Frontiers in Neurology, Год журнала: 2024, Номер 15
Опубликована: Июль 11, 2024
Seizures have a profound impact on quality of life and mortality, in part because they can be challenging both to detect forecast. Seizure detection relies upon accurately differentiating transient neurological symptoms caused by abnormal epileptiform activity from similar with different causes. forecasting aims identify when person has high or low likelihood seizure, which is related seizure prediction. Machine learning artificial intelligence are data-driven techniques integrated neurodiagnostic monitoring technologies that attempt accomplish those tasks. In this narrative review, we describe the existing software hardware approaches for forecasting, as well concepts how evaluate performance new future application clinical practice. These include long-term without electroencephalography (EEG) report very sensitivity reduced false positive detections. addition, implications evaluation novel treatments seizures within trials. Based these data, machine could fundamentally change care people seizures, but there multiple validation steps necessary rigorously demonstrate their benefits costs, relative current standard.
Язык: Английский
Процитировано
9Can ChatGPT 4.0 Diagnose Epilepsy? A Study on Artificial Intelligence’s Diagnostic Capabilities
Journal of Clinical Medicine, Год журнала: 2025, Номер 14(2), С. 322 - 322
Опубликована: Янв. 7, 2025
Objectives: This study investigates the potential of artificial intelligence (AI), specifically large language models (LLMs) like ChatGPT, to enhance decision support in diagnosing epilepsy. AI tools can improve diagnostic accuracy, efficiency, and decision-making speed. The aim this was compare level agreement epilepsy diagnosis between human experts (epileptologists) (ChatGPT), using 2014 International League Against Epilepsy (ILAE) criteria, identify predictors errors made by ChatGPT. Methods: A retrospective analysis conducted on data from 597 patients who visited emergency department for either a first epileptic seizure or recurrence. Diagnoses experienced epileptologists were compared with those ChatGPT 4.0, which trained ILAE definition. diagnoses assessed Cohen's kappa statistic. Sensitivity specificity 2 × contingency tables, multivariate analyses performed variables associated errors. Results: Neurologists diagnosed 216 (36.2%), while it 109 (18.2%). neurologists very low, value -0.01 (95% confidence intervals, CI: -0.08 0.06). ChatGPT's sensitivity 17.6% 14.5-20.6), 81.4% 78.2-84.5), positive predictive 34.8% 31.0-38.6), negative 63.5% 59.6-67.4). 41.7% cases, more frequent older specific medical conditions. correct classification acute symptomatic seizures unknown etiology. Conclusions: 4.0 does not reach clinicians' performance epilepsy, showing poor identifying but better at recognizing non-epileptic cases. overall concordance clinicians is extremely low. Further research needed accuracy other LLMs.
Язык: Английский
Процитировано
1The use of AI in epilepsy and its applications for people with intellectual disabilities: commentary
Acta Epileptologica, Год журнала: 2025, Номер 7(1)
Опубликована: Фев. 19, 2025
Abstract Epilepsy is one of the most common neurological disorders, affecting more than 50 million people worldwide. Management particularly complex in individuals with intellectual disabilities, who are at a much higher risk having severe seizures compared to general population. People disabilities regularly excluded from epilepsy research, despite significantly risks negative health outcomes and early mortality. Recent advances artificial intelligence (AI) have shown great potential improving diagnosis, monitoring, management epilepsy. Machine learning techniques been used analysing electroencephalography data for efficient seizure detection prediction, as well individualised treatment, which facilitates timely customised intervention Research implementation AI-based solutions still remains limited due lack accessible long-term clinical model training, difficulties communicating ethical challenges ensuring safety AI systems this This paper presents an overview recent applications highlighting key necessity including research on epilepsy, strategies promote development use vulnerable Given prevalence consequences associated application care has significant positive impact. To achieve impact avoid increasing existing inequity, there urgent need greater inclusion around management.
Язык: Английский
Процитировано
1The promise of AI Large Language Models for Epilepsy care
Epilepsy & Behavior, Год журнала: 2024, Номер 154, С. 109747 - 109747
Опубликована: Март 21, 2024
Язык: Английский
Процитировано
7Unsupervised learning from EEG data for epilepsy: A systematic literature review
Artificial Intelligence in Medicine, Год журнала: 2025, Номер 162, С. 103095 - 103095
Опубликована: Фев. 22, 2025
Язык: Английский
Процитировано
1Machine learning applied to epilepsy: bibliometric and visual analysis from 2004 to 2023
Frontiers in Neurology, Год журнала: 2024, Номер 15
Опубликована: Апрель 2, 2024
Background Epilepsy is one of the most common serious chronic neurological disorders, which can have a negative impact on individuals, families and society, even death. With increasing application machine learning techniques in medicine recent years, integration with epilepsy has received close attention, potential to provide reliable optimal performance for clinical diagnosis, prediction, precision through use various types mathematical algorithms, promises make better parallel advances. However, no bibliometric assessment been conducted evaluate scientific progress this area. Therefore, study aims visually analyze trend current state research related bibliometrics visualization. Methods Relevant articles reviews were searched 2004–2023 using Web Science Core Collection database, analyses visualizations performed VOSviewer, CiteSpace, Bibliometrix (R-Tool R-Studio). Results A total 1,284 papers retrieved from Wo SCC database. The number shows an year by year. These mainly 1,957 organizations 87 countries/regions, majority United States China. journal highest published EPILEPSIA. Acharya, U. Rajendra (Ngee Ann Polytechnic, Singapore) authoritative author field his paper “Deep Convolutional Neural Networks Automated Detection Diagnosis Epileptic Seizures Using EEG Signals” was cited. Literature keyword analysis that seizure management neuroimaging are hotspots developments. Conclusions This first methods visualize areas epilepsy, revealing trends frontiers field. information will useful reference researchers focusing learning.
Язык: Английский
Процитировано
6Evaluating the Clinical Validity and Reliability of Artificial Intelligence-Enabled Diagnostic Tools in Neuropsychiatric Disorders
Cureus, Год журнала: 2024, Номер unknown
Опубликована: Окт. 16, 2024
Neuropsychiatric disorders (NPDs) pose a substantial burden on the healthcare system. The major challenge in diagnosing NPDs is subjective assessment by physician which can lead to inaccurate and delayed diagnosis. Recent studies have depicted that integration of artificial intelligence (AI) neuropsychiatry could potentially revolutionize field precisely complex neurological mental health timely fashion providing individualized management strategies. In this narrative review, authors examined current status AI tools assessing neuropsychiatric evaluated their validity reliability existing literature. analysis various datasets including MRI scans, EEG, facial expressions, social media posts, texts, laboratory samples accurate diagnosis conditions using machine learning has been profoundly explored article. recent trials tribulations encouraging future scope utility application discussed. Overall proved be feasible applicable it about time research translates clinical settings for favorable patient outcomes. Future should focus presenting higher quality evidence superior adaptability establish guidelines providers maintain standards.
Язык: Английский
Процитировано
5Neuropathology and epilepsy surgery - 2024 update.
PubMed, Год журнала: 2024, Номер 5
Опубликована: Янв. 1, 2024
Neuropathology-based studies in neurosurgically resected brain tissue obtained from carefully examined patients with focal epilepsies remain a treasure box for excellent insights into human neuroscience, including avenues to better understand the neurobiology of organization and neuronal hyperexcitability at cellular level glio-neuronal interaction. It also allows translate results animal models order develop personalized treatment strategies near future. A nice example this is discovery new disease entity 2017, termed mild malformation cortical development oligodendroglial hyperplasia epilepsy or MOGHE, frontal lobe young children intractable seizures. In 2021, somatic missense mutation galactose transporter SLC35A2 leading altered glycosylation lipoproteins Golgi apparatus was detected 50 % MOGHE samples. 2023, first clinical trial evaluated supplementation histopathologically confirmed carrying
Язык: Английский
Процитировано
4Supervised machine learning compared to large language models for identifying functional seizures from medical records
Epilepsia, Год журнала: 2025, Номер unknown
Опубликована: Фев. 17, 2025
Abstract Objective The Functional Seizures Likelihood Score (FSLS) is a supervised machine learning–based diagnostic score that was developed to differentiate functional seizures (FS) from epileptic (ES). In contrast this targeted approach, large language models (LLMs) can identify patterns in data for which they were not specifically trained. To evaluate the relative benefits of each we compared performance FSLS two LLMs: ChatGPT and GPT‐4. Methods total, 114 anonymized cases constructed based on patients with documented FS, ES, mixed ES or physiologic seizure‐like events (PSLEs). Text‐based presented three sequential prompts LLMs, showing history present illness (HPI), electroencephalography (EEG) results, neuroimaging results. We accuracy (number correct predictions/number cases) area under receiver‐operating characteristic (ROC) curves (AUCs) LLMs using mixed‐effects logistic regression. Results 74% (95% confidence interval [CI] 65%–82%) AUC 85% CI 77%–92%). GPT‐4 superior both ( p <.001), an 77%–91%) 87% 79%–95%). Cohen's kappa between 40% (fair). provided different predictions days when same note 33% patients, LLM's self‐rated certainty moderately correlated observed variability (Spearman's rho 2 : 30% [fair, ChatGPT] 63% [substantial, GPT‐4]). Significance Both identified substantial subset FS clinical history. fair agreement highlights differently structured score. inconsistency LLMs' across incomplete insight into their own consistency concerning. This comparison cautions about how learning artificial intelligence could practice.
Язык: Английский
Процитировано
0Electroencephalography-driven brain-network models for personalized interpretation and prediction of neural oscillations
Clinical Neurophysiology, Год журнала: 2025, Номер unknown
Опубликована: Март 1, 2025
Язык: Английский
Процитировано
0