3 results for Cumin, David, Dataset

  • Intraoperative physiology recorded for investigating methods of diagnosis by an expert system

    Harrison, Michael; Cumin, David (2015-04-29)

    Dataset
    The University of Auckland Library

    Data in this collection was originally collected, with local ethics committee and patient consent, at 0.1 Hz from the Datex-Ohmeda S5 anaesthetic monitor via a serial connection using an ‘in-house’ data-collection program (a modification of that A.Lowe, which is incorporated into EBMi software - http://www.custos.co.nz). The data collected were converted to text files. The EBMi software generates an assessment of the state of blood volume, whether there was sympathetic activity or a fall in cardiac output, amongst other indices and were stored separate files. This data was collected for the purpose of investigating methods of enhancing intra-operative diagnoses by an expert system (http://dx.doi.org/10.1016/j.compbiomed.2014.12.017). An R script was used to merge and convert these files into the XML schema. With the exception of heart rate, pulse rate and respiratory rate all the values in the raw data are 100x the true values.

    View record details
  • Physiological changes during anaesthesia for surgery with potential for moderate blood loss

    Harrison, Michael; Cumin, David (2012)

    Dataset
    The University of Auckland Library

    This dataset is made available under the Open Database License: http://opendatacommons.org/licenses/odbl/1.0/. Any rights in individual contents of the dataset are licensed under the Database Contents License: http://opendatacommons.org/licenses/dbcl/1.0/ Data in this collection was originally collected, with local ethics committee and patient consent, at 0.1 Hz from the Datex-Ohmeda S5 anaesthetic monitor via a serial connection using an ‘in-house’ data-collection program (A.Lowe). The data collected were converted to text files and stored in a Microsoft Excel™ file. The concurrent timed comments from clinicians included an assessment of the state of blood volume, whether there was sympathetic activity or a fall in cardiac output, amongst others, were collected in a separate file. This data was collected for the purpose of investigating methods of enhancing intra-operative diagnoses by an expert system [41]. A VBA (v6.5, Microsoft) script was used to merge and convert these files into the XML schema. Note that blood pressure values are 100 x the true value.

    View record details
  • Anaesthetic Shoulder Arthroscopy Cases (ASAC) Dataset

    Cumin, David (2009)

    Dataset
    The University of Auckland Library

    This data is made available under the Open Database License: http://opendatacommons.org/licenses/odbl/1.0/. Any rights in individual contents of the database are licensed under the Database Contents License: http://opendatacommons.org/licenses/dbcl/1.0/ Manuscript accepted to JAMIA, http://dx.doi.org/10.1136/amiajnl-2012-001087 1.1. PURPOSE There is currently a lack of high quality anaesthetic records from operations which can be used for research purposes (especially to do with physiologic signals). The purpose of this database is to allow an easily accessible resource for researchers and clinicians to deposit and/or obtain complete clinical recordings. 1.2. DATA Ethics approval was obtained to collect and make accessible the data in this database. Data was collected using the SAFERSleep system [1] set to the highest data sampling rate (0.2Hz). Recordings include patient data, time-stamped medical events/interventions (e.g. drug, gas, vapour administration, repositioning of patient etc.) and physiologic data (e.g. heart rate, oxygen saturation etc.). The database contains 20 high quality anaesthetic cases made up of 13 male and 7 female patients aged between 21 and 70 years of age and weighing between 57 and 110kg undergoing shoulder arthroscopy operations. Each patient had between 17 and 26 measured variables over their proceedure, between 5 and 24 events, and between 18 and 58 drug administrations. In total, there are 400 signals over approximately 20 hours with 274 events and 597 drug administrations. Each case is a separate XML file and is easily navigated (Figure 0 1). The time series (physiologic) data is contained as comma-separated values (CSV) within the XML format. Combining the CSV and the XML makes the data easily read and allows for ease of creation should other researchers wish to add cases.

    View record details