Leveraging IoTs and Machine Learning for Patient Diagnosis and Ventilation Management in the Intensive Care Unit.

Future healthcare systems will rely heavily on clinical decision support systems (CDSS) to improve the decision-making processes of clinicians. To explore the design of future CDSS, we developed a research-focused CDSS for the management of patients in the intensive care unit that leverages Internet of Things (IoT) devices capable of collecting streaming physiologic data from ventilators and other medical devices. We then created machine learning (ML) models that could analyze the collected physiologic data to determine if the ventilator was delivering potentially harmful therapy and if a deadly respiratory condition, acute respiratory distress syndrome (ARDS), was present. We also present work to aggregate these models into a mobile application that can provide responsive, real-time alerts of changes in ventilation to providers. As illustrated in the recent COVID-19 pandemic, being able to accurately predict ARDS in newly infected patients can assist in prioritizing care. We show that CDSS may be used to analyze physiologic data for clinical event recognition and automated diagnosis, and we also highlight future research avenues for hospital CDSS.

Mapping novel immunogenic epitopes in IgA nephropathy.

BACKGROUND AND OBJECTIVES: IgA plays a key role in IgA nephropathy (IgAN) by forming immune complexes and depositing in the glomeruli, leading to an inflammatory response. However, the antigenic targets and functional characterization of IgA have been incompletely defined in this disease. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: This study was performed in sera from patients who were studied as part of a prospective, observational study of IgAN. These patients (n=22) all had biopsy-proven IgAN within 3 years of study initiation, complete clinical data, annual urinary inulin clearance for GFRs, and at least 5 years of follow-up. Progression was defined as loss of >5 ml/min per 1.73 m(2) per year of inulin clearance measured over at least 5 years. A protein microarray was used for detection of IgAN-specific IgA autoantibodies in blood across approximately 9000 human antigens to specifically identify the most immunogenic protein targets that drive IgA antibodies in IgAN (n=22), healthy controls (n=10), and non-IgAN glomerular diseases (n=17). Results were validated by ELISA assays in sera and by immunohistochemistry in IgAN kidney biopsies. IgA-specific antibodies were correlated with clinical and histologic variables to assess their effect on disease progression and prognosis. RESULTS: Fifty-four proteins mounted highly significant IgA antibody responses in patients with IgAN with a false discovery rate (q value) of ≤10%; 325 antibodies (P≤0.05) were increased overall. Antitissue transglutaminase IgA was significantly elevated in IgAN (P<0.001, q value of 0%). IgA antibodies to DDX4 (r=-0.55, P=0.01) and ZADH2 (r=-0.48, P=0.02) were significantly correlated with the decline of renal function. Specific IgA autoantibodies are elevated in IgAN compared with normal participants and those with other glomerular diseases. CONCLUSIONS: In this preliminary study, IgA autoantibodies target novel proteins, highly expressed in the kidney glomerulus and tubules. These IgA autoantibodies may play important roles in the pathogenesis of IgAN.

Profiling of autoantibodies in IgA nephropathy, an integrative antibiomics approach.

BACKGROUND AND OBJECTIVES: IgG commonly co-exists with IgA in the glomerular mesangium of patients with IgA nephropathy (IgAN) with unclear clinical relevance. Autoantibody (autoAb) biomarkers to detect and track progression of IgAN are an unmet clinical need. The objective of the study was to identify IgA-specific autoAbs specific to IgAN. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: High-density protein microarrays were evaluated IgG autoAbs in the serum of IgAN patients (n = 22) and controls (n = 10). Clinical parameters, including annual GFR and urine protein measurements, were collected on all patients over 5 years. Bioinformatic data analysis was performed to select targets for further validation by immunohistochemistry (IHC). RESULTS: One hundred seventeen (1.4%) specific antibodies were increased in IgAN. Among the most significant were the autoAb to the Ig family of proteins. IgAN-specific autoAbs (approximately 50%) were mounted against proteins predominantly expressed in glomeruli and tubules, and selected candidates were verified by IHC. Receiver operating characteristic analysis of our study demonstrated that IgG autoAb levels (matriline 2, ubiquitin-conjugating enzyme E2W, DEAD box protein, and protein kinase D1) might be used in combination with 24-hour proteinuria to improve prediction of the progression of IgAN (area under the curve = 0.86, P = 0.02). CONCLUSIONS: IgAN is associated with elevated IgG autoAbs to multiple proteins in the kidney. This first analysis of the repertoire of autoAbs in IgAN identifies novel, immunogenic protein targets that are highly expressed in the kidney glomerulus and tubules that may bear relevance in the pathogenesis and progression of IgAN.