Marcus Ng

Dr. Marcus Ng

orcid

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 June 2013 in Epilepsy Research and Treatment
Verified
Journal Article
Open access
PDF file
MNMarcus Ng
Marcus Ng
MPMilena Pavlova
Milena Pavlova
Since the formal characterization of sleep stages, there have been reports that seizures may preferentially occur in certain phases of sleep. Through ascending cholinergic connections from the brainstem, rapid eye movement (REM) sleep is physiologically characterized by low voltage fast activity on the electroencephalogram, REMs, and muscle atonia. Multiple independent studies confirm that, in REM sleep, there is a strikingly low proportion of seizures (~1% or less). We review a total of 42 distinct conventional and intracranial studies in the literature which comprised a net of 1458 patients. Indexed to duration, we found that REM sleep was the most protective stage of sleep against focal seizures, generalized seizures, focal interictal discharges, and two particular epilepsy syndromes. REM sleep had an additional protective effect compared to wakefulness with an average 7.83 times fewer focal seizures, 3.25 times fewer generalized seizures, and 1.11 times fewer focal interictal discharges. In further studies REM sleep has also demonstrated utility in localizing epileptogenic foci with potential translation into postsurgical seizure freedom. Based on emerging connectivity data in sleep, we hypothesize that the influence of REM sleep on seizures is due to a desynchronized EEG pattern which reflects important connectivity differences unique to this sleep stage.

25 April 2023 in Neurology
Verified
Journal Article
Closed access
PDF file
JJJin Jing
Jin Jing
WGWendong Ge
Wendong Ge
ASAaron F. Struck
Aaron F. Struck
MFMarta Fernandes
Marta Fernandes
SHShenda Hong
Shenda Hong
MWMichael Brandon Westover
Michael Brandon Westover
Background and Objectives: The validity of brain monitoring using electroencephalography (EEG), particularly to guide care in patients with acute or critical illness, requires that experts can reliably identify seizures and other potentially harmful rhythmic and periodic brain activity, collectively referred to as “ictal-interictal-injury continuum" (IIIC). Prior inter-rater reliability (IRR) studies are limited by small samples and selection bias. This study was conducted to assess the reliability of experts in identifying IIIC. Methods: This prospective analysis included 30 experts with subspecialty clinical neurophysiology training from 18 institutions. Experts independently scored varying numbers of ten-second EEG segments as: “Seizure (SZ)”, “Lateralized Periodic Discharges (LPD)”, “Generalized Periodic Discharges (GPD)”, “Lateralized Rhythmic Delta Activity (LRDA)”, “Generalized Rhythmic Delta Activity (GRDA)”, or “Other”. EEGs were performed for clinical indications at Massachusetts General Hospital between 2006 to 2020. Primary outcome measures were pairwise IRR (average percent agreement (PA) between pairs of experts) and majority IRR (average PA with group consensus) for each class; and beyond chance agreement (κ). Secondary outcomes were calibration of expert scoring to group consensus, and latent trait analysis to investigate contributions of bias and noise to scoring variability. Results: Among 2,711 EEGs, 49% were from females, and median (IQR) age was 55 (41). In total experts scored 50,697 EEG segments; the median [range] number scored by each expert was 6,287.5 [1,002, 45,267]. Overall pairwise IRR was moderate (PA 52%, κ 42%), and majority IRR was substantial (PA 65%, κ 61%). Noise-bias analysis demonstrated that a single underlying receiver operating curve can account for most variation in experts9 false positive vs true positive characteristics (median [range] of variance explained (R2): 95 [93, 98]%), and for most variation in experts’ precision vs sensitivity characteristics (R2: 75 [59, 89]%). Thus, variation between experts is mostly attributable not to differences in expertise, but rather to variation in decision thresholds. Discussion: Our results provide precise estimates of expert reliability from a large and diverse sample, and a parsimonious theory to explain the origin of disagreements between experts. The results also establish a standard for how well an automated IIIC classifier must perform to match experts. Classification of Evidence: This study provides Class II evidence that independent expert review reliably identifies ictal-interictal injury continuum patterns on EEG compared to expert consensus.

24 July 2020 in Frontiers in Neurology
Verified
Journal Article
Open access
PDF file
GMGraham A. McLeod
Graham A. McLeod
AGAmirhossein Ghassemi
Amirhossein Ghassemi
MNMarcus Ng
Marcus Ng
Epilepsy is a common and debilitating neurological disease. When medication cannot control seizures in up to 40% of cases, surgical resection of epileptogenic tissue is a clinically and cost- effective therapy to achieve seizure freedom. To simultaneously resect minimal yet sufficient cortex, exquisite localization of the epileptogenic zone (EZ) is crucial. However, localization is not straightforward, given relative difficulty of capturing seizures, constraints of the inverse problem in source localization, and possible disparate locations of symptomatogenic vs. epileptogenic regions. Thus, attention has been paid to which state of vigilance best localizes the EZ, in the hopes that one or another sleep-wake state may hold the key to improved accuracy of localization. Studies investigating this topic have employed diverse methodologies and produced diverse results. Nonetheless, rapid eye movement sleep (REM) has emerged as a promising sleep-wake state, as epileptic phenomena captured in REM may spatially correspond more closely to the EZ. Cortical neuronal asynchrony in REM may spatially constrain epileptic phenomena to reduce propagation away from the source generator, rendering them of high localizing value. However, some recent work demonstrates best localization in sleep-wake states other than REM, and there are reports of REM providing clearly false localization. Moreover, synchronistic properties and basic mechanisms of human REM remain to be fully characterized. Amidst these uncertainties, there is an urgent need for recording and analytical techniques to improve accuracy of localization. Here we present a systematic review and quantitative analysis of pertinent literature on whether and how REM may help localize epileptogenic foci. To help streamline and accelerate future work on the intriguing anti-epileptic properties of REM, we also introduce a simple, conceptually clear set-theoretic framework to conveniently and rigorously describe the spatial properties of epileptic phenomena in the brain.

26 June 2025 in NEJM AI
Verified
Journal Article
Closed access
JLJun Li
Jun Li
Daniel M GoldenholzDaniel M Goldenholz
Daniel M Goldenholz
MAMoritz Alkofer
Moritz Alkofer
CSChenxi Sun
Chenxi Sun
FNFabio A. Nascimento
Fabio A. Nascimento
JJJin Jing
Jin Jing
Epileptiform discharges, or spikes, within electroencephalogram (EEG) recordings are essential for diagnosing epilepsy and localizing seizure origins. Artificial intelligence (AI) offers a promising approach to automating detection, but current models are often hindered by artifact-related false positives and often target either event- or EEG-level classification, thus limiting clinical utility. We developed SpikeNet2, a deep-learning model based on a residual network architecture, and enhanced it with hard-negative mining to reduce false positives. Our study analyzed 17,812 EEG recordings from 13,523 patients across multiple institutions, including Massachusetts General Brigham (MGB) hospitals. Data from the Human Epilepsy Project (HEP) and SCORE-AI (SAI) were also included. A total of 32,433 event-level samples, labeled by experts, were used for training and evaluation. Performance was assessed using the area under the receiver operating characteristic curve (AUROC), the area under the precision–recall curve (AUPRC), calibration error, and a modified area under the curve (mAUC) metric. The model’s generalizability was evaluated using external datasets. SpikeNet2 demonstrated strong performance in event-level spike detection, achieving an AUROC of 0.973 and an AUPRC of 0.995, with 44% of experts surpassing the model on the MGB dataset. In external validation, the model achieved an AUROC of 0.942 and an AUPRC of 0.948 on the HEP dataset. For EEG-level classification, SpikeNet2 recorded an AUROC of 0.958 and an AUPRC of 0.959 on the MGB dataset, an AUROC of 0.888 and an AUPRC of 0.823 on the HEP dataset, and an AUROC of 0.995 and an AUPRC of 0.991 on the SAI dataset, with 32% of experts outperforming the model. The false-positive rate was reduced to an average of nine spikes per hour. SpikeNet2 offers expert-level accuracy in both event-level spike detection and EEG-level classification, while significantly reducing false positives. Its dual functionality and robust performance across diverse datasets make it a promising tool for clinical and telemedicine applications, particularly in resource-limited settings. (Funded by the National Institutes of Health and others.)

2 November 2022 in Neurotherapeutics
Verified
Journal Article
Closed access
PDF file
MNMarcus Ng
Marcus Ng
HEHussam El-Alawi
Hussam El-Alawi
DTDarion Toutant
Darion Toutant
ECEun Hyung Choi
Eun Hyung Choi
NWNatalie Wright
Natalie Wright
JKJi Hyun Ko
Ji Hyun Ko

12 September 2021 in Epilepsia Open
Verified
Journal Article
Open access
PDF file
MNMarcus Ng
Marcus Ng
MPMilena Pavlova
Milena Pavlova
Objective The World Health Organization, International League Against Epilepsy (ILAE), and International Bureau for Epilepsy have called epilepsy a public health imperative, with appropriate emphasis on low-to-middle income countries (LMIC). Although Canada is a high-income country (HIC), income is not distributed uniformly. Furthermore, epilepsy data from the national statistical agency explicitly overlooks the Arctic by excluding these territories. A common neurologic emergency, status epilepticus (SE) is a life-threatening manifestation of epilepsy that demands prompt treatment to avoid death and long-term sequelae. Therefore, we examined the rate of SE in a well-defined Canadian Arctic region. Methods This study takes epidemiologic advantage of the Kivalliq Region’s geographical isolation, which is accessible only by air. All SE patients requiring emergency care are consistently flown 1,200-1,900 kilometres to a single designated hospital in a distinct southern part of Canada for further management and electroencephalography (EEG). We conducted a retrospective database and chart review at this “bottleneck” hospital to identify patients with seizure(s) severe enough to justify emergency airborne medical evacuation over a 11.25 year period from 2009-2020. Results We screened 40,392 EEGs to yield 117 distinct medical evacuations for “operational SE” from 99 patients to derive estimated SE incidences of 99.9 evacuations per 100,000/year, and 84.5 patients per 100,000/year. The average time from seizure onset to EEG was 3.2 days. Only 16.2% of SE patients had known epilepsy. For “confirmed SE” cases meeting ILAE criteria, or cases with persistently epileptiform EEG despite days of empiric treatment, estimated incidence was 77.7 evacuations per 100,000/year, and 64.9 patients per 100,000/year. Significance High SE and epilepsy rates in the Canadian Arctic are consistent with LMIC rather than HIC. Our findings demonstrate the paradox of LMIC-equivalent epilepsy populations camouflaged within HIC. Our findings also highlight the longstanding plight of these under-served and overlooked populations hidden in plain sight.