Dataset Information

Data analytics and clinical feature ranking of medical records of patients with sepsis.

ABSTRACT:

Background

Sepsis is a life-threatening clinical condition that happens when the patient's body has an excessive reaction to an infection, and should be treated in one hour. Due to the urgency of sepsis, doctors and physicians often do not have enough time to perform laboratory tests and analyses to help them forecast the consequences of the sepsis episode. In this context, machine learning can provide a fast computational prediction of sepsis severity, patient survival, and sequential organ failure by just analyzing the electronic health records of the patients. Also, machine learning can be employed to understand which features in the medical records are more predictive of sepsis severity, of patient survival, and of sequential organ failure in a fast and non-invasive way.

Dataset and methods

In this study, we analyzed a dataset of electronic health records of 364 patients collected between 2014 and 2016. The medical record of each patient has 29 clinical features, and includes a binary value for survival, a binary value for septic shock, and a numerical value for the sequential organ failure assessment (SOFA) score. We disjointly utilized each of these three factors as an independent target, and employed several machine learning methods to predict it (binary classifiers for survival and septic shock, and regression analysis for the SOFA score). Afterwards, we used a data mining approach to identify the most important dataset features in relation to each of the three targets separately, and compared these results with the results achieved through a standard biostatistics approach.

Results and conclusions

Our results showed that machine learning can be employed efficiently to predict septic shock, SOFA score, and survival of patients diagnoses with sepsis, from their electronic health records data. And regarding clinical feature ranking, our results showed that Random Forests feature selection identified several unexpected symptoms and clinical components as relevant for septic shock, SOFA score, and survival. These discoveries can help doctors and physicians in understanding and predicting septic shock. We made the analyzed dataset and our developed software code publicly available online.

SUBMITTER: Chicco D

PROVIDER: S-EPMC7860202 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Data analytics and clinical feature ranking of medical records of patients with sepsis.

Chicco Davide D Oneto Luca L

BioData mining 20210203 1

<h4>Background</h4>Sepsis is a life-threatening clinical condition that happens when the patient's body has an excessive reaction to an infection, and should be treated in one hour. Due to the urgency of sepsis, doctors and physicians often do not have enough time to perform laboratory tests and analyses to help them forecast the consequences of the sepsis episode. In this context, machine learning can provide a fast computational prediction of sepsis severity, patient survival, and sequential o ...[more]

PMID: 33536030

Similar Datasets

Project description:ImportanceDespite the large health burden, reliable data on sepsis epidemiology are lacking; studies using International Statistical Classification of Diseases and Related Health Problems (ICD)-coded hospital discharge diagnosis for sepsis identification suffer from limited sensitivity. Also, ICD data do not allow investigation of underlying pathogens and antimicrobial resistance.ObjectivesTo generate reliable epidemiological estimates by linking data from a population-based database to a reference standard of clinical medical record review.Design, setting, and participantsThis was a retrospective, observational cohort study using a population-based administrative database including all acute care hospitals of the Scania region in Sweden in 2019 and 2020 to identify hospital-treated sepsis cases by ICD codes. From this database, clinical medical records were also selected for review within 6 strata defined by ICD discharge diagnosis (both with and without sepsis diagnosis). Data were analyzed from April to October 2022.Main outcomes and measuresHospital and population incidences of sepsis, case fatality, antimicrobial resistance, and temporal dynamics due to COVID-19 were assessed, as well as validity of ICD-10 case identification methods compared with the reference standard of clinical medical record review.ResultsOut of 295 531 hospitalizations in 2019 in the Scania region of Sweden, 997 patient medical records were reviewed, among which 457 had sepsis according to clinical criteria. Of the patients with clinical sepsis, 232 (51%) were female, and 357 (78%) had at least 1 comorbidity. The median (IQR) age of the cohort was 76 (67-85) years. The incidence of sepsis in hospitalized patients according to the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) criteria in 2019 was 4.1% (95% CI, 3.6-4.5) by medical record review. This corresponds to an annual incidence rate of 747 (95% CI, 663-832) patients with sepsis per 100 000 population. No significant increase in sepsis during the COVID-19 pandemic nor a decrease in sepsis incidence when excluding COVID-19 sepsis was observed. Few sepsis cases caused by pathogens with antimicrobial resistance were found. The validity of ICD-10-based case identification in administrative data was low.Conclusions and relevanceIn this cohort study of sepsis epidemiology, sepsis was a considerable burden to public health in Sweden. Supplying administrative data with information from clinical medical records can help to generate reliable data on sepsis epidemiology.

Project description:BackgroundPersonal electronic health records (PEHRs) allow patients to view, generate, and manage their personal and medical data that are relevant across illness episodes, such as their medications, allergies, immunizations, and their medical, social, and family health history. Thus, patients can actively participate in the management of their health care by ensuring that their health care providers have an updated and accurate overview of the patients' medical records. However, the uptake of PEHRs remains low, especially in terms of patients entering and managing their personal and medical data in their PEHR.ObjectiveThis scoping review aimed to explore the barriers and facilitators that patients face when deciding to review, enter, update, or modify their personal and medical data in their PEHR. This review also explores the extent to which patient-generated and -managed data affect the quality and safety of care, patient engagement, patient satisfaction, and patients' health and health care services.MethodsWe searched the MEDLINE, Embase, CINAHL, PsycINFO, Cochrane Library, Web of Science, and Google Scholar web-based databases, as well as reference lists of all primary and review articles using a predefined search query.ResultsOf the 182 eligible papers, 37 (20%) provided sufficient information about patients' data management activities. The results showed that patients tend to use their PEHRs passively rather than actively. Patients refrain from generating and managing their medical data in a PEHR, especially when these data are complex and sensitive. The reasons for patients' passive data management behavior were related to their concerns about the validity, applicability, and confidentiality of patient-generated data. Our synthesis also showed that patient-generated and -managed health data ensures that the medical record is complete and up to date and is positively associated with patient engagement and patient satisfaction.ConclusionsThe findings of this study suggest recommendations for implementing design features within the PEHR and the construal of a dedicated policy to inform both clinical staff and patients about the added value of patient-generated data. Moreover, clinicians should be involved as important ambassadors in informing, reminding, and encouraging patients to manage the data in their PEHR.

Project description:BackgroundDeep learning models have attracted significant interest from health care researchers during the last few decades. There have been many studies that apply deep learning to medical applications and achieve promising results. However, there are three limitations to the existing models: (1) most clinicians are unable to interpret the results from the existing models, (2) existing models cannot incorporate complicated medical domain knowledge (eg, a disease causes another disease), and (3) most existing models lack visual exploration and interaction. Both the electronic health record (EHR) data set and the deep model results are complex and abstract, which impedes clinicians from exploring and communicating with the model directly.ObjectiveThe objective of this study is to develop an interpretable and accurate risk prediction model as well as an interactive clinical prediction system to support EHR data exploration, knowledge graph demonstration, and model interpretation.MethodsA domain-knowledge-guided recurrent neural network (DG-RNN) model is proposed to predict clinical risks. The model takes medical event sequences as input and incorporates medical domain knowledge by attending to a subgraph of the whole medical knowledge graph. A global pooling operation and a fully connected layer are used to output the clinical outcomes. The middle results and the parameters of the fully connected layer are helpful in identifying which medical events cause clinical risks. DG-Viz is also designed to support EHR data exploration, knowledge graph demonstration, and model interpretation.ResultsWe conducted both risk prediction experiments and a case study on a real-world data set. A total of 554 patients with heart failure and 1662 control patients without heart failure were selected from the data set. The experimental results show that the proposed DG-RNN outperforms the state-of-the-art approaches by approximately 1.5%. The case study demonstrates how our medical physician collaborator can effectively explore the data and interpret the prediction results using DG-Viz.ConclusionsIn this study, we present DG-Viz, an interactive clinical prediction system, which brings together the power of deep learning (ie, a DG-RNN-based model) and visual analytics to predict clinical risks and visually interpret the EHR prediction results. Experimental results and a case study on heart failure risk prediction tasks demonstrate the effectiveness and usefulness of the DG-Viz system. This study will pave the way for interactive, interpretable, and accurate clinical risk predictions.

Dataset Information

Data analytics and clinical feature ranking of medical records of patients with sepsis.

Background

Dataset and methods

Results and conclusions

Publications

Data analytics and clinical feature ranking of medical records of patients with sepsis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets