Selective prediction for extracting unstructured clinical data.
Journal of the American Medical Informatics Association : JAMIA
While there are currently approaches to handle unstructured clinical data, such as manual abstraction and structured proxy variables, these methods may be time-consuming, not scalable, and imprecise. This article aims to determine whether selective prediction, which gives a model the option to abstain from generating a prediction, can improve the accuracy and efficiency of unstructured clinical data abstraction.We trained selective classifiers (logistic regression, random forest, support vector machine) to extract 5 variables from clinical notes: depression (n = 1563), glioblastoma (GBM, n = 659), rectal adenocarcinoma (DRA, n = 601), and abdominoperineal resection (APR, n = 601) and low anterior resection (LAR, n = 601) of adenocarcinoma. We varied the cost of false positives (FP), false negatives (FN), and abstained notes and measured total misclassification cost.The depression selective classifiers abstained on anywhere from 0% to 97% of notes, and the change in total misclassification cost ranged from -58% to 9%. Selective classifiers abstained on 5%-43% of notes across the GBM and colorectal cancer models. The GBM selective classifier abstained on 43% of notes, which led to improvements in sensitivity (0.94 to 0.96), specificity (0.79 to 0.96), PPV (0.89 to 0.98), and NPV (0.88 to 0.91) when compared to a non-selective classifier and when compared to structured proxy variables.We showed that selective classifiers outperformed both non-selective classifiers and structured proxy variables for extracting data from unstructured clinical notes.Selective prediction should be considered when abstaining is preferable to making an incorrect prediction.
View details for DOI 10.1093/jamia/ocad182
View details for PubMedID 37769323
Translational Models in Glioma Immunotherapy Research.
Current oncology (Toronto, Ont.)
2023; 30 (6): 5704-5718
Immunotherapy is a promising therapeutic domain for the treatment of gliomas. However, clinical trials of various immunotherapeutic modalities have not yielded significant improvements in patient survival. Preclinical models for glioma research should faithfully represent clinically observed features regarding glioma behavior, mutational load, tumor interactions with stromal cells, and immunosuppressive mechanisms. In this review, we dive into the common preclinical models used in glioma immunology, discuss their advantages and disadvantages, and highlight examples of their utilization in translational research.
View details for DOI 10.3390/curroncol30060428
View details for PubMedID 37366911
Opioid usage in lumbar disc herniation patients with nonsurgical, early, and late surgical treatments.
Assess opioid usage in surgical and non-surgical patients with lumbar disc herniation receiving different treatment approaches and timing.Individuals with newly diagnosed lumbar intervertebral disc without myelopathy were queried from Optum Clinformatics DataMart. Patients were categorized into 3 cohorts: nonsurgical, early surgery, and late surgery. Early surgery cohort patients had surgery within 30-days post-diagnosis; late surgery cohort patients had surgery after 30 days but before 1-year post-diagnosis. The index date was defined as the diagnosis date for nonsurgical patients, and the initial surgery date for surgical patients. The primary outcome was the average daily opioid morphine milligram equivalent (MME) prescribed. Additional outcomes included the percentage of opioid-using patients and cumulative opioid burden.A total of 573,082 patients met inclusion criteria: 533,226 patients received nonsurgical treatments, 22,312 patients received early surgery, and 17,544 patients received late surgery. Both surgical cohorts experienced a "post-surgical hump" of opioid usage, which then sharply declined and gradually plateaued, with daily opioid MME consistently lower in the early as opposed to late surgery cohort. The early surgery cohort also consistently had a lower prevalence of opioid-using patients than the late surgery cohort. Patients receiving nonsurgical demonstrated the highest one-year post-index cumulative opioid burden, and the early surgery cohort consistently had lower cumulative opioid MME than the late surgery cohort.Early surgery in lumbar disc herniation patients is associated with lower long-term average daily MME, incidence of opioid use, and one-year cumulative MME burden compared to nonsurgical and late surgery treatment approaches.
View details for DOI 10.1016/j.wneu.2023.02.029
View details for PubMedID 36775237
Brain injury in COVID-19 is associated with dysregulated innate and adaptive immune responses.
Brain : a journal of neurology
COVID-19 is associated with neurological complications including stroke, delirium and encephalitis. Furthermore, a post-viral syndrome dominated by neuropsychiatric symptoms is common, and is seemingly unrelated to COVID-19 severity. The true frequency and underlying mechanisms of neurological injury are unknown, but exaggerated host inflammatory responses appear to be a key driver of COVID-19 severity. We investigated the dynamics of, and relationship between, serum markers of brain injury (neurofilament light [NfL], glial fibrillary acidic protein [GFAP] and total tau) and markers of dysregulated host response (autoantibody production and cytokine profiles) in 175 patients admitted with COVID-19 and 45 patients with influenza. During hospitalisation, sera from patients with COVID-19 demonstrated elevations of NfL and GFAP in a severity-dependent manner, with evidence of ongoing active brain injury at follow-up 4 months later. These biomarkers were associated with elevations of pro-inflammatory cytokines and the presence of autoantibodies to a large number of different antigens. Autoantibodies were commonly seen against lung surfactant proteins but also brain proteins such as myelin associated glycoprotein. Commensurate findings were seen in the influenza cohort. A distinct process characterised by elevation of serum total tau was seen in patients at follow-up, which appeared to be independent of initial disease severity and was not associated with dysregulated immune responses unlike NfL and GFAP. These results demonstrate that brain injury is a common consequence of both COVID-19 and influenza, and is therefore likely to be a feature of severe viral infection more broadly. The brain injury occurs in the context of dysregulation of both innate and adaptive immune responses, with no single pathogenic mechanism clearly responsible.
View details for DOI 10.1093/brain/awac321
View details for PubMedID 36065116
Immunometabolism, a new therapeutic development for immunotherapies of high-grade gliomas: a narrative review.
Chinese clinical oncology
2022; 11 (4): 26
Immunotherapy has yielded significant improvements in survival for many cancer types, but its impact on glioblastoma (GBM) has been relatively muted. There is a growing interest in understanding the role of cancer metabolism and its role in tumor growth and therapeutic response. Thus, it is equally important to consider the clinical implications of immune cell metabolism on cancer progression and implications for therapeutic development. Our objective is to present new developments in immunometabolic research that are relevant to immunotherapy development for high-grade gliomas.A literature search and review was conducted, regarding original research articles studying metabolic pathways of immune cells in high-grade gliomas. Searches were conducted in PubMed and Embase databases on May 15 and June 13, 2022. English-language original research articles were selected and prioritized based on their inclusion of findings related to metabolic changes in myeloid and lymphoid cells in the glioma tumor microenvironment.There are many metabolic mechanisms by which immune cells in high-grade gliomas, like GBM, contribute to tumor growth and persistence via immunosuppression and high therapeutic resistance. There are also several ways that metabolic optimization has already been shown to improve immunotherapies already in clinical trials or in use, including dendritic cell vaccines and chimeric antigen receptor T cells.The implications of immunometabolic research presented here should be taken into consideration in future research and immunotherapy development of high-grade gliomas for our best chances at improving patient survival.
View details for DOI 10.21037/cco-22-58
View details for PubMedID 36098097
Health Care Resource Utilization in Management of Opioid-Naive Patients With Newly Diagnosed Neck Pain.
JAMA network open
2022; 5 (7): e2222062
Importance: Research has uncovered heterogeneity and inefficiencies in the management of idiopathic low back pain, but few studies have examined longitudinal care patterns following newly diagnosed neck pain.Objective: To understand health care utilization in patients with new-onset idiopathic neck pain.Design, Setting, and Participants: This cross-sectional study used nationally sourced longitudinal data from the IBM Watson Health MarketScan claims database (2007-2016). Participants included adult patients with newly diagnosed neck pain, no recent opioid use, and at least 1 year of continuous postdiagnosis follow-up. Exclusion criteria included prior or concomitant diagnosis of traumatic cervical disc dislocation, vertebral fractures, myelopathy, and/or cancer. Only patients with at least 1 year of prediagnosis lookback were included. Data analysis was performed from January 2021 to January 2022.Main Outcomes and Measures: The primary outcome of interest was 1-year postdiagnosis health care expenditures, including costs, opioid use, and health care service utilization. Early services were those received within 30 days of diagnosis. Multivariable regression models and regression-adjusted statistics were used.Results: In total, 679 030 patients (310 665 men [45.6%]) met the inclusion criteria, of whom 7858 (1.2%) underwent surgery within 1 year of diagnosis. The mean (SD) age was 44.62 (14.87) years among nonsurgical patients and 49.69 (9.53) years among surgical patients. Adjusting for demographics and comorbidities, 1-year regression-adjusted health care costs were $24 267.55 per surgical patient and $515.69 per nonsurgical patient. Across all health care services, $95 379 949 was accounted for by nonsurgical patients undergoing early imaging who did not receive any additional conservative therapy or epidural steroid injections, for a mean (SD) of $477.53 ($1375.60) per patient and median (IQR) of $120.60 ($20.70-$452.37) per patient. On average, patients not undergoing surgery, physical therapy, chiropractic manipulative therapy, or epidural steroid injection, who underwent either early advanced imaging (magnetic resonance imaging or computed tomography) or both early advanced and radiographic imaging, accumulated significantly elevated health care costs ($850.69 and $1181.67, respectively). Early conservative therapy was independently associated with 24.8% (95% CI, 23.5%-26.2%) lower health care costs.Conclusions and Relevance: In this cross-sectional study, early imaging without subsequent intervention was associated with significantly increased health care spending among patients with newly diagnosed idiopathic neck pain. Early conservative therapy was associated with lower costs, even with increased frequency of therapeutic services, and may have reduced long-term care inefficiency.
View details for DOI 10.1001/jamanetworkopen.2022.22062
View details for PubMedID 35816312
Neurological update: COVID-19.
Journal of neurology
2021; 268 (11): 4379-4387
Coronavirus Disease 2019 is predominantly a disorder of the respiratory system, but neurological complications have been recognised since early in the pandemic. The major pathophysiological processes leading to neurological damage in COVID-19 are cerebrovascular disease, immunologically mediated neurological disorders and the detrimental effects of critical illness on the nervous system. It is still unclear whether direct invasion of the nervous system by the Severe Acute Respiratory Syndrome Coronavirus 2 occurs; given the vast numbers of people infected at this point, this uncertainty suggests that nervous system infection is unlikely to represent a significant issue if it occurs at all. In this review, we explore what has been learnt about the neurological complications of COVID-19 over the course of the pandemic, and by which mechanisms these complications most commonly occur.
View details for DOI 10.1007/s00415-021-10581-y
View details for PubMedID 33929617
View details for PubMedCentralID PMC8085652