14, wherein the adjusting a dosage of the GABAergic anesthetic comprises manually adjusting the dosage of the GABAergic anesthetic by an anesthesia care giver.

16. The method of claim 14, wherein the adjusting a dosage of the GABAergic anesthetic comprises automatically adjusting the dosage of the GABAergic anesthetic.

17. The method of claim 14, wherein the GABAergic anesthetic is at least one of propofol or sevoflurane.

18. A system comprising: an electroencephalogram (EEG) recording device to measure an EEG of a subject sedated with a GABAergic anesthetic; an infusion line connected to the subject to provide the GABAergic anesthetic to the subject; a monitor, operably connected to the electroencephalogram (EEG) recording device, toAttorney Docket No. MIT-25121WO01 filter oscillatory signals out of the EEG from desired frequency bands; a controller, operably connected to the monitor to determine a state of anesthetic- mediated unconsciousness of the subject based on the oscillatory signals and to adjust a dosage of the GABAergic anesthetic based on the state of anesthetic-mediated unconsciousness of the subject and a desired level of unconsciousness of the subject; a graphical user interface (GUI), operably connected to the monitor and controller, to set the desired level of unconsciousness of the subject; and an infusion pump, operably connected to the monitor, controller, and GUI, to pump the dosage of the GABAergic anesthetic into the subject through the infusion line.

19. The system of claim 18, wherein the electroencephalogram (EEG) recording device comprises one or more electrodes in electrical communication with the subject.

20. The system of claim 18, wherein the desired level of unconsciousness of the subject may be set automatically.

21. The system of claim 18, wherein the desired level of unconsciousness of the subject may be set by an anesthesia care giver.

Images