How artificial intelligence could transform emergency care.

Artificial intelligence (AI) in healthcare is the ability of a computer to perform tasks typically associated with clinical care (e.g. medical decision-making and documentation). AI will soon be integrated into an increasing number of healthcare applications, including elements of emergency department (ED) care. Here, we describe the basics of AI, various categories of its functions (including machine learning and natural language processing) and review emerging and potential future use-cases for emergency care. For example, AI-assisted symptom checkers could help direct patients to the appropriate setting, models could assist in assigning triage levels, and ambient AI systems could document clinical encounters. AI could also help provide focused summaries of charts, summarize encounters for hand-offs, and create discharge instructions with an appropriate language and reading level. Additional use cases include medical decision making for decision rules, real-time models that predict clinical deterioration or sepsis, and efficient extraction of unstructured data for coding, billing, research, and quality initiatives. We discuss the potential transformative benefits of AI, as well as the concerns regarding its use (e.g. privacy, data accuracy, and the potential for changing the doctor-patient relationship).

A machine learning-based prediction of the micropapillary/solid growth pattern in invasive lung adenocarcinoma with radiomics.

BACKGROUND: Micropapillary/solid (MP/S) growth patterns of lung adenocarcinoma are vital for making clinical decisions regarding surgical intervention. This study aimed to predict the presence of a MP/S component in lung adenocarcinoma using radiomics analysis. METHODS: Between January 2011 and December 2013, patients undergoing curative invasive lung adenocarcinoma resection were included. Using the "PyRadiomics" package, we extracted 90 radiomics features from the preoperative computed tomography (CT) images. Subsequently, four prediction models were built by utilizing conventional machine learning approaches fitting into radiomics analysis: a generalized linear model (GLM), Naïve Bayes, support vector machine (SVM), and random forest classifiers. The models' accuracy was assessed using a receiver operating curve (ROC) analysis, and the models' stability was validated both internally and externally. RESULTS: A total of 268 patients were included as a primary cohort, and 36.6% (98/268) of them had lung adenocarcinoma with an MP/S component. Patients with an MP/S component had a higher rate of lymph node metastasis (18.4% versus 5.3%) and worse recurrence-free and overall survival. Five radiomics features were selected for model building, and in the internal validation, the four models achieved comparable performance of MP/S prediction in terms of area under the curve (AUC): GLM, 0.74 [95% confidence interval (CI): 0.65-0.83]; Naïve Bayes, 0.75 (95% CI: 0.65-0.85); SVM, 0.73 (95% CI: 0.61-0.83); and random forest, 0.72 (95% CI: 0.63-0.81). External validation was performed using a test cohort with 193 patients, and the AUC values were 0.70, 0.72, 0.73, and 0.69 for Naïve Bayes, SVM, random forest, and GLM, respectively. CONCLUSIONS: Radiomics-based machine learning approach is a very strong tool for preoperatively predicting the presence of MP/S growth patterns in lung adenocarcinoma, and can help customize treatment and surveillance strategies.

Updated Cost-Effectiveness of Intravitreal Ocriplasmin for Vitreomacular Adhesion and Macular Hole.

BACKGROUND AND OBJECTIVE: The purpose of this study is to provide an updated assessment of cost-efficacy of intravitreal ocriplasmin (IVO) for vitreomacular adhesion (VMA) and macular holes (MH). PATIENTS AND METHODS: This was a single-center, multiple-physician, institutional review board-approved, retrospective, 15-month cost-effectiveness analysis study (January 2015 to April 2016). Clinical charts and billing records of 247 patients with VMA and MH were reviewed. Patients were divided into group 1 (VMA and MH treated by pars plana vitrectomy [PPV]), group 2 (VMA and MH treated by IVO), and group 3 (VMA treated by IVO). Success rates of interventions in each group were compared, including cost-effectiveness, cost per line-year, and cost per quality-adjusted life-year (QALY). RESULTS: Success rates for initial intervention were 98% in group 1, 55.6% in group 2, and 67.7% in group 3. Cost of PPV at our institution was $6,538.00 and cost of IVO (2016) was $3,480.00. Using a cohort-based computer Markov model, the treatment decision tree demonstrated group 1 was less cost-effective, with cost per line of $2,654.39, cost per line-year saved of $185.62, and cost per QALY of $6,187.00. Group 2 was cost-effective with cost per line of $2,456.25, cost per line-year saved of $171.77, and cost per QALY of $5,726.00. The difference in cost-effectiveness showed IVO was more cost-effective than PPV, with a difference in cost per line of $198.14, cost per line-year saved of $13.85, and cost per QALY of $461.00. CONCLUSIONS: IVO is a more cost-effective intervention than vitrectomy for the treatment of VMA and MH in the setting of judicious use in appropriate patients. The success rate of IVO in our patient population was greater than currently published rates and most certainly impacted probability of cost-efficacy. Further research targeting optimizing IVO success rate is needed. [Ophthalmic Surg Lasers Imaging Retina. 2018;49:e240-e248.].

Insights Into the Molecular Mechanisms of T Follicular Helper-Mediated Immunity and Pathology.

T follicular helper (Tfh) cells play key role in providing help to B cells during germinal center (GC) reactions. Generation of protective antibodies against various infections is an important aspect of Tfh-mediated immune responses and the dysregulation of Tfh cell responses has been implicated in various autoimmune disorders, inflammation, and malignancy. Thus, their differentiation and maintenance must be closely regulated to ensure appropriate help to B cells. The generation and function of Tfh cells is regulated by multiple checkpoints including their early priming stage in T zones and throughout the effector stage of differentiation in GCs. Signaling pathways activated downstream of cytokine and costimulatory receptors as well as consequent activation of subset-specific transcriptional factors are essential steps for Tfh cell generation. Thus, understanding the mechanisms underlying Tfh cell-mediated immunity and pathology will bring into spotlight potential targets for novel therapies. In this review, we discuss the recent findings related to the molecular mechanisms of Tfh cell differentiation and their role in normal immune responses and antibody-mediated diseases.

STAT4 Regulates the CD8+ Regulatory T Cell/T Follicular Helper Cell Axis and Promotes Atherogenesis in Insulin-Resistant Ldlr-/- Mice.

The metabolic syndrome and diabetic conditions support atherosclerosis, but the exact mechanisms for accelerated atherogenesis remain unclear. Although the proinflammatory role of STAT4 in atherosclerosis and diet-induced insulin resistance (IR) was recently established, the impact of STAT4 on atherogenesis in conditions of IR is not known. In this study, we generated Stat4-/-Ldlr-/- mice that were fed a diabetogenic diet with added cholesterol (DDC). DDC-fed Stat4-/-Ldlr-/- mice demonstrated improved glucose tolerance, insulin sensitivity, and a 36% reduction in atherosclerosis compared with Ldlr-/- controls. Interestingly, we detected a reduction in T follicular helper (Tfh) cells and plasma B cells but a sharp elevation in CD8+ regulatory T cells (Tregs) in spleens and aortas of Stat4-/-Ldlr-/- mice compared with Ldlr-/- mice. Similarly, STAT4 deficiency supported CD8+ Treg differentiation in vitro. STAT4-deficient CD8+ Tregs suppressed Tfh cell and germinal center B cell development upon immunization with keyhole limpet hemocyanin, indicating an important role for STAT4 in CD8+ Treg functions in vivo. Furthermore, adoptive transfer of Stat4-/-Ldlr-/- CD8+ Tregs versus Ldlr-/- CD8+ Tregs resulted in a significant reduction in plaque burden and suppression of Tfh cell and germinal center B cells in DDC-fed Ldlr-/- recipients. STAT4 expression in macrophages (MΦs) also affected the Tfh/CD8+ Treg axis, because conditioned media from Stat4-/-Ldlr-/- MΦs supported CD8+ Treg differentiation, but not Tfh cell differentiation, in a TGF-ß-dependent manner. These findings suggest a novel mechanism by which STAT4 supports atherosclerosis in IR Ldlr-/- mice via STAT4-dependent MΦs, as well as cell-intrinsic suppression of CD8+ Treg generation and functions and maintenance of Tfh cell generation and the accompanying humoral immune response.

Comparative quantitative proteomic analysis of disease stratified laser captured microdissected human islets identifies proteins and pathways potentially related to type 1 diabetes.

Type 1 diabetes (T1D) is a chronic inflammatory disease that is characterized by autoimmune destruction of insulin-producing pancreatic beta cells. The goal of this study was to identify novel protein signatures that distinguish Islets from patients with T1D, patients who are autoantibody positive without symptoms of diabetes, and from individuals with no evidence of disease. High resolution high mass accuracy label free quantitative mass spectrometry analysis was applied to islets isolated by laser capture microdissection from disease stratified human pancreata from the Network for Pancreatic Organ Donors with Diabetes (nPOD), these included donors without diabetes, donors with T1D-associated autoantibodies in the absence of diabetes, and donors with T1D. Thirty-nine proteins were found to be differentially regulated in autoantibody positive cases compared to the no-disease group, with 25 upregulated and 14 downregulated proteins. For the T1D cases, 63 proteins were differentially expressed, with 24 upregulated and 39 downregulated, compared to the no disease controls. We have identified functional annotated enriched gene families and multiple protein-protein interaction clusters of proteins are involved in biological and molecular processes that may have a role in T1D. The proteins that are upregulated in T1D cases include S100A9, S100A8, REG1B, REG3A and C9 amongst others. These proteins have important biological functions, such as inflammation, metabolic regulation, and autoimmunity, all of which are pathways linked to the pathogenesis of T1D. The identified proteins may be involved in T1D development and pathogenesis. Our findings of novel proteins uniquely upregulated in T1D pancreas provides impetus for further investigations focusing on their expression profiles in beta cells/ islets to evaluate their role in the disease pathogenesis. Some of these molecules may be novel therapeutic targets T1D.

Atherosclerosis-Driven Treg Plasticity Results in Formation of a Dysfunctional Subset of Plastic IFNγ+ Th1/Tregs.

RATIONALE: Forkhead box P3+ T regulatory cells (Tregs) are key players in maintaining immune homeostasis. Evidence suggests that Tregs respond to environmental cues to permit or suppress inflammation. In atherosclerosis, Th1-driven inflammation affects Treg homeostasis, but the mechanisms governing this phenomenon are unclear. OBJECTIVE: Here, we address whether atherosclerosis impacts Treg plasticity and functionality in Apoe-/- mice, and what effect Treg plasticity might have on the pathology of atherosclerosis. METHODS AND RESULTS: We demonstrate that atherosclerosis promotes Treg plasticity, resulting in the reduction of CXCR3+ Tregs and the accumulation of an intermediate Th1-like interferon (IFN)-γ+CCR5+ Treg subset (Th1/Tregs) within the aorta. Importantly, Th1/Tregs arise in atherosclerosis from bona fide Tregs, rather than from T-effector cells. We show that Th1/Tregs recovered from atherosclerotic mice are dysfunctional in suppression assays. Using an adoptive transfer system and plasticity-prone Mir146a-/- Tregs, we demonstrate that elevated IFNγ+ Mir146a-/- Th1/Tregs are unable to adequately reduce atherosclerosis, arterial Th1, or macrophage content within Apoe-/- mice, in comparison to Mir146a+/+ Tregs. Finally, via single-cell RNA-sequencing and real-time -polymerase chain reaction, we show that Th1/Tregs possess a unique transcriptional phenotype characterized by coexpression of Treg and Th1 lineage genes and a downregulation of Treg-related genes, including Ikzf2, Ikzf4, Tigit, Lilrb4, and Il10. In addition, an ingenuity pathway analysis further implicates IFNγ, IFNα, interleukin-2, interleukin-7, CTLA-4 (cytotoxic T-lymphocyte-associated protein 4), T-cell receptor, and Csnk2b-related pathways in regulating Treg plasticity. CONCLUSIONS: Atherosclerosis drives Treg plasticity, resulting in the accumulation of dysfunctional IFNγ+ Th1/Tregs that may permit further arterial inflammation and atherogenesis.

Smooth Muscle Cell-Derived Interleukin-17C Plays an Atherogenic Role via the Recruitment of Proinflammatory Interleukin-17A+ T Cells to the Aorta.

OBJECTIVE: Atherosclerosis is characterized by frequent communication between infiltrating leukocytes and vascular cells, through chemokine and cytokine networks. Interleukin-17C (IL-17C) is detectable within atherosclerotic lesions; however, the potential involvement of this cytokine has not been examined. Thus, we sought to investigate the role of IL-17C in atherosclerosis. APPROACH AND RESULTS: The expression of IL-17 cytokines was profiled within aortas of apolipoprotein E double knockout (Apoe(-/-)) mice, and Il17c expression was elevated. Flow cytometry experiments revealed a major population of aortic IL-17C-producing smooth muscle cells. Next, we generated Il17c(-/-)Apoe(-/-) mice and demonstrated that atherosclerotic lesion and collagen content was diminished within Western diet-fed Il17c(-/-)Apoe(-/-) aortas and aortic roots in comparison to Apoe(-/-) controls. Smooth muscle cells and fibroblasts were mainly responsible for the reduced Col1A1 expression in the aorta of Il17c(-/-)Apoe(-/-) mice. Importantly, IL-17C-treated Apoe(-/-) aortas showed upregulated Col1A1 expression ex vivo. Il17c(-/-)Apoe(-/-) mice displayed a proportional reduction in aortic macrophages, neutrophils, T cells, T helper 1 cells, and T regulatory cells, without corresponding changes in the peripheral immune composition. Examination of aortic IL-17A(+) T-cell receptor γδ T cells and Th17 cells demonstrated a stark reduction in the percentage and number of these subsets within Il17c(-/-)Apoe(-/-) versus Apoe(-/-) mice. Explanted 12-week Western diet-fed Apoe(-/-) aortas treated with IL-17C resulted in the induction of multiple vascular chemokines and cytokines. Th17 cells demonstrated attenuated migration toward supernatants from cultures of Il17c(-/-)Apoe(-/-) smooth muscle cells, and short-term homing experiments revealed diminished recruitment of Th17 cells to the aorta of Il17c(-/-)Apoe(-/-) recipients. CONCLUSIONS: Smooth muscle cell-derived IL-17C plays a proatherogenic role by supporting the recruitment of Th17 cells to atherosclerotic lesions.

CXCR6 regulates the recruitment of pro-inflammatory IL-17A-producing T cells into atherosclerotic aortas.

The adaptive immune response is involved in the development and progression of atherosclerosis and IL-17A(+) cells play a role in this disease. Although elevated number of CD4(+) IL-17A(+) (Th17) and IL-17A(+)TCRγδ(+) T cells are found within murine atherosclerotic aortas and human plaques, the mechanisms governing IL-17A(+) T-cell migration to atherosclerotic lesions are unclear. The chemokine receptor CXCR6 is expressed on several T-cell subsets and plays a pro-atherogenic role in atherosclerosis. Here, we used CXCR6-deficient (Cxcr6 (GFP/GFP) ) apolipoprotein E-deficient (Apoe (-/-) ) mice to investigate the involvement of CXCR6 in the recruitment IL-17A(+) T cells to atherosclerotic aortas. Flow cytometric analyses revealed reductions in Th17 and IL-17A(+)TCRγδ(+) T cells within aged Cxcr6 (GFP/GFP) Apoe (-/-) aortas, in comparison with age-matched Cxcr6 (GFP/+) Apoe (-/-) aortas. Although CXCR6-sufficient IL-17A(+) T cells efficiently migrated toward CXCL16, the migration of CXCR6-deficient IL-17A(+) T cells was abolished in transwell assays. Importantly, the recruitment of Cxcr6 (GFP/GFP) Apoe (-/-) IL-17A(+) T cells into the aortas of Apoe (-/-) recipients was markedly reduced in short-term adoptive transfer experiments. Altogether these results demonstrate an important role of CXCR6 in the regulation of pathological Th17 and IL-17A(+)TCRγδ(+) T-cell recruitment into atherosclerotic lesions.