Machine learning model to predict evolution of pulseless electrical activity during in-hospital cardiac arrest.

Background: During pulseless electrical activity (PEA) the cardiac mechanical and electrical functions are dissociated, a phenomenon occurring in 25-42% of in-hospital cardiac arrest (IHCA) cases. Accurate evaluation of the likelihood of a PEA patient transitioning to return of spontaneous circulation (ROSC) may be vital for the successful resuscitation. The aim: We sought to develop a model to automatically discriminate between PEA rhythms with favorable and unfavorable evolution to ROSC. Methods: A dataset of 190 patients, 120 with ROSC, were acquired with defibrillators from different vendors in three hospitals. The ECG and the transthoracic impedance (TTI) signal were processed to compute 16 waveform features. Logistic regression models where designed integrating both automated features and characteristics annotated in the QRS to identify PEAs with better prognosis leading to ROSC. Cross validation techniques were applied, both patient-specific and stratified, to evaluate the performance of the algorithm. Results: The best model consisted in a three feature algorithm that exhibited median (interquartile range) Area Under the Curve/Balanced accuracy/Sensitivity/Specificity of 80.3(9.9)/75.6(8.0)/ 77.4(15.2)/72.3(16.4) %, respectively. Conclusions: Information hidden in the waveforms of the ECG and TTI signals, along with QRS complex features, can predict the progression of PEA. Automated methods as the one proposed in this study, could contribute to assist in the targeted treatment of PEA in IHCA.

Non-shockable rhythms: A parametric model for the immediate probability of return of spontaneous circulation.

BACKGROUND: Cardiac arrest can present with asystole, Pulseless Electrical Activity (PEA), or Ventricular Fibrillation/Tachycardia (VF/VT). We investigated the transition intensity of Return of spontaneous circulation (ROSC) from PEA and asystole during in-hospital resuscitation. MATERIALS AND METHODS: We included 770 episodes of cardiac arrest. PEA was defined as ECG with >12 QRS complexes per min, asystole by an isoelectric signal >5 seconds. The observed times of PEA to ROSC transitions were fitted to five different parametric time-to-event models. At values ≤0.1, transition intensities roughly represent next-minute probabilities allowing for direct interpretation. Different entities of PEA and asystole, dependent on whether it was the primary or a secondary rhythm, were included as covariates. RESULTS: The transition intensities to ROSC from primary PEA and PEA after asystole were unimodal with peaks of 0.12 at 3 min and 0.09 at 6 min, respectively. Transition intensities to ROSC from PEA after VF/VT, or following transient ROSC, exhibited high initial values of 0.32 and 0.26 at 3 minutes, respectively, but decreased. The transition intensity to ROSC from initial asystole and asystole after PEA were both about 0.01 and 0.02; while asystole after VF/VT had an intensity to ROSC of 0.15 initially which decreased. The transition intensity from asystole after temporary ROSC was constant at 0.08. CONCLUSION: The immediate probability of ROSC develops differently in PEA and asystole depending on the preceding rhythm and the duration of the resuscitation attempt. This knowledge may aid simple bedside prognostication and electronic resuscitation algorithms for monitors/defibrillators.

Characteristics and patients' portrayals of Norwegian social media memes. A mixed methods analysis.

Background: Despite reports on troublesome contents created and shared online by healthcare professionals, a systematic inquiry of this potential problem has been missing. Our objective was to characterize the content of healthcare-associated social media memes in terms of common themes and how patients were portrayed. Materials and methods: This study applied a mixed methods approach to characterize the contents of Instagram memes from popular medicine- or nursing-associated accounts in Norway. In total, 2,269 posts from 18 Instagram accounts were included and coded for thematic contents. In addition, we conducted a comprehensive thematic analysis of 30 selected posts directly related to patients. Results: A fifth of all posts (21%) were related to patients, including 139 posts (6%) related to vulnerable patients. Work was, however, the most common theme overall (59%). Nursing-associated accounts posted more patient-related contents than medicine-associated accounts (p < 0.01), but the difference may be partly explained by the former focusing on work life rather than student life. Patient-related posts often thematized (1) trust and breach of trust, (2) difficulties and discomfort at work, and (3) comical aspects of everyday life as a healthcare professional. Discussion: We found that a considerable number of Instagram posts from healthcare-associated accounts included patients and that these posts were diverse in terms of contents and offensiveness. Awareness that professional values also apply online is important for both healthcare students and healthcare providers. Social media memes can act as an educational resource to facilitate discussions about (e-)professionalism, the challenges and coping of everyday life, and ethical conflicts arising in healthcare settings.