Mauricio Villarroel's research while affiliated with University of Oxford and other places

... Moreover, being supported by additional markers such as isovolumetric contraction time (IVCT) and relaxation time (IVRT), it derives the Myocardial context of the TDW's origin. Currently, the proposed origin of this fiducial point is attributed to blood pressure wave reflection [32]. As result of the afore-mentioned findings, it is evident that the extraction of ejection time from the PPG signal is indeed feasible. ...

... Most of the existing research utilizes the diffuse reflection from the face region to estimate the rPPG signal, with the exception of a few works analyzing the hand [6,17,33,41], arms [50], and thigh [11]. A gap in current rPPG re- Figure 2. The subjects remained seated while collecting data (a). ...

... We have previously shown that kinematic features derived from the data collected by wearable devices placed on the participant's trunk, wrists, and feet, together with machine learning algorithms can be used to track the progression of Progressive Supranuclear Palsy motor symptoms 18 . The current study aims to extend this knowledge, using walking and postural sway features derived from data collected by six IMUs, to identify the earliest signal of motor symptom progression in individuals with PD measured every three months, over a period of 18 months. ...

... In a study of ICU patients, contact PPG with cuff calibration was shown to accurately predict invasive arterial BP using a DL model [25] . A preliminary study for rPPG in the ICU demonstrated a promising correlation (r = 0.49 ± 0.083 standard error) with invasively measured pulse pressure, though BP was not included as an outcome [92] . Real-time HR tracking using rPPG has been shown to be reliable in the post-operative unit [93] , neonatal intensive care unit [94] , and hemodialysis unit [95] . ...

... Poets et al. developed an interesting apnoea score to assess the severity of apnoea, but this score is not validated yet and is mostly a reflection of current clinical practice [57]. The rate of apnoea and hypoxic events can also fluctuate largely within and between patients, and detection can be difficult [51,58]. We therefore chose for a more pragmatic design with less strict inclusion criteria. ...

... Long-and short-term memory (LSTM) networks, dense networks, random forests, and linear regression models were employed to track the trends in BP. Finnegan et al. [15] investigated the presence of circadian rhythm in PAT and its applications in nocturnal BP dip or rise classification, as it is a strong indicator for CVDs. ...

... This fundamental principle underlies the measurement of pulse wave velocity (PWV) to estimate arterial pressure. Pulse transit time (PTT), the time interval required for an electrical pulse to traverse between peripheral points, serves as an indicator of BP variation [23]. Researchers can use numerous approaches according to different wave properties from signals, especially heuristic modeling with regression [24], which generally uses machine learning algorithms for modeling. ...

... Lee et al. (91) developed a new method for the detection of apnoea from the impedance pneumograph by first removing cardiac interference. Our group recently built on this work and developed a new method to identify inter-breath intervals and apnoeas in infants which included an automated classifier to distinguish between periods of true apnoea and signal which is low amplitude due to artefacts or poor electrode placement (93). Using these methods could improve apnoea detection and consequently our understanding of apnoea in preterm infants; 74% of apnoeic events across 276 infants identified using the algorithm of Lee et al. were not documented in clinical notes (94). ...

... More insight into the postoperative course of vital parameters would be especially interesting, as the literature describes disruption of the diurnal rhythm following surgery and anaesthesia [9][10][11][12], and the disturbance of the diurnal timing system might negatively affect an individual's ability to recover after surgery [12]. This hypothesis is supported by a study by Davidson et al., who were the first to study diurnal rhythms in continuous measurements of vital parameters in the ICU, where they described potential prognostic value for patient recovery [13]. New innovations, such as continuous monitoring using wearables, have been increasingly investigated and recent studies have shown the potential for deterioration detection in perioperative care [14][15][16][17]. ...

... These commonplace gadgets are equipped with a wide range of sensors that may be applied to smart health applications. For instance, apps that measure pulse rate using a camera and LED have previously proven to be accurate in this regard [28,29] . Spirometry tests, which play a crucial part in the identification of respiratory disorders, have previously been conducted using pressure applications coupled to mobile phones. ...