Quantifying Clinical Outcome Measures in Patients with Epilepsy Using the Electronic Health Record

Xie, Kevin; Litt, Brian; Roth, Dan; Ellis, Colin A

doi:10.18653/v1/2022.bionlp-1.36

Cited by 9 publications

(16 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, we summarized the preictal data with node strength, a global network measure that is robust to the incomplete sampling of the iEEG brain network, 46,47 . Baseline seizure frequency and severity were not controlled for when assessing an individual's propensity for severe seizures following medication taper, and future studies might additionally incorporate an individual's trajectory in seizure severity over the course of their disease using clinical notes from the electronic health record 48 . People admitted to the EMU are often medically refractory, thus seizure severity scores may differ in ambulatory EEG, where an individual may be exposed to more seizure triggers.…”

Section: Discussionmentioning

confidence: 99%

A quantitative tool for seizure severity: diagnostic and therapeutic applications

Pattnaik

Ghosn

Ong

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Objective: More than one-third of the people with focal epilepsy do not achieve seizure freedom with medication, neuromodulation, or neurosurgery therapies. Palliative care with the goal of reducing epilepsy burden is an alternative for these patients. Minimizing severe seizures is essential for reducing morbidity. Existing seizure severity scales are qualitative and rely on patient reports, limiting our ability to rigorously track and intervene to curb severe seizures. The goal of this study is to develop and validate a quantitative metric for seizure severity. Methods: We retrospectively analyzed preictal and ictal intracranial-EEG (iEEG) recordings from 54 people with drug-resistant epilepsy undergoing pre-surgical evaluation. We developed a new metric that objectively combines seizure duration, spread, and semiology to quantify seizure severity. We calculated preictal iEEG network features and fit a linear mixed-effects model to quantify patient-specific associations between preictal networks and seizure severity. Results: We evaluated 256 seizures from 54 patients using the quantitative seizure severity score. Seizure severity was consistent with clinical seizure type. Medication taper strategy was associated with seizure severity (p = 0.018, 97.5% confidence interval = [-1.242, -0.116]) and lower pre-surgical seizure severity was associated with better post-surgical seizure outcome (U = 465, p = 0.042). A linear mixed-effects model with preictal network features as regressors and seizure severity as response revealed a group-level positive trend. In 12 out of 14 patients with multiple types of seizures, more severe seizures were preceded by more abnormal preictal networks. Significance: We present a quantitative metric for seizure severity that correlates with clinical and electrographic features. We found that the seizure severity score was associated with abnormal preictal networks. We propose this measure to holistically capture patient condition and guide incremental changes in therapy to improve patient outcome over time.

show abstract

Section: Discussionmentioning

confidence: 99%

A quantitative tool for seizure severity: diagnostic and therapeutic applications

Pattnaik

Ghosn

Ong

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We then used a combination of neural summarization with the T5 language model 10 and custom rules-based quantification to convert the extracted text spans into quantitative frequencies and date-time objects, respectively. 5 In this study, we applied this pipeline to our dataset of outpatient office notes. For each note, we classified patients as seizure-free or having recent seizures, and we quantified the seizure frequency and date of most recent seizure.…”

Section: Nlp Model Development and Implementationmentioning

confidence: 99%

“…We recently developed and validated a natural language processing (NLP) algorithm to extract seizure freedom, seizure frequency, and date of most recent seizure from the text of outpatient progress notes for patients with epilepsy. 4,5 Using annotated clinical notes, we fine-tuned and applied state-of-the-art transformer language models to rapidly read and comprehend clinical note text. The algorithm achieved near-human performance at classifying patients as seizure-free at each clinic visit (median accuracy = 84%) and human performance at determining seizure frequency (accuracy = 88%, F 1 score = 85%) and the date of most recent seizure (accuracy = 86%, F 1 score = 83%).…”

Section: Introductionmentioning

confidence: 99%

“…The algorithm achieved near-human performance at classifying patients as seizure-free at each clinic visit (median accuracy = 84%) and human performance at determining seizure frequency (accuracy = 88%, F 1 score = 85%) and the date of most recent seizure (accuracy = 86%, F 1 score = 83%). 4,5 In this study, we sought to (1) determine the feasibility of extracting these epilepsy outcomes measures from an EHR at large scale and (2) use this approach to characterize epilepsy outcomes over time at our academic epilepsy center. We applied our method to 55 630 outpatient progress notes from the epilepsy center at our institution over a 12-year period.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Long‐term epilepsy outcome dynamics revealed by natural language processing of clinic notes

et al. 2023

Self Cite

View full text Add to dashboard Cite

Objective Electronic medical records allow for retrospective clinical research with large patient cohorts. However, epilepsy outcomes are often contained in free text notes that are difficult to mine. We recently developed and validated novel natural language processing (NLP) algorithms to automatically extract key epilepsy outcome measures from clinic notes. In this study, we assessed the feasibility of extracting these measures to study the natural history of epilepsy at our center. Methods We applied our previously validated NLP algorithms to extract seizure freedom, seizure frequency, and date of most recent seizure from outpatient visits at our epilepsy center from 2010 to 2022. We examined the dynamics of seizure outcomes over time using Markov model‐based probability and Kaplan–Meier analyses. Results Performance of our algorithms on classifying seizure freedom was comparable to that of human reviewers (algorithm F1 = .88 vs. human annotator κ = .86). We extracted seizure outcome data from 55 630 clinic notes from 9510 unique patients written by 53 unique authors. Of these, 30% were classified as seizure‐free since the last visit, 48% of non‐seizure‐free visits contained a quantifiable seizure frequency, and 47% of all visits contained the date of most recent seizure occurrence. Among patients with at least five visits, the probabilities of seizure freedom at the next visit ranged from 12% to 80% in patients having seizures or seizure‐free at the prior three visits, respectively. Only 25% of patients who were seizure‐free for 6 months remained seizure‐free after 10 years. Significance Our findings demonstrate that epilepsy outcome measures can be extracted accurately from unstructured clinical note text using NLP. At our tertiary center, the disease course often followed a remitting and relapsing pattern. This method represents a powerful new tool for clinical research with many potential uses and extensions to other clinical questions.

show abstract

“…We then used a previously validated text mining pipeline to automatically extract and quantify each patient's monthly seizure frequency from the clinical note text. 22,23 For patients in whom the baseline seizure frequency could not be determined with text mining (n = 37), the baseline seizure frequency was determined by a board-certified epileptologist (E.C.C.) by manually reviewing the EMU admission note.…”

Section: Baseline Seizure Frequencymentioning

confidence: 99%

A pharmacokinetic model of antiseizure medication load to guide care in the epilepsy monitoring unit

et al. 2023

Self Cite

View full text Add to dashboard Cite

Objective: Evaluating patients with drug-resistant epilepsy often requires inducing seizures by tapering antiseizure medications (ASMs) in the epilepsy monitoring unit (EMU). The relationship between ASM taper strategy, seizure timing, and severity remains unclear. In this study, we developed and validated a pharmacokinetic model of total ASM load and tested its association with seizure occurrence and severity in the EMU. Methods:We studied 80 patients who underwent intracranial electroencephalographic recording for epilepsy surgery planning. We developed a first order pharmacokinetic model of the ASMs administered in the EMU to generate a continuous metric of overall ASM load. We then related modeled ASM load to seizure likelihood and severity. We determined the association between the rate of ASM load reduction, the length of hospital stay, and the probability of having a severe seizure. Finally, we used modeled ASM load to predict oncoming seizures.Results: Seizures occurred in the bottom 50th percentile of sampled ASM loads across the cohort (p < .0001, Wilcoxon signed-rank test), and seizures requiring rescue therapy occurred at lower ASM loads than seizures that did not require rescue therapy (logistic regression mixed effects model, odds ratio = .27, p = .01).Greater ASM decrease early in the EMU was not associated with an increased likelihood of having a severe seizure, nor with a shorter length of stay.

show abstract

Quantifying Clinical Outcome Measures in Patients with Epilepsy Using the Electronic Health Record

Cited by 9 publications

References 0 publications

A quantitative tool for seizure severity: diagnostic and therapeutic applications

A quantitative tool for seizure severity: diagnostic and therapeutic applications

Long‐term epilepsy outcome dynamics revealed by natural language processing of clinic notes

A pharmacokinetic model of antiseizure medication load to guide care in the epilepsy monitoring unit

Contact Info

Product

Resources

About