Browsing by Author "Soares, Joana Isabel"
Now showing 1 - 10 of 10
Results Per Page
Sort Options
- Comparative evaluation of artificial intelligence chatbots in answering electroencephalography-related questionsPublication . Proença, Soraia; Soares, Joana Isabel; Parra, Joana; Maia, Gisela; Leite, Juliana; Beniczky, Sándor; Jesus-Ribeiro, Joana; Henrique Maia, Gisela MariaAs large language models (LLMs) become more accessible, they may be used to explain challenging EEG concepts to nonspecialists. This study aimed to compare the accuracy, completeness, and readability of EEG-related responses from three LLM-based chatbots and to assess inter-rateragreement. One hundred questions, covering 10 EEG categories, were entered into ChatGPT, Copilot, and Gemini. Six raters from the clinical neurophysiology field (two physicians, two teachers, and two technicians) evaluated the responses. Accuracy was rated on a 6-point scale, completeness on a 3-point scale, and readability was assessed using the Automated Readability Index (ARI). We used a repeated-measures ANOVA for group differences in accuracy and readability, the intraclass correlation coefficient (ICC) for inter-raterreliability, and a two way ANOVA, with chatbot and raters as factors, for completeness. Total accuracy was significantly higher for ChatGPT (mean ± SD 4.54 ± .05) compared with Copilot (mean ± SD 4.11 ± .08) and Gemini (mean ± SD 4.16 ± .13) (p < .001). ChatGPT's lowest performance was in normal variants and patterns of uncertain significance (mean ± SD 3.10 ± .14), while Copilot and Gemini performed lowest in ictal EEG patterns (mean ± SD 2.93 ± .11 and 3.37 ± .24, respectively). Although inter-rater agreement for accuracy was excellent among physicians (ICC = .969) and teachers (ICC = .926), it was poor for technicians in several EEG categories. ChatGPT achieved significantly higher completeness scores than Copilot (p < .001) and Gemini (p = .01). ChatGPT text (ARI − mean ± SD 17.41 ± 2.38) was less readable than Copilot (ARI −mean ± SD 11.14 ± 2.60) (p < .001) and Gemini (ARI − mean ± SD 14.16 ± 3.33). Chatbots achieved relatively high accuracy, but not without flaws, emphasizing that the information provided requires verification. ChatGPT outperformed the other chatbots in accuracy and completeness, though at the expense of readability. The lower inter-rater agreement among technicians may reflect a gap in standardized training or practical experience, potentially impacting the consistency of EEG-related content assessment.
- A custom-built single-channel in-ear electroencephalography sensor for sleep phase detection: an interdependent solution for at-home sleep studiesPublication . Borges, Daniel Filipe; Soares, Joana Isabel; Silva, Heloísa; Felgueiras, João; Batista, Carla; Ferreira, Simão; Rocha, Nuno; Leal, AlbertoSleep is vital for health. It has regenerative and protective functions. Its disruption reduces the quality of life and increases susceptibility to disease. During sleep, there is a cyclicity of distinct phases that are studied for clinical purposes using polysomnography (PSG), a costly and technically demanding method that compromises the quality of natural sleep. The search for simpler devices for recording biological signals at home addresses some of these issues. We have reworked a single-channel in-ear electroencephalography (EEG) sensor grounded to a commercially available memory foam earplug with conductive tape. A total of 14 healthy volunteers underwent a full night of simultaneous PSG, in-ear EEG and actigraphy recordings. We analysed the performance of the methods in terms of sleep metrics and staging. In another group of 14 patients evaluated for sleep-related pathologies, PSG and in-ear EEG were recorded simultaneously, the latter in two different configurations (with and without a contralateral reference on the scalp). In both groups, the in-ear EEG sensor showed a strong correlation, agreement and reliability with the ‘gold standard’ of PSG and thus supported accurate sleep classification, which is not feasible with actigraphy. Single-channel in-ear EEG offers compelling prospects for simplifying sleep parameterisation in both healthy individuals and clinical patients and paves the way for reliable assessments in a broader range of clinical situations, namely by integrating Level 3 polysomnography devices. In addition, addressing the recognised overestimation of the apnea-hypopnea index, due to the lack of an EEG signal, and the sparse information on sleep metrics could prove fundamental for optimised clinical decision making.
- Hippotherapy improves gross motor function in children with cerebral palsy: evidence from a systematic reviewPublication . Bernardino, Inês; Borges, Daniel Filipe; Casalta Lopes, João; Soares, Joana Isabel; Borges, Daniel Filipe"Hippotherapy uses horse movement to promote physical and psychosocial rehabilitation and may benefit children with cerebral palsy (CP). Standardised instruments such as the Activity Scale for Kids-Performance (ASK©), the Gross Motor Function Classification System (GMFCS) and the Gross Motor Function Measure (GMFM) are needed to quantify effects on motor function. To systematically review the effects of hippotherapy on gross motor skills in children with CP. Although autism spectrum disorder (ASD) was included in the search strategy, no eligible ASD studies were identified. Following PRISMA guidelines, six databases (PubMed, EMBASE, Web of Science, SCOPUS, Cochrane and SciELO) were searched for English, Portuguese or Spanish studies employing ASK©, GMFCS or GMFM. Two reviewers independently screened records, extracted data and assessed risk of bias. Twenty-five studies (602 participants, mean age 7.1 years, 3–14) met inclusion criteria; all involved CP, none ASD. Interventions lasted 8–24 weeks (1–3 sessions/week). Two ASK© studies showed significant motor gains (Hedges g=0.48–0.62). GMFM was used in 22 studies; 20 reported clinically relevant improvements, particularly in dimensions D (standing) and E (walking, running, jumping). The sole GMFCS study reported no change in classification. Methodological quality was moderate, limited by small samples and lack of blinding. Hippotherapy improves gross motor function in CP, best demonstrated with GMFM. Evidence for ASD is absent, highlighting a research gap. Broader application of ASK© and GMFCS is still needed to better define benefits across neurodevelopmental disorders."
- Response to: Does reliability benefit from superior visualization of epileptiform discharges on inferior temporal electrodes?Publication . Batista, Carla; Soares, Joana Isabel; Coelho, Paulo; Ferreira, Simão; Rosenzweig, Ivana; Borges, Daniel Filipe; Borges, Daniel Filipe; Ferreira, SimãoWe thank Dr. Kleine for his thoughtful letter regarding our study and for highlighting both the anatomical rationale and the recognized clinical relevance of inferior-temporal electrodes. We welcome the opportunity to clarify specific numerical points and to provide the inter-reviewer agreement data he requested. Our study was intentionally designed as a prospective, blinded, within-subject technical validation comparing the IFCN-recommended 25-electrode 19-channel 10–20 International System (10–20 IS) in consecutive(IFCN-25) array with the conventional 19-channel 10–20 International System (10–20 IS) in consecutive adult EEGs. Importantly, it was not framed as a diagnostic or superiority trial: no external gold standard was applied, and both montages were treated methodologically as equivalent acquisition systems. The primary endpoint was patient-level detection of any abnormality (IEDs or focal slow activity) under blinded real-world conditions. In this reply, we address exclusively the point raised by Dr. Kleine concerning inter-reviewer agreement for temporal-lobe IEDs, providing the 2 × 2 tables and κ values requested. These analyses represent a clear subset of the broader dataset already published and are provided here for transparency.
- Spike detection in the wild: Screening of suspected temporal lobe epilepsy cases using a tailored 2-channel wearable EEGPublication . Borges, Daniel Filipe; Soares, Joana Isabel; Dias, Daniela; Cordeiro, Helena; Leal, Alberto; Borges, Daniel FilipeTo clinically validate the contribution of a custom-built-wearable device (waEEG) compared to a full 10–20 electrode array ambulatory EEG (aEEG) for screening epilepsy cases in patients with suspected temporal lobe epilepsy (TLE) but negative routine EEGs. Patients (aged 16–91 years) with clinically suspected TLE who were referred for a 24 h aEEG were fitted with an additional 2-channel bipolar waEEG device and prospectively enrolled in the study until 20 TLE diagnoses were confirmed by aEEG. 41 patients were included and their waEEG was blindly reviewed by two experienced clinical neurophysiologists and a semi-automated spike detection software to categorize patients into TLE (spikes present) and non-TLE (no spikes) groups. The experts achieved good sensitivity (95%–100%) and accuracy (98%–93%) with excellent interrater agreement (kappa>0.80) in patient labelling. The semi-automated software performed poorly (40% sensitivity, 68% accuracy) and failed to classify TLE in more than half the cases. Classification was not affected by restricting spike detection to the evening and night time, which reduced the average length of the analyzed EEG from 23.4 to 10.4 h. Three false-positive spike detections were thoroughly analyzed and reclassified as artifacts due to eye and body movements and electrocardiographic contamination. To better control cardiac artifacts, the addition of an ECG channel to the waEEG is recommended. Detection of spikes with waEEG allows accurate detection of epilepsy in suspected TLE cases, with less technical and professional effort and improved acceptance. This screening tool could improve the yield of follow-up with a conventional aEEG and provide an accessible method for monitoring interictal epileptiform activity in TLE. Epilepsy is a chronic short circuit in the brain. In adults, it most often affects the temporal lobes, resulting in temporal lobe epilepsy (TLE). Seizures are infrequent but difficult to treat. Electroencephalography (EEG) is the best method to detect the electrical disturbances and is crucial to distinguish epilepsy from other non-epileptic disorders. Developing simple, inexpensive and easily accessible portable EEG methods that complement in-hospital assessment could significantly impact patient care. Our study aims to clinically validate a wearable epilepsy screening device to aid in TLE management, reduce delays in diagnosis and enable straightforward assessment of epileptic activity.
- The prevalence of post-therapy epilepsy in patients treated for high-grade glial tumors: a systematic review and meta-analysisPublication . Ferreira, Marta Pereira; Carvalho, Ruben Lopes; Soares, Joana Isabel; Casalta‑Lopes, João; Borges, Daniel Filipe; Borges, Daniel Filipe; Soares, Joana I.Gliomas are the most prevalent type of primary brain tumor of the adult central nervous system. High-grade gliomas (HGG) are the most common type of glioma. Epilepsy is often the first clinical manifestation of HGG. Since epilepsy leads to increased morbidity and mortality rates, seizure control is one of the main therapeutic goals for patients with glioma-related epilepsy. Post-therapy epilepsy is observed in a significant percentage of patients, hence, this work aimed to quantify the prevalence of post-therapy epilepsy after HGG treatment. Our search was conducted across PubMed®, EMBASE®, Web of Science™, Cochrane Library, Sicelo and Scopus, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. This review included articles published in Portuguese or English that evaluate adult patients with newly diagnosed HGG, who were treated with at least surgery or radiation. Thirty-six studies reporting on 4036 HGG patients were included in our meta-analysis. The mean age ranged from 44 to 73 years. Glioblastoma was the most commonly observed HGG, representing 77,8% of all glioma patients. The pre-treatment seizure frequency was observed in 21,2%. All patients underwent surgery as the main therapy, and 1842 patients received standard adjuvant therapy. We also observed a pooled prevalence of post-therapy seizures of 25.5% (95% confidence interval of [19.9%; 31.1%]). Substantial heterogeneity in all assessed variables was observed. Conducting larger prospective studies with suitable epilepsy diagnostic methods would help provide a more precise estimate of the number of HGG patients who develop post-therapy epilepsy.
- The role of actigraphy in the assessment of central disorders of Hypersomnolence: A systematic review and meta-analysisPublication . Maia, Susana; Soares, Joana Isabel; Borges, Daniel Filipe; Lopes, João Casalta; Gonçalves, Marta; Borges, Daniel FilipeActigraphy provides an objective measure of sleepiness and is recommended by the American Academy of Sleep Medicine for use 7–14 days prior to multiple sleep latency testing. It plays a valuable role in the differential diagnosis of hypersomnolence. Our aim was to provide a comprehensive summary of actigraphy features in central disorders of hypersomnolence (CDH). Data were sourced from six bibliographic databases. Fixed- or random-effects models were applied to compare patients with narcolepsy type 1 (NT1) to controls. Of the 1,737 publications identified in our search, 8 studies met the inclusion criteria. The total sample consisted of 473 participants, encompassing patients with NT1, idiopathic hypersomnia (IH), hypersomnolence with normal CSF hypocretin-1 levels, Kleine–Levin syndrome (KLS), traumatic brain injury (TBI), major depressive disorder (MDD), myotonic dystrophy (MD), primary insomnia and healthy controls. Actigraphy devices varied across studies. Compared to control subjects, NT1 patients had lower total sleep time (TST), sleep efficiency and daytime motor activity, with increased wake after sleep onset, awakenings, nocturnal motor activity and longest nap duration. In KLS, TST was higher during hypersomnia episodes than during asymptomatic phases. TBI and MDD patients had a higher TST than the control group, while MD patients had a lower TST than patients with IH. Actigraphy is a valuable tool for objectively assessing sleep and can assist in detecting CDH. However, the absence of standardized guidelines limits their broader implementation in clinical practice.
- The sound of silence: Quantification of typical absence seizures by sonifying EEG signals from a custom‐built wearable devicePublication . Borges, Daniel Filipe; Fernandes, João; Soares, Joana Isabel; Casalta‐Lopes, João; Carvalho, Daniel; Beniczky, Sándor; Leal, AlbertoObjective: To develop and validate a method for long- term (24- h) objective quantification of absence seizures in the EEG of patients with childhood absence epilepsy (CAE) in their real home environment using a wearable device (waEEG), comparing automatic detection methods with auditory recognition after seizure sonification. Methods: The waEEG recording was acquired with two scalp electrodes. Automatic analysis was performed using previously validated software (Persyst® 14) and then fully reviewed by an experienced clinical neurophysiologist. The EEG data were converted into an audio file in waveform format with a 60- fold time compression factor. The sonified EEG was listened to by three inexperienced observers and the number of seizures and the processing time required for each data set were recorded blind to other data. Quantification of seizures from the patient diary was also assessed. Results: Eleven waEEG recordings from seven CAE patients with an average age of 8.18 ± 1.60 years were included. No differences in the number of seizures were found between the recordings using automated methods and expert audio assessment, with significant correlations between methods (ρ > .89, p < .001) and between observers (ρ > .96, p < .001). For the entire data set, the audio assessment yielded a sensitivity of .830 and a precision of .841, resulting in an F1 score of .835. Significance: Auditory waEEG seizure detection by lay medical personnel provided similar accuracy to post- processed automatic detection by an experienced clinical neurophysiologist, but in a less time- consuming procedure and without the need for specialized resources. Sonification of long- term EEG recordings in CAE provides a user- friendly and cost- effective clinical workflow for quantifying seizures in clinical practice, minimizing human and technical constraints.
- The sound of silence: Quantification of typical absence seizures by sonifying EEG signals from a custom‐built wearable devicePublication . Borges, Daniel Filipe; Fernandes, João; Soares, Joana Isabel; Casalta‐Lopes, João; Carvalho, Daniel; Beniczky, Sándor; Leal, AlbertoTo develop and validate a method for long-term (24-h) objective quantification of absence seizures in the EEG of patients with childhood absence epilepsy (CAE) in their real home environment using a wearable device (waEEG), comparing automatic detection methods with auditory recognition after seizure sonification. The waEEG recording was acquired with two scalp electrodes. Automatic analysis was performed using previously validated software (Persyst® 14) and then fully reviewed by an experienced clinical neurophysiologist. The EEG data were converted into an audio file in waveform format with a 60-fold time compression factor. The sonified EEG was listened to by three inexperienced observers and the number of seizures and the processing time required for each data set were recorded blind to other data. Quantification of seizures from the patient diary was also assessed. Eleven waEEG recordings from seven CAE patients with an average age of 8.18 ± 1.60 years were included. No differences in the number of seizures were found between the recordings using automated methods and expert audio assessment, with significant correlations between methods (ρ > .89, p < .001) and between observers (ρ > .96, p < .001). For the entire data set, the audio assessment yielded a sensitivity of .830 and a precision of .841, resulting in an F1 score of .835. Auditory waEEG seizure detection by lay medical personnel provided similar accuracy to post-processed automatic detection by an experienced clinical neurophysiologist, but in a less time-consuming procedure and without the need for specialized resources. Sonification of long-term EEG recordings in CAE provides a user-friendly and cost-effective clinical workflow for quantifying seizures in clinical practice, minimizing human and technical constraints.
- Wearable sleep staging technology as an alternative to polysomnography: a systematic review and meta-analysisPublication . Borges, Maria; Pereira, Telmo; Borges, Daniel Filipe; Soares, Joana IsabelIntroduction:Sleep is vital for health as it has regenerative and protective functions. During sleep, there is a cyclicity of different phases that are analysed and classified for clinical purposes using polysomnography (PSG), a costly and technically demanding method. The tremendous growth of sleep medicine, where demand for studies far outstrips supply, opens a window for the development of accurate, low-threshold sleep monitoring solutions that can be self-administered at home and could help avoid these issues of convenience, accessibility and reproducibility. Objective:This study aims to analyse the existing literature on the feasibility of wearable devices as an alternative to PSG for the classification of sleep stages. Methods: The literature search was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA). All studies published in English or Portuguese with healthy adults who used wearables to record sleep were included. A meta-analysis was also performed to assess the mean values of the sleep metrics: total sleep time (TST), sleep latency (SL) and wake after sleep onset (WASO), all in minutes and time per stage (in minutes and as relative frequency of TST), as well as sleep efficiency (SE) (in %) and the corresponding statistics between the wearables used and the PSG. Results: Given the high variability of wearables, the analysed metrics whose values were closest to the PSG came from different devices. Nevertheless, the meta-analysis revealed that most wearables tend to overestimate these variables Conclusions: The performance of wearables demonstrates remarkable accuracy in sleep staging, rivalling the gold standard PSG in some variables while providing a more convenient and unobtrusive alternative. This review enriches our global knowledge of sleep measurement and summarizes the limitations that need to be overcome, as informed decision making depends on understanding the different device options, validation contexts and cost implications.