The PhD defence and trial lecture are fully digital and streamed using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.
Ex auditorio questions: the chair of the defence will invite the audience to ask ex auditorio questions. This can be requested by clicking 'Participants -> Raise hand'.
Deep learning for medical data analysis: Opportunities and challenges
Main research findings
Invasive monitoring of intracranial pressure (ICP) plays an essential role in diagnostics and surveillance of patients with brain injury and brain disease. While tracking of mean ICP is a cornerstone of cerebrovascular monitoring worldwide, an increasing body of research highlights the added clinical benefit of also considering the cardiac-induced pressure variations in the ICP signal, namely the pulsatile ICP.
Current measurement modalities of mean and pulsatile ICP are all invasive with associated risks of intracranial bleeds and infections. The primary aim of the presented research was, therefore, to investigate avenues for non-invasive pulsatile ICP estimation in order to safely provide the clinicians with crucial information about the intracranial condition.
Non-invasive pulsatile ICP estimation has been performed in two different studies. The first study utilizes the central aortic blood pressure waveforms and a statistical model to estimate pulsatile ICP. The second study uses the pressure oscillations measured in the outer ear as a source signal. The studies both gave reasonably good visual estimates of the ICP morphology. The majority of patients investigated in the presented research suffer from a disorder not sufficiently understood. The pulsatile ICP’s dependence on hemodynamic events was, therefore, explored in a third study.
Contact information to Department: Pernille Adine Nordby