A novel artificial intelligence based intensive care unit monitoring system: using physiological waveforms to identify sepsis

2021 ◽  
Vol 379 (2212) ◽  
Author(s):  
Maximiliano Mollura ◽  
Li-Wei H. Lehman ◽  
Roger G. Mark ◽  
Riccardo Barbieri
Keyword(s):  
Artificial Intelligence ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Vital Sign ◽  
Relevant Information ◽  
Great Opportunity ◽  
Diagnosis And Prognosis ◽  
Challenges And Opportunities ◽  
Multivariate Modelling ◽  
Decision Making Processes

A massive amount of multimodal data are continuously collected in the intensive care unit (ICU) along each patient stay, offering a great opportunity for the development of smart monitoring devices based on artificial intelligence (AI). The two main sources of relevant information collected in the ICU are the electronic health records (EHRs) and vital sign waveforms continuously recorded at the bedside. While EHRs are already widely processed by AI algorithms for prompt diagnosis and prognosis, AI-based assessments of the patients’ pathophysiological state using waveforms are less developed, and their use is still limited to real-time monitoring for basic visual vital sign feedback at the bedside. This study uses data from the MIMIC-III database (PhysioNet) to propose a novel AI approach in ICU patient monitoring that incorporates features estimated by a closed-loop cardiovascular model, with the specific goal of identifying sepsis within the first hour of admission. Our top benchmark results (AUROC = 0.92, AUPRC = 0.90) suggest that features derived by cardiovascular control models may play a key role in identifying sepsis, by continuous monitoring performed through advanced multivariate modelling of vital sign waveforms. This work lays foundations for a deeper data integration paradigm which will help clinicians in their decision-making processes. This article is part of the theme issue ‘Advanced computation in cardiovascular physiology: new challenges and opportunities’.

scholarly journals Inter-professional collaboration in an Intensive Care Unit: Challenges and opportunities

Rev Rene ◽  
2016 ◽  
Vol 17 (1) ◽  
pp. 10
Author(s):  
Eveline Rodrigues da Silva Barros ◽  
Ana Ecilda Lima Ellery
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Health Workers ◽  
Quality Care ◽  
Team Leadership ◽  
Professional Collaboration ◽  
Organizational Boundaries ◽  
Challenges And Opportunities ◽  
Provide Quality Care ◽  
The Times

To understand the relationship between health professionals in an intensive care unit, to explore the inter-professional collaboration. Methods: it is a qualitative study, inspired by the Hermeneutics Phenomenology of Paul Ricoeur, for the production of knowledge. Interviews were conducted with 36 intensive care professionals of a tertiary public hospital. Results: the professionals are satisfied with the work, and there is a commitment to provide quality care despite organizational boundaries such as precarious employment relationships and turnover of professionals. The inter-professional collaboration is an indispensable factor for assistance, but in practice is not effective most of the times by the absence of provisions for the integration of the team, leadership presence, as well as the overcrowding of services that overwhelm health workers. Conclusion: while recognizing the need for inter-professional collaboration, professionals do their work even in a very individualized way, with no strategies to boost this cooperation.

Predicting Mortality in the Surgical Intensive Care Unit Using Artificial Intelligence and Natural Language Processing of Physician Documentation

2018 ◽  
Vol 84 (7) ◽  
pp. 1190-1194 ◽  
Author(s):  
Joshua Parreco ◽  
Antonio Hidalgo ◽  
Robert Kozol ◽  
Nicholas Namias ◽  
Rishi Rattan
Keyword(s):  
Artificial Intelligence ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Severity Of Illness ◽  
Surgical Intensive Care Unit ◽  
Surgical Intensive Care ◽  
Severity Of Injury

The purpose of this study was to use natural language processing of physician documentation to predict mortality in patients admitted to the surgical intensive care unit (SICU). The Multiparameter Intelligent Monitoring in Intensive Care III database was used to obtain SICU stays with six different severity of illness scores. Natural language processing was performed on the physician notes. Classifiers for predicting mortality were created. One classifier used only the physician notes, one used only the severity of illness scores, and one used the physician notes with severity of injury scores. There were 3838 SICU stays identified during the study period and 5.4 per cent ended with mortality. The classifier trained with physician notes with severity of injury scores performed with the highest area under the curve (0.88 ± 0.05) and accuracy (94.6 ± 1.1%). The most important variable was the Oxford Acute Severity of Illness Score (16.0%). The most important terms were “dilated” (4.3%) and “hemorrhage” (3.7%). This study demonstrates the novel use of artificial intelligence to process physician documentation to predict mortality in the SICU. The classifiers were able to detect the subtle nuances in physician vernacular that predict mortality. These nuances provided improved performance in predicting mortality over physiologic parameters alone.

Antibiotic stewardship in the intensive care unit: Challenges and opportunities

2019 ◽  
Vol 40 (6) ◽  
pp. 693-698 ◽  
Author(s):  
Kathleen Chiotos ◽  
Pranita D. Tamma ◽  
Jeffrey S. Gerber
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Antibiotic Stewardship ◽  
Audit And Feedback ◽  
Molecular Diagnostic ◽  
Future Research ◽  
Antibiotic Overuse ◽  
Challenges And Opportunities ◽  
Prospective Audit And Feedback ◽  
The Impact

AbstractInfections due to antibiotic-resistant organisms are increasing in prevalence and represent a major public health threat. Antibiotic overuse is a major driver of this epidemic, and antibiotic stewardship an important means of limiting antibiotic resistance. The intensive care unit (ICU) setting presents an intersection of opportunities and challenges for effective antibiotic stewardship, but limited data inform optimal stewardship interventions in this setting. In this review, we present unique considerations for stewardship interventions the ICU setting and summarize available data evaluating the impact of prospective audit and feedback, diagnostic test stewardship, rapid molecular diagnostic tests, and procalcitonin-guided algorithms for antibiotic discontinuation. The existing knowledge gaps ripe for future research are emphasized.

scholarly journals Introduction of the identification, situation, background, assessment, recommendations tool to improve the quality of information transfer during medical handover in intensive care

2016 ◽  
Vol 18 (1) ◽  
pp. 17-23 ◽  
Author(s):  
Benjamin Ramasubbu ◽  
Emma Stewart ◽  
Rosalba Spiritoso
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Patient Information ◽  
Information Transfer ◽  
Relevant Information ◽  
Interquartile Range ◽  
Quality And Safety ◽  
Significant Difference ◽  
Handover Process

Objective To audit the quality and safety of the current doctor-to-doctor handover of patient information in our Cardiothoracic Intensive Care Unit. If deficient, to implement a validated handover tool to improve the quality of the handover process. Methods In Cycle 1 we observed the verbal handover and reviewed the written handover information transferred for 50 consecutive patients in St George’s Hospital Cardiothoracic Intensive Care Unit. For each patient’s handover, we assessed whether each section of the Identification, Situation, Background, Assessment, Recommendations tool was used on a scale of 0–2. Zero if no information in that category was transferred, one if the information was partially transferred and two if all relevant information was transferred. Each patient’s handover received a score from 0 to 10 and thus, each cycle a total score of 0–500. Following the implementation of the Identification, Situation, Background, Assessment, Recommendations handover tool in our Intensive Care Unit in Cycle 2, we re-observed the handover process for another 50 consecutive patients hence, completing the audit cycle. Results There was a significant difference between the total scores from Cycle 1 and 2 (263/500 versus 457/500, p < 0.001). The median handover score for Cycle 1 was 5/10 (interquartile range 4–6). The median handover score for Cycle 2 was 9/10 (interquartile range 9–10). Patient handover scores increased significantly between Cycle 1 and 2, U = 13.5, p < 0.001. Conclusions The introduction of a standardised handover template (Identification, Situation, Background, Assessment, Recommendations tool) has improved the quality and safety of the doctor-to-doctor handover of patient information in our Intensive Care Unit.

Monitoring of Potential Safety Events and Vital Signs during Active Mobilization of Patients Undergoing Continuous Renal Replacement Therapy in a Medical Intensive Care Unit

2016 ◽  
Vol 42 (1) ◽  
pp. 83-90 ◽  
Author(s):  
Hyun Lee ◽  
Young Jun Ko ◽  
Jinhee Jung ◽  
Aeng Ja Choi ◽  
Gee Young Suh ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Renal Replacement Therapy ◽  
Continuous Renal Replacement Therapy ◽  
Vital Sign ◽  
Medical Intensive Care Unit ◽  
Sign Changes ◽  
Potential Safety ◽  
Active Mobilization ◽  
Medical Intensive Care

Background/Aims: This study aims to evaluate potential safety events and vital sign changes during active mobilization physical therapy (PT) in critically ill patients undergoing continuous renal replacement therapy (CRRT). Methods: A retrospective review was performed on 29 patients who were treated with CRRT and who underwent 81 PT sessions in a medical intensive care unit at a single referral hospital; 15 patients underwent 33 sessions with passive range of motion (PROM) and 17 patients underwent 48 active mobilization PT sessions. Three patients received both types of PT including 8 PROM and 5 active mobilization PT sessions. The occurrences of safety events and vital sign changes during active mobilization PT sessions were evaluated. Results: The safety events did not develop during 33 sessions with PROM. However, there were 2 safety events (4.1%) during 48 active mobilization PT sessions including one session with mobilization in the bed and the other in a sitting position on the edge of the bed. These safety events exclusively developed during active mobilization PT sessions, in which concomitant extracorporeal membrane oxygenation (ECMO) support and CRRT were delivered. Regarding vital sign changes during PT sessions, there were no significant differences in systolic blood pressure (BP), diastolic BP, mean arterial pressure, heart rate, respiratory rate, or peripheral oxygen saturation before and after both PROM and active mobilization PT sessions. Conclusions: This study showed that active mobilization PT can be performed safely in patients who are being treated with CRRT without a significant hemodynamic change. However, the development of potential safety events in patients with ECMO needs to be monitored carefully.

scholarly journals Artificial Intelligence Algorithm for Predicting Cardiac Arrest Using Electrocardiography

2020 ◽  
Author(s):  
Joon-myoung Kwon ◽  
Kyung-Hee Kim ◽  
Ki-Hyun Jeon ◽  
Soo Youn Lee ◽  
Jinsik Park ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Intensive Care Unit ◽  
Cardiac Arrest ◽  
Intensive Care ◽  
External Validation ◽  
High Risk Group ◽  
Validation Data ◽  
Intelligence Algorithm ◽  
Artificial Intelligence Algorithm ◽  
Intensive Care Unit Transfer

Abstract Background: In-hospital cardiac arrest is a major burden in health care. Although several track-and-trigger systems are used to predict cardiac arrest, they often have unsatisfactory performances. We hypothesized that a deep-learning-based artificial intelligence algorithm (DLA) could effectively predict cardiac arrest using electrocardiography (ECG). We developed and validated a DLA for predicting cardiac arrest using ECG. Methods: We conducted a retrospective study that included 47,505 ECGs of 25,672 adult patients admitted to two hospitals, who underwent at least one ECG from October 2016 to September 2019. The endpoint was occurrence of cardiac arrest within 24 hours from ECG. Using subgroup analyses in patients who were initially classified as non-event, we confirmed the delayed occurrence of cardiac arrest and unexpected intensive care unit transfer over 14 days.Results: We used 32,294 ECGs of 10,461 patients and 4,483 ECGs of 4,483 patients from a hospital were used as development and internal validation data, respectively. Additionally, 10,728 ECGs of 10,728 patients from another hospital were used as external validation data, which confirmed the robustness of the developed DLA. During internal and external validation, the areas under the receiver operating characteristic curves of the DLA in predicting cardiac arrest within 24 hours were 0.913 and 0.948, respectively. The high risk group of the DLA showed a significantly higher hazard for delayed cardiac arrest (5.74% vs. 0.33%, P < 0.001) and unexpected intensive care unit transfer (4.23% vs. 0.82%, P < 0.001). A sensitivity map of the DLA displayed the ECG regions used to predict cardiac arrest, with the DLA focused most on the QRS complex. Conclusions: Our DLA successfully predicted cardiac arrest using diverse formats of ECG. The results indicate that cardiac arrest could be screened and predicted not only with a conventional 12-lead ECG, but also with a single-lead ECG using a wearable device that employs our DLA.

Sign in / Sign up

Export Citation Format

Share Document