Intraoperative vital signs such as electrocardiography, blood pressure, percutaneous oxygen saturation, and body temperature are objective measures of physiologic function and are tracked with high-acuity patient monitors during surgery and anesthesiaJaipur Stock. These vital signs are usually used as-is, but sometimes converted into clinically useful secondary parameters developed through mathematical, engineering, and medical algorithms. Modern anesthesia widely adopts advanced patient monitors that present a variety of secondary parameters such as electroencephalogram-based anesthesia depth index, arterial pressure-derived cardiac output, and electrocardiography and photoplethysmography-based analgesia index. Numerous studies have shown that these secondary parameters are useful for optimizing patient care during surgery and greatly improve postoperative outcomes1,2,3.
Recent advances in machine learning technologies such as one-dimensional convolutional neural network allowed more accurate interpretation of the complex time-series biosignals4. The relationship between various vital signs was also elucidated using artificial intelligence resulting in practical high-performance algorithms in the medical field5,6. However, the lack of large-scale, high-resolution biosignal data required for machine learning has been a major obstacle to the development or improvement of biosignal algorithmsGuoabong Investment. Electronic medical records (EMR) systems and automated anesthesia records (AAR) are important sources of biosignal big datasets, however, they have limited capabilities because (1) most EMR systems and AARs only store low time resolution data that are insufficient for interpretation of dynamic physiological changes during surgery; (2) essential waveform data such as electrocardiography, photoplethysmography, electroencephalography, and airway pressure waves are not stored on most systems due to cost or technical limitations, and (3) current recording systems do not fully support integrated recording of data from multiple devices7,8. In general, obtaining high-quality vital signs data in surgical patients is considered technically difficult or very expensive.
Previously, we developed the Vital Recorder program, a data capture software that records time-synchronized high-resolution data from various anesthesia devices including patient monitors, anesthesia machines, brain monitors, cardiac monitors, target-controlled infusion pumps, and rapid infusion system9. All parameters of multiple monitoring devices applied simultaneously to one patient are recorded as time-synchronized data tracks and stored as a single case file. Automatic recording function of this program has enabled massive collection of intraoperative biosignals in our tertiary, university hospital. The Vital signs DataBase (VitalDB) was constructed using (1) de-identified case files that were automatically recorded by the Vital Recorder program during daily surgery and anesthesia, and (2) perioperative patient information retrieved from our EMR system.
Unlike the previously reported public multi-parameter biosignal datasets10,11,12, the VitalDB is the first public biosignal dataset specifically focused on perioperative patient care and is characterized by containing multi-parameter high-resolution waveform and numeric data13. Since the VitalDB dataset was first released in 2017, it has been used for various big data research such as: deep learning for arterial pressure waveform-based cardiac output algorithm, deep learning-based pharmacokinetic-pharmacodynamic study of intravenous anesthetics, machine learning for bispectral index algorithm, statistical analysis of the relationship between intraoperative bispectral index and postoperative mortality, and deep learning algorithm to predict intraoperative hypotension from arterial waveforms14,15,16,17,18.
Perioperative clinical information, laboratory results and surgical outcomes in this dataset may facilitate a variety of clinical outcomes or clinical decision support studiesChennai Investment. Studies that elucidate the relationship between biosignal parameters and clinical variables will also be feasible. For instance, the effects of intraoperative variables such as hypotension, hypothermia, and low cardiac output on clinical outcomes such as acute kidney injury, the length of hospital stay, or in-hospital mortality can be examined. The physiologic effects of various interventions such as vasoactive drugs, fluids, anesthetics, and anesthesia machine settings may be sought from the dataset. This dataset may simply be used as data samples for developing signal processing algorithms. However, we argue that this big data is better suited for a training dataset for machine learning of biosignals or for external validation of biosignal algorithms created using other datasets.
Surat Investment