Automatic ARDS surveillance with chest X-ray recognition using convolutional neural networks.

OBJECTIVE: This study aims to design, validate and assess the accuracy a deep learning model capable of differentiation Chest X-Rays between pneumonia, acute respiratory distress syndrome (ARDS) and normal lungs. MATERIALS AND METHODS: A diagnostic performance study was conducted using Chest X-Ray images from adult patients admitted to a medical intensive care unit between January 2003 and November 2014. X-ray images from 15,899 patients were assigned one of three prespecified categories: "ARDS", "Pneumonia", or "Normal". RESULTS: A two-step convolutional neural network (CNN) pipeline was developed and tested to distinguish between the three patterns with sensitivity ranging from 91.8% to 97.8% and specificity ranging from 96.6% to 98.8%. The CNN model was validated with a sensitivity of 96.3% and specificity of 96.6% using a previous dataset of patients with Acute Lung Injury (ALI)/ARDS. DISCUSSION: The results suggest that a deep learning model based on chest x-ray pattern recognition can be a useful tool in distinguishing patients with ARDS from patients with normal lungs, providing faster results than digital surveillance tools based on text reports. CONCLUSION: A CNN-based deep learning model showed clinically significant performance, providing potential for faster ARDS identification. Future research should prospectively evaluate these tools in a clinical setting.

Machine Learning for Automated Classification of Abnormal Lung Sounds Obtained from Public Databases: A Systematic Review.

Pulmonary auscultation is essential for detecting abnormal lung sounds during physical assessments, but its reliability depends on the operator. Machine learning (ML) models offer an alternative by automatically classifying lung sounds. ML models require substantial data, and public databases aim to address this limitation. This systematic review compares characteristics, diagnostic accuracy, concerns, and data sources of existing models in the literature. Papers published from five major databases between 1990 and 2022 were assessed. Quality assessment was accomplished with a modified QUADAS-2 tool. The review encompassed 62 studies utilizing ML models and public-access databases for lung sound classification. Artificial neural networks (ANN) and support vector machines (SVM) were frequently employed in the ML classifiers. The accuracy ranged from 49.43% to 100% for discriminating abnormal sound types and 69.40% to 99.62% for disease class classification. Seventeen public databases were identified, with the ICBHI 2017 database being the most used (66%). The majority of studies exhibited a high risk of bias and concerns related to patient selection and reference standards. Summarizing, ML models can effectively classify abnormal lung sounds using publicly available data sources. Nevertheless, inconsistent reporting and methodologies pose limitations to advancing the field, and therefore, public databases should adhere to standardized recording and labeling procedures.

Evaluation of Digital Health Strategy to Support Clinician-Led Critically Ill Patient Population Management: A Randomized Crossover Study.

To investigate whether a novel acute care multipatient viewer (AMP), created with an understanding of clinician information and process requirements, could reduce time to clinical decision-making among clinicians caring for populations of acutely ill patients compared with a widely used commercial electronic medical record (EMR). DESIGN: Single center randomized crossover study. SETTING: Quaternary care academic hospital. SUBJECTS: Attending and in-training critical care physicians, and advanced practice providers. INTERVENTIONS: AMP. MEASUREMENTS AND MAIN RESULTS: We compared ICU clinician performance in structured clinical task completion using two electronic environments-the standard commercial EMR (Epic) versus the novel AMP in addition to Epic. Twenty subjects (10 pairs of clinicians) participated in the study. During the study session, each participant completed the tasks on two ICUs (7-10 beds each) and eight individual patients. The adjusted time for assessment of the entire ICU and the adjusted total time to task completion were significantly lower using AMP versus standard commercial EMR (-6.11; 95% CI, -7.91 to -4.30 min and -5.38; 95% CI, -7.56 to -3.20 min, respectively; p < 0.001). The adjusted time for assessment of individual patients was similar using both the EMR and AMP (0.73; 95% CI, -0.09 to 1.54 min; p = 0.078). AMP was associated with a significantly lower adjusted task load (National Aeronautics and Space Administration-Task Load Index) among clinicians performing the task versus the standard EMR (22.6; 95% CI, -32.7 to -12.4 points; p < 0.001). There was no statistically significant difference in adjusted total errors when comparing the two environments (0.68; 95% CI, 0.36-1.30; p = 0.078). CONCLUSIONS: When compared with the standard EMR, AMP significantly reduced time to assessment of an entire ICU, total time to clinical task completion, and clinician task load. Additional research is needed to assess the clinicians' performance while using AMP in the live ICU setting.

Pressure Injury Prediction Model Using Advanced Analytics for At-Risk Hospitalized Patients.

OBJECTIVE: Analyzing pressure injury (PI) risk factors is complex because of multiplicity of associated factors and the multidimensional nature of this injury. The main objective of this study was to identify patients at risk of developing PI. METHOD: Prediction performances of multiple popular supervised learning were tested. Together with the typical steps of a machine learning project, steps to prevent bias were carefully conducted, in which analysis of correlation covariance, outlier removal, confounding analysis, and cross-validation were used. RESULT: The most accurate model reached an area under receiver operating characteristic curve of 99.7%. Ten-fold cross-validation was used to ensure that the results were generalizable. Random forest and decision tree had the highest prediction accuracy rates of 98%. Similar accuracy rate was obtained on the validation cohort. CONCLUSIONS: We developed a prediction model using advanced analytics to predict PI in at-risk hospitalized patients. This will help address appropriate interventions before the patients develop a PI.

The Impact of Health Information Technology for Early Detection of Patient Deterioration on Mortality and Length of Stay in the Hospital Acute Care Setting: Systematic Review and Meta-Analysis.

OBJECTIVE: To evaluate the impact of health information technology (HIT) for early detection of patient deterioration on patient mortality and length of stay (LOS) in acute care hospital settings. DATA SOURCES: We searched MEDLINE and Epub Ahead of Print, In-Process & Other Non-Indexed Citations and Daily, Embase, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Scopus from 1990 to January 19, 2021. STUDY SELECTION: We included studies that enrolled patients hospitalized on the floor, in the ICU, or admitted through the emergency department. Eligible studies compared HIT for early detection of patient deterioration with usual care and reported at least one end point of interest: hospital or ICU LOS or mortality at any time point. DATA EXTRACTION: Study data were abstracted by two independent reviewers using a standardized data extraction form. DATA SYNTHESIS: Random-effects meta-analysis was used to pool data. Among the 30 eligible studies, seven were randomized controlled trials (RCTs) and 23 were pre-post studies. Compared with usual care, HIT for early detection of patient deterioration was not associated with a reduction in hospital mortality or LOS in the meta-analyses of RCTs. In the meta-analyses of pre-post studies, HIT interventions demonstrated a significant association with improved hospital mortality for the entire study cohort (odds ratio, 0.78 [95% CI, 0.70-0.87]) and reduced hospital LOS overall. CONCLUSIONS: HIT for early detection of patient deterioration in acute care settings was not significantly associated with improved mortality or LOS in the meta-analyses of RCTs. In the meta-analyses of pre-post studies, HIT was associated with improved hospital mortality and LOS; however, these results should be interpreted with caution. The differences in patient outcomes between the findings of the RCTs and pre-post studies may be secondary to confounding caused by unmeasured improvements in practice and workflow over time.

Implementation and evaluation of sepsis surveillance and decision support in medical ICU and emergency department.

OBJECTIVE: To improve the timely diagnosis and treatment of sepsis many institutions implemented automated sepsis alerts. Poor specificity, time delays, and a lack of actionable information lead to limited adoption by bedside clinicians and no change in practice or clinical outcomes. We aimed to compare sepsis care compliance before and after a multi-year implementation of a sepsis surveillance coupled with decision support in a tertiary care center. DESIGN: Single center before and after study. SETTING: Large academic Medical Intensive Care Unit (MICU) and Emergency Department (ED). POPULATION: Patients 18 years of age or older admitted to *** Hospital MICU and ED from 09/4/2011 to 05/01/2018 with severe sepsis or septic shock. INTERVENTIONS: Electronic medical record-based sepsis surveillance system augmented by clinical decision support and completion feedback. MEASUREMENTS AND MAIN RESULTS: There were 1950 patients admitted to the MICU with the diagnosis of severe sepsis or septic shock during the study period. The baseline characteristics were similar before (N = 854) and after (N = 1096) implementation of sepsis surveillance. The performance of the alert was modest with a sensitivity of 79.9%, specificity of 76.9%, positive predictive value (PPV) 27.9%, and negative predictive value (NPV) 97.2%. There were 3424 unique alerts and 1131 confirmed sepsis patients after the sniffer implementation. During the study period average care bundle compliance was higher; however after taking into account improvements in compliance leading up to the intervention, there was no association between intervention and improved care bundle compliance (Odds ratio: 1.16; 95% CI: 0.71 to 1.89; p-value 0.554). Similarly, the intervention was not associated with improvement in hospital mortality (Odds ratio: 1.55; 95% CI: 0.95 to 2.52; p-value: 0.078). CONCLUSIONS: A sepsis surveillance system incorporating decision support or completion feedback was not associated with improved sepsis care and patient outcomes.

Classification of Respiratory Conditions using Auscultation Sound.

Management of respiratory conditions relies on timely diagnosis and institution of appropriate management. Computerized analysis and classification of breath sounds has a potential to enhance reliability and accuracy of diagnostic modality while making it suitable for remote monitoring, personalized uses, and self-management uses. In this paper, we describe and compare sound recognition models aimed at automatic diagnostic differentiation of healthy persons vs patients with COPD vs patients with pneumonia using deep learning approaches such as Multi-layer Perceptron Classifier (MLPClassifier) and Convolutional Neural Networks (CNN).Clinical Relevance-Healthcare providers and researchers interested in the field of medical sound analysis, specifically automatic detection/classification of auscultation sound and early diagnosis of respiratory conditions may benefit from this paper.

Improvement of constrictive bronchiolitis (bronchiolitis obliterans) after rituximab therapy in 2 patients with primary sjögren syndrome.

Constrictive bronchiolitis is one of the manifestations of small-airway involvement in primary Sjögren syndrome (SS) and is associated with fixed airflow obstruction despite treatment with bronchodilators, macrolides, corticosteroids, and corticosteroid-sparing agents. Reports have shown a beneficial effect of rituximab on interstitial lung disease associated with SS, but the effect of rituximab on constrictive bronchiolitis is unknown. Herein, we present 2 cases of patients with constrictive bronchiolitis associated with SS who experienced symptomatic improvement and stabilization of pulmonary function testing (PFT) after rituximab therapy. Lung function declined in one of the patients when B cells reconstituted, with improved PFT results on re-administration of rituximab. Our case reports suggest that B cells may be involved in the pathogenesis of SS-associated constrictive bronchiolitis. Therapy targeting B cells may therefore be helpful in treating this debilitating and refractory condition. Further research is warranted.

Early vs late oral nutrition in patients with diabetic ketoacidosis admitted to a medical intensive care unit.

BACKGROUND: Diabetic ketoacidosis (DKA) has an associated mortality of 1% to 5%. Upon admission, patients require insulin infusion and close monitoring of electrolyte and blood sugar levels with subsequent transitioning to subcutaneous insulin and oral nutrition. No recommendations exist regarding the appropriate timing for initiation of oral nutrition. AIM: To assess short-term outcomes of oral nutrition initiated within 24 h of patients being admitted to a medical intensive care unit (MICU) for DKA. METHODS: A retrospective observational cohort study was conducted at a single academic medical center. The patient population consisted of adults admitted to the MICU with the diagnosis of DKA. Baseline characteristics and outcomes were compared between patients receiving oral nutrition within (early nutrition group) and after (late nutrition group) the first 24 h of admission. The primary outcome was 28-d mortality. Secondary outcomes included 90-d mortality, MICU and hospital lengths of stay (LOS), and time to resolution of DKA. RESULTS: There were 128 unique admissions to the MICU for DKA with 67 patients receiving early nutrition and 61 receiving late nutrition. The APACHE (Acute Physiology and Chronic Health Evaluation) IV mortality and LOS scores and DKA severity were similar between the groups. No difference in 28- or 90-d mortality was found. Early nutrition was associated with decreased hospital and MICU LOS but not with prolonged DKA resolution, anion gap closure, or greater rate of DKA complications. CONCLUSION: In patients with DKA, early nutrition was associated with a shorter MICU and hospital LOS without increasing the rate of DKA complications.

Management of bioprosthetic cardiac valve thrombosis.

Valvular thrombosis is a known complication of bioprosthetic valve replacement that usually occurs within a year of implantation. Four-dimensional computed tomography is quickly becoming the gold standard to directly visualize thrombus. Current guidelines for prophylaxis include aspirin for all bioprosthetic valve replacements, dual antiplatelet therapy for aortic valves, and anticoagulation for mitral valves. However, new trials have suggested single-agent antiplatelet therapy or anticoagulation for treatment of bioprosthetic valve thrombosis. Three cases are presented that illustrate the use of anticoagulants and new techniques for detecting thrombosis on imaging.

Pulmonary Mucormycosis in a Patient With Uncontrolled Diabetes.

Prompt diagnosis of a patient presenting with rhinocerebral, pulmonary, gastrointestinal, and central nervous system manifestations is critical for treatment of this potentially fatal fungal infection.

Analysis of 24-hour pulse wave velocity in patients with renal transplantation.

The aim of our study was to assess the feasibility of using an approach to 24-hour pulse wave velocity (PWV) analysis similar to ambulatory blood pressure monitoring analysis in the management of patients with renal transplantation. Overall, 41 patients aged between 18 and 55 years who had end-stage renal disease resulting from glomerulopathy were recruited from the kidney transplant waiting list. All the measurements were performed before kidney transplantation and at 1 and 20 weeks after transplantation. The Pulse Time Index of Norm (PTIN) was calculated with the Vasotens® technology for the estimation of the 24-hour PWV, defined as the percentage of the 24-hour period during which the PWV does not exceed 10 m/second. Before kidney transplantation, the mean PTIN in the whole group was 56.3 (standard deviation, 18.4). Then, a week after the renal transplantation, a decrease in the PTIN was observed in most cases, going to 27.6 (standard deviation, 11.1). After 20 weeks, the mean PTIN in the whole group increased again to 52.0 (standard deviation, 23.6). In our study, we found that the persistence of arterial stiffness disturbances after kidney transplantation appears to be relatively predictable. We determined the cutoff value of PTIN that could predict the two states of PTIN: a state of improvement or a state of decline/without change. The cutoff value of PTIN at 45% had a sensitivity of 69%, specificity of 76%, and area under the curve of 0.65. The analysis of variance showed that in the group with an initial PTIN of 45% or higher, the PTIN in the remote period after transplantation changed significantly (P < 0.05), whereas in the group with an initial PTIN lower than 45%, there were no significant changes. Thus, the analysis of 24-hour pulse wave velocity in the management of patients with renal transplantation using PTIN is feasible.