A5050324885- EEG and Brain-Computer Interfaces
- Epilepsy research and treatment
- Functional Brain Connectivity Studies
- Nutrition and Health in Aging
- Neural dynamics and brain function
- Machine Learning in Healthcare
- Blind Source Separation Techniques
- Natural Language Processing Techniques
- Nutritional Studies and Diet
- Biomedical Text Mining and Ontologies
- Cancer Genomics and Diagnostics
- Topic Modeling
- ECG Monitoring and Analysis
- Frailty in Older Adults
- Colorectal Cancer Treatments and Studies
- Bacterial Identification and Susceptibility Testing
- Neutropenia and Cancer Infections
- Data Stream Mining Techniques
- Urinary and Genital Oncology Studies
- Oral health in cancer treatment
- Vascular Procedures and Complications
- Cancer survivorship and care
- Big Data and Business Intelligence
- Neural Networks and Applications
- Colorectal and Anal Carcinomas
University of Coimbra
2019-2024
University of Freiburg
2022-2024
University Medical Center Freiburg
2022-2024
Hospital Beatriz Ângelo
2018-2023
Hospital da Luz
2023
ORCID
2021
Instituto de Pesquisas Tecnológicas
2019
Abstract Seizure prediction may improve the quality of life patients suffering from drug-resistant epilepsy, which accounts for about 30% total epileptic patients. The pre-ictal period determination, characterized by a transitional stage between normal brain activity and seizure, is critical step. Past approaches failed to attain real-world applicability due lack generalization capacity. More recently, deep learning techniques outperform traditional classifiers handle time dependencies....
Abstract The development of seizure prediction models is often based on long-term scalp electroencephalograms (EEGs) since they capture brain electrical activity, are non-invasive, and come at a relatively low-cost. However, suffer from major shortcomings. First, EEG usually highly contaminated with artefacts. Second, changes in the signal over long intervals, known as concept drift, neglected. We evaluate influence these problems deep neural networks using time series shallow widely-used...
Seizure prediction might be the solution to tackle apparent unpredictability of seizures in patients with drug-resistant epilepsy, which comprise about a third all epilepsy. Designing seizure models involves defining pre-ictal period, transition stage between inter-ictal brain activity and discharge. This period is typically fixed interval, some recent studies reporting evaluation different patient-specific intervals. Recently, researchers have aimed determine regular seizure. Authors been...
Scalp electroencephalogram (EEG) is a non-invasive measure of brain activity. It widely used in several applications including cognitive tasks, sleep stage detection, and seizure prediction. When recorded over hours, this signal usually corrupted by noisy disturbances such as experimental errors, environmental interferences, physiological artifacts. These may generate confounding factors and, therefore, lead to false results. Models able minimise EEG artifacts are then necessary for...
Abstract Electrocardiogram (ECG) recordings, lasting hours before epileptic seizures, have been studied in the search for evidence of existence a preictal interval that follows normal ECG trace and precedes seizure’s clinical manifestation. The has not yet clinically parametrized. Furthermore, duration this varies seizures both among patients from same patient. In study, we performed heart rate variability (HRV) analysis to investigate discriminative power features HRV identification...
Abstract Typical seizure prediction models aim at discriminating interictal brain activity from pre-seizure electrographic patterns. Given the lack of a preictal clinical definition, fixed interval is widely used to develop these models. Recent studies reporting selection among range intervals show inter- and intra-patient variability, possibly reflecting heterogeneity generation process. Obtaining accurate labels can be train supervised and, hence, avoid setting for all seizures within same...
Abstract Almost one-third of epileptic patients fail to achieve seizure control through anti-epileptic drug administration. In the scarcity completely controlling a patient’s epilepsy, prediction plays significant role in clinical management and providing new therapeutic options such as warning or intervention devices. Seizure algorithms aim identify preictal period that Electroencephalogram (EEG) signals can capture. However, this is associated with substantial heterogeneity, varying among...
Abstract According to the literature, seizure prediction models should be developed following a patient-specific approach. However, seizures are usually very rare events, meaning number of events that may used optimise approaches is limited. To overcome such constraint, we analysed possibility using data from patients an external database improve models. We present trained transfer learning procedure. deep convolutional autoencoder electroencephalogram 41 collected EPILEPSIAE database. Then,...
Abstract Background Venous thromboembolism (VTE) is a frequent complication in patients with cancer and causes considerable morbidity mortality. The risk of VTE higher pancreatic often associated treatment delays or interruptions. Recently, the ONKOTEV score was proposed as predictor model for cancer, but its validation still ongoing. Patients Methods We conducted retrospective study to determine incidence evaluate predictive tool population cancer. Results Between February 2012 May 2017,...
Abstract The administration of antiepileptic drugs or surgical interventions fails to control seizures in about 30% patients. Seizure prediction is a viable strategy for enhancing their quality life because it can be used intervention warning systems. These systems may disarm or, at the very least, lessen adverse effects. Identifying preictal interval, which marks change from regular brain activity seizure, critical this research field. Even though several predictive studies applied various...
Abstract Seizure prediction remains a challenge, with approximately 30% of patients unresponsive to conventional treatments. Addressing this issue is crucial for improving patients’ quality life, as timely intervention can mitigate the impact seizures. In research field, it critical identify preictal interval, transition from regular brain activity seizure. While previous studies have explored various Electroencephalogram (EEG) based methodologies prediction, few been clinically applicable....
Having in mind that different languages might present challenges, this paper presents the following contributions to area of Information Extraction from clinical text, targeting Portuguese language: a collection 281 texts language, with manually-annotated named entities; word embeddings trained larger similar texts; results using BiLSTM-CRF neural networks for entity recognition on annotated collection, including comparison in-domain or out-of-domain task. Although learned much less data,...
Guidelines recommend regional lymphadenectomy with a lymph node yield (LNY) of at least 12 nodes (LN) for adequate colon cancer (CC) staging. LNY ≥22LN may improve survival, especially in right-sided CC [Lee et al., Surg Oncol, 27(3), 2018]. This multicentric retrospective cohort study evaluated the impact and tumor laterality on staging survival.Patients stage I-III that underwent surgery from 2012 to 2018 were grouped according LNY: <22 ≥ 22. Primary outcomes LN positivity (N+ rate)...
Objective: Independent component analysis (ICA) is commonly used to remove noisy artifacts from multi-channel scalp electroencephalogram (EEG) signals. ICA decomposes EEG into different independent components (ICs) and then, experts the ones. This process highly time-consuming are not always available. To surpass this drawback, research going on develop models automatically conduct IC classification. Current state-of-the-art use power spectrum densities (PSDs) topoplots classify ICs. The...
<b><i>Introduction:</i></b> Febrile neutropenia (FN) is a potentially life-threatening complication of systemic chemotherapy (CT) that often requires hospital admission. Delay in diagnosis and treatment are associated with higher morbidity mortality. <b><i>Objective:</i></b> We aimed to determine the factors influence FN episodes outcomes emergency room (ER). <b><i>Methods:</i></b> This was retrospective study all (with...
Seizure prediction is a promising solution to improve the quality of life for drug-resistant patients, which concerns nearly 30% patients with epilepsy. The present study aimed ascertain impact incorporating sleep-wake information in seizure prediction.
Abstract Background Liquid biopsy (LB) is a non-invasive tool to evaluate the heterogeneity of tumors. Since RAS mutations (RAS-mut) play major role in resistance antiepidermal growth factor receptor inhibitors (EGFR) monoclonal antibodies (Mabs), serial monitoring RAS-mut with LB may be useful guide treatment. The main aim this study was prognostic value loss (NeoRAS-wt) LB, during treatment metastatic colorectal cancer (mCRC). Methods A retrospective conducted on patients mCRC between...
Abstract Scalp electroencephalogram is a non-invasive multi-channel biosignal that records the brain’s electrical activity. It highly susceptible to noise might overshadow important data. Independent component analysis one of most used artifact removal methods. separates data into different components, although it can not automatically reject noisy ones. Therefore, experts are needed decide which components must be removed before reconstructing To automate this method, researchers have...