Published:
Machine learning in EEG and SEEG can look deceptively straightforward: preprocess the signals, extract features or train a network, and report accuracy. In practice, many pipelines fail because the modeling choices are cleaner than the data assumptions underneath them.
Published:
The central clinical question in drug-resistant focal epilepsy is not only where a seizure begins, but also how it evolves over time. In stereotactic EEG (SEEG), the transition from seizure onset to propagation and then to termination carries information that can help separate the true seizure onset zone from regions that are recruited later.
Published:
One reason I keep returning to changepoint methods in epilepsy research is that the underlying signals are fundamentally temporal. In stereotactic EEG (SEEG), clinically relevant events are not only defined by spatial location, but also by transitions in dynamics: onset, spread, reorganization, and termination.
Published:
Machine learning in EEG and SEEG can look deceptively straightforward: preprocess the signals, extract features or train a network, and report accuracy. In practice, many pipelines fail because the modeling choices are cleaner than the data assumptions underneath them.
Published:
Machine learning in EEG and SEEG can look deceptively straightforward: preprocess the signals, extract features or train a network, and report accuracy. In practice, many pipelines fail because the modeling choices are cleaner than the data assumptions underneath them.
Published:
The central clinical question in drug-resistant focal epilepsy is not only where a seizure begins, but also how it evolves over time. In stereotactic EEG (SEEG), the transition from seizure onset to propagation and then to termination carries information that can help separate the true seizure onset zone from regions that are recruited later.
Published:
One reason I keep returning to changepoint methods in epilepsy research is that the underlying signals are fundamentally temporal. In stereotactic EEG (SEEG), clinically relevant events are not only defined by spatial location, but also by transitions in dynamics: onset, spread, reorganization, and termination.
Published:
Machine learning in EEG and SEEG can look deceptively straightforward: preprocess the signals, extract features or train a network, and report accuracy. In practice, many pipelines fail because the modeling choices are cleaner than the data assumptions underneath them.
Published:
The central clinical question in drug-resistant focal epilepsy is not only where a seizure begins, but also how it evolves over time. In stereotactic EEG (SEEG), the transition from seizure onset to propagation and then to termination carries information that can help separate the true seizure onset zone from regions that are recruited later.
Published:
One reason I keep returning to changepoint methods in epilepsy research is that the underlying signals are fundamentally temporal. In stereotactic EEG (SEEG), clinically relevant events are not only defined by spatial location, but also by transitions in dynamics: onset, spread, reorganization, and termination.