About

Emery N. Brown is the Edward Hood Professor of Medical Engineering and Computational Neuroscience at the Massachusetts Institute of Technology (MIT), the Warren M. Zapol Professor of Anaesthesia at Harvard Medical School, and a practicing anesthesiologist at Massachusetts General Hospital. He holds a B.A. in Applied Mathematics from Harvard College, an M.A. and Ph.D. in statistics from Harvard University, and an M.D. from Harvard Medical School. Dr. Brown is an anesthesiologist-statistician whose experimental research has significantly contributed to understanding how anesthetics act in the brain to induce general anesthesia. His work involves developing signal processing algorithms to analyze neuroscience data, characterizing neural representations during learning, and understanding the neurophysiological mechanisms underlying anesthesia. His research employs systems neuroscience approaches using fMRI, EEG, neurophysiological recordings, and mathematical modeling in interdisciplinary collaborations. Dr. Brown's long-term goal is to establish a neurophysiological definition of anesthesia, develop safer anesthetic drugs, and improve methods for measuring anesthesia depth. He has served on various NIH and NSF advisory committees and is a member of prestigious organizations including the Institute of Medicine, the National Academy of Sciences, and the National Academy of Engineering.

Research topics

• Psychology
• Neuroscience
• Computer Science
• Medicine
• Internal medicine
• Anesthesia
• Sociology
• Political Science
• Biology
• Engineering ethics
• Physics
• Social psychology
• Surgery
• Cognitive psychology
• Communication
• Telecommunications
• Acoustics
• Data science
• Engineering
• Epistemology
• Management science

Selected publications

• Electroencephalographic Monitoring in the Recovery Room for Identification of Patients at Risk for Postoperative Delirium

  Anesthesiology · 2026-04-15

  articleOpen access

  BACKGROUND: Electroencephalography (EEG), especially the assessment of alpha oscillations (8-12 Hz), has gained attention as a non-invasive marker of neural network integrity, with reduced alpha power linked to thalamocortical disconnectivity and increased vulnerability to postoperative delirium (POD) in elderly patients. Recent studies have demonstrated that altered EEG dynamics in the alpha-band frequencies during anesthesia emergence in the post-anesthesia care unit (PACU) are associated with POD. However, EEG data postoperatively from the recovery room are scarce. METHODS: This prospective observational study (2019-2022) at Charité-Universitätsmedizin Berlin investigates frontal EEG signatures and postoperative delirium (POD) in older adults undergoing surgery. Postoperative POD screening was performed in the recovery room and twice daily for five days. While the overall study collected pre- and intraoperative EEGs, we focus here on frontal EEG files recorded in the recovery room while patients were awake. Spectral-domain EEG features and monitor-derived indices were compared by multivariable regression analysis between those who developed POD and those who did not. RESULTS: A total of 184 patients had sufficient EEG data for analysis; 57 (31%) developed POD. Recovery-room recordings showed significantly reduced alpha- and beta-band power in POD vs NoPOD patients with cumulative 10-20 Hz power providing the best discrimination (AUC 0.69, 95% CI 0.60-0.77). NoPOD patients exhibited increases in alpha- and beta-band power (8-20 Hz) relative to preoperative baseline, with highest augmentation at approximately 16 Hz, whereas POD patients retained spectral patterns comparable to baseline across all frequency bands. CONCLUSIONS: These data suggest that frontal EEG monitoring in the recovery room can help identify a "vulnerable brain" phenotype associated with increased POD risk. Integrating this monitoring into postoperative care may support risk stratification and more individualized management.

  PublisherDOI

• Time-Frequency Domain Classifier for Propofol-Mediated Unconsciousness

  2025-07-14

  articleSenior author

  Wavelet analysis is a powerful tool in signal processing, enabling precise time-frequency localization that is well-suited for non-stationary signals such as electroencephalograms (EEG). We developed a wavelet-based classifier to assess the level of propofol-mediated unconsciousness. For feature extraction, we used a 6-level Discrete Wavelet Transform (DWT) decomposition followed by simple time domain transformations. The multi-class gradient-boosting classifier generated probability estimates for three classes: conscious, unconscious, and burst suppression. For each window, a continuous class estimate (CCE) was calculated based on these probabilities, and the class with the highest probability was selected as the predicted label. We compared the performance of the DWT-based classifier with other methods that use frequency and time-frequency feature extraction in terms of accuracy and computational efficiency. The DWT-based classifier achieved 92.9% accuracy, comparable to other methods, while providing improved detection of burst suppression events. Additionally, an optimized version of the DWT-based classifier, using a reduced set of features, achieved a similar accuracy of 92.3%, while offering improvements in computational efficiency.

  PublisherDOI

• Long‐term effects of gamma frequency sensory stimulation after 30 months in patients with mild Alzheimer’s disease: Evidence supporting potential disease modification

  Alzheimer s & Dementia · 2025-12-01 · 1 citations

  articleOpen access

  BACKGROUND: Non-invasive gamma-frequency light and sound stimulation at 40Hz has been shown to reduce Alzheimer's disease (AD) pathology and enhance behavioral performance in AD mouse models. Building on these findings, we hypothesized that induced gamma neural activity through light and sound stimulation could serve as a disease-modifying therapy for AD. In the open-label extension phase of this trial, our objective is to assess the effects of long-term daily 40Hz stimulation. METHODS: We conducted a placebo-controlled, randomized control trial in subjects with probable mild AD to use our light and sound device at home for one hour daily (NCT04042922). In the open-label extension phase, we report results after 30 months of daily 40Hz synchronized light and sound in late-onset AD (n=3) and early-onset AD (n=2) patients. Longitudinal electroencephalogram (EEG) was used to evaluate for safety and neural activity when using the 40Hz stimulation. Magnetic resonance imaging (MRI) was used to evaluate brain structure and actigraphy was used to record sleep. Longitudinal neuropsychiatric testing and actigraphy were used to assess cognition and sleep. Plasma pTau217 levels were evaluated longitudinally. RESULTS: Chronic 40Hz light and sound stimulation was safe and well-tolerated after 30 months of daily usage. EEG data show that our novel light and sound device continue to safely and effectively induce 40Hz neural activity. Daily 40Hz stimulation slowed brain atrophy as compared to historical age-matched controls. There was improvement of intradaily stability in 3 out of the 5 participants. Performance on cognitive testing scores improved in late-onset AD patients but not as much in early-onset AD patients. There was a reduction of plasma pTau217 by 47% in 1 participant as compared to baseline. CONCLUSIONS: Gamma frequency light and sound stimulation can be safely used daily for 30 months and slows AD-related degeneration. Induced neural activity using sensory stimulation at 40Hz shows promise as a novel disease modifying therapeutic for Alzheimer's dementia.

  PublisherOA PDFDOI

• Clinical Electroencephalography for Anesthesiologists and Intensivists: Part 2—Physiologic Signatures and Active Management

  Anesthesiology · 2025-11-11 · 4 citations

  articleOpen accessSenior authorCorresponding

  The electroencephalogram (EEG) offers physicians a window into their patients' neurologic and broader physiologic states. Frontal EEG interpretation is emerging as a core competency for anesthesiologists and intensivists. Part 1 of this series reviewed the basics of frontal EEG, including relevant biophysics, neuroanatomy, and anesthetic-mediated frontal EEG signatures of sedation and unconsciousness. In part 2, the authors outline a set of physiologic signatures that integrate the basics of EEG interpretation presented in part 1 with the systemic pathophysiology commonly seen in critical illness. They organize these signatures into a systematic framework to facilitate use of frontal EEG monitoring in the active management of critically ill patients. Finally, the authors use case examples to illustrate how the signatures and framework can guide patient-specific management.

  PublisherDOI

• 5-year follow-up of a fully implanted brain–computer interface in a spinal cord injury patient

  Journal of Neural Engineering · 2025-03-24 · 12 citations

  articleOpen access

  Abstract Spinal cord injury (SCI) affects over 250 000 individuals in the US. Brain–computer interfaces (BCIs) may improve quality of life by controlling external devices. Invasive intracortical BCIs have shown promise in clinical trials but degrade in the chronic period and tether patients to acquisition hardware. Alternatively, electrocorticography (ECoG) records data from electrodes on the cortex, and studies evaluating fully implanted BCI-ECoG systems are scarce. Objective. We seek to address this need using a fully implanted ECoG-based BCI that allows for home use in SCI. Approach. The patient used a long-term BCI system, initially controlling an functional electrical stimulation orthosis in the lab and later using an external mechanical orthosis at home. To evaluate its long-term viability, electrode contact impedance, signal quality, and decoder performance were measured. Signal quality was assessed using signal-to-noise ratio and maximum bandwidth of the signal. Decoder performance was monitored using the area under the receiver operator characteristic curve (AUROC). Main results. The study analyzed data from the patient’s home environment over 54 months, revealing that the device was used at home for 38 ± 24 min on average daily. After six months, we observed stable event-related desynchronization that aided in determining the onset of motor intention. The decoder’s average AUROC across months was 0.959. Importantly, 40 months of the data collected was gather from the subject’s home or community environment. The results indicate long-term ECoG recordings were stable for motor-imagery classification and motor control in the community environment in a case of an individual with SCI. Significance. This study presents the long-term feasibility and viability of an ECoG-based BCI system that persists in the home environment in a case of SCI. Future research should explore larger electrode counts with more participants to confirm this stability. Understanding these trends is crucial for clinical utility and chronic viability in broader patient populations.

  PublisherDOI

• Perioperative Multimodal General Anesthesia Focusing on Specific Central Nervous System Targets Does Not Reduce Postoperative Interleukin-6 and Neurofilament Light Levels in Patients Undergoing Cardiac Surgeries: A Randomized Controlled Trial

  Journal of Cardiothoracic and Vascular Anesthesia · 2025-11-05 · 1 citations

  article
  PublisherDOI

• Convergent effects of different anesthetics on changes in phase alignment of cortical oscillations

  Cell Reports · 2025-05-01 · 8 citations

  articleOpen access

  Many anesthetics cause loss of consciousness despite having diverse underlying molecular and circuit actions. To explore the convergent effects of these drugs, we examine how anesthetic doses of ketamine and dexmedetomidine affect bilateral oscillations in the prefrontal cortex of nonhuman primates. Both anesthetics increase phase locking in the ventrolateral and dorsolateral prefrontal cortex, within and across hemispheres. However, the nature of the phase locking varies. Neighboring prefrontal subregions within a hemisphere show decreased phase alignment with both drugs. Local analyses within a region suggest that this finding could be explained by broad cortical distance-based effects, such as large traveling waves. In contrast, homologous areas across hemispheres become more aligned in phase. Our results suggest that both anesthetics induce strong patterns of cortical phase alignment that are markedly different from those during waking and that these patterns may be a common feature driving loss of responsiveness from different anesthetic drugs.

  PublisherOA PDFDOI

• Evaluation of Two Asynchronous Envelope Detection Techniques for Photoplethysmography

  2025-07-14

  articleSenior author

  Photoplethysmography (PPG) is widely used in both clinical and consumer settings. It is often used to measure heart rate using methods that rely on heart beat detection. However, additional features of the waveform, including the relative levels of the signal envelopes and the resulting amplitude (PPGA), encode valuable physiological information.This work describes and compares two asynchronous techniques for envelope detection in PPG that do not rely on peak detection: the Asymmetric Root-Mean-Square (RMS) detector and the Peak Injection Low-Pass detector. These methods are developed to enhance real-time patient monitoring and facilitate decision-making systems. Both methods are simple to implement and incur minimal time delay. Further, the detectors offer values between beats, simplifying downstream processing by removing an interpolation step.The envelope detectors are evaluated on the CapnoBase TBME benchmark dataset. Across all subjects, the absolute errors for the upper envelope at beat times are 0.60 and 0.02 (arbitrary units) for the half-wave RMS and peak injection methods, respectively. A simple method using envelopes to facilitate peak detection is also presented. This work demonstrates the feasibility of extracting additional information from a widely used physiological measurement.

  PublisherDOI

• Electroencephalogram signatures of propanidid-mediated sedation and unconsciousness

  Journal of Neurophysiology · 2025-07-17 · 1 citations

  articleOpen accessSenior author

  Propanidid is a short-acting intravenous anesthetic of the eugenol group that has been safely used in Mexico for over 20 years. Like propofol, it acts via GABAergic mechanisms; however, its effects on neural circuits remain uncharacterized. Therefore, we compared EEG signatures between patients anesthetized with propanidid and propofol. Our work suggests similar neural mechanisms for propanidid and propofol. As with propofol, propanidid's EEG signatures could be used to monitor sedation and unconsciousness.

  PublisherDOI

• Consciousness Is a Continuum

  Scientific American · 2025-09-01

  article1st authorCorresponding
  PublisherDOI

Recent grants

• NIH Grant R01GM104948

  NIH · $4.4M · 2018

• Thalamocortical Dynamics and Consciousness

  NIH · $2.0M · 2017–2023

• NIH Grant R01GM053559

  NIH · $942k · 2003

• NIH Grant R01MH059733

  NIH · $3.0M · 2010

• Integrated Systems Neuroscience Studies of Anaesthesia

  NIH · $17.8M · 2017–2024

Frequent coauthors

• Patrick L. Purdon

  Stanford University

  531 shared

• Riccardo Barbieri

  Politecnico di Milano

  365 shared

• Zhe Chen

  Guilin University of Electronic Technology

  224 shared

• Oluwaseun Akeju

  Boston University

  186 shared

• Brian L. Edlow

  Massachusetts General Hospital

  175 shared

• Ken Solt

  Massachusetts General Hospital

  167 shared

• Yelena G. Bodien

  136 shared

• Wasim Q. Malik

  124 shared

Labs

• Brain and Cognitive SciencesPI

Awards & honors

• NIH Director’s Pioneer Award
• NIH Director’s Transformative Research Award
• 2011 Jerome Sacks Award for Outstanding Cross Disciplinary R…
• Fellow of the American Institute for Medical and Biological…
• Fellow of the American Statistical Association