Healthcare artificial intelligence (AI) holds the potential to increase patient safety, augment efficiency and improve patient outcomes, yet research is often limited by data access, cohort curation, ...and tools for analysis. Collection and translation of electronic health record data, live data, and real-time high-resolution device data can be challenging and time-consuming. The development of clinically relevant AI tools requires overcoming challenges in data acquisition, scarce hospital resources, and requirements for data governance. These bottlenecks may result in resource-heavy needs and long delays in research and development of AI systems. We present a system and methodology to accelerate data acquisition, dataset development and analysis, and AI model development. We created an interactive platform that relies on a scalable microservice architecture. This system can ingest 15,000 patient records per hour, where each record represents thousands of multimodal measurements, text notes, and high-resolution data. Collectively, these records can approach a terabyte of data. The platform can further perform cohort generation and preliminary dataset analysis in 2-5 minutes. As a result, multiple users can collaborate simultaneously to iterate on datasets and models in real time. We anticipate that this approach will accelerate clinical AI model development, and, in the long run, meaningfully improve healthcare delivery.
Contrast-induced nephropathy (CIN) has long been observed in both experimental and clinical studies. However, recent observational studies have questioned the prevalence and severity of CIN following ...intravenous contrast exposure. Initial studies of acute kidney injury following intravenous contrast were limited by the absence of control groups or contained control groups that did not adjust for additional acute kidney injury risk factors, including prevalent chronic kidney disease, as well as accepted prophylactic strategies. More contemporary use of propensity score–adjusted models have attempted to minimize the risk for selection bias, although bias cannot be completely eliminated without a prospective randomized trial. Based on existing data, we recommend the following CIN risk classification: patients with estimated glomerular filtration rates (eGFRs) ≥ 45mL/min/1.73m2 are at negligible risk for CIN, while patients with eGFRs<30mL/min/1.73m2 are at high risk for CIN. Patients with eGFRs between 30 and 44mL/min/1.73m2 are at an intermediate risk for CIN unless diabetes mellitus is present, which would further increase the risk. In all patients at any increased risk for CIN, the risk for CIN needs to be balanced by the risk of not performing an intravenous contrast-enhanced study.
Preeclampsia, a pregnancy-specific condition associated with new-onset hypertension after 20-weeks gestation, is a leading cause of maternal and neonatal morbidity and mortality. Predictive tools to ...understand which individuals are most at risk are needed.
We identified a cohort of N=1125 pregnant individuals who delivered between May 2015 and May 2022 at Mass General Brigham Hospitals with available electronic health record data and linked genetic data. Using clinical electronic health record data and systolic blood pressure polygenic risk scores derived from a large genome-wide association study, we developed machine learning (XGBoost) and logistic regression models to predict preeclampsia risk.
Pregnant individuals with a systolic blood pressure polygenic risk score in the top quartile had higher blood pressures throughout pregnancy compared with patients within the lowest quartile systolic blood pressure polygenic risk score. In the first trimester, the most predictive model was XGBoost, with an area under the curve of 0.74. In late pregnancy, with data obtained up to the delivery admission, the best-performing model was XGBoost using clinical variables, which achieved an area under the curve of 0.91. Adding the systolic blood pressure polygenic risk score to the models did not improve the performance significantly based on De Long test comparing the area under the curve of models with and without the polygenic score.
Integrating clinical factors into predictive models can inform personalized preeclampsia risk and achieve higher predictive power than the current practice. In the future, personalized tools can be implemented to identify high-risk patients for preventative therapies and timely intervention to improve adverse maternal and neonatal outcomes.
Current AI-driven research in radiology requires resources and expertise that are often inaccessible to small and resource-limited labs. The clinicians who are able to participate in AI research are ...frequently well-funded, well-staffed, and either have significant experience with AI and computing, or have access to colleagues or facilities that do. Current imaging data is clinician-oriented and is not easily amenable to machine learning initiatives, resulting in inefficient, time consuming, and costly efforts that rely upon a crew of data engineers and machine learning scientists, and all too often preclude radiologists from driving AI research and innovation. We present the system and methodology we have developed to address infrastructure and platform needs, while reducing the staffing and resource barriers to entry. We emphasize a data-first and modular approach that streamlines the AI development and deployment process while providing efficient and familiar interfaces for radiologists, such that they can be the drivers of new AI innovations.
As part of its Extremely Brilliant Source (EBS) upgrade project, the ESRF's BM29 BioSAXS beamline was subject to a significant upgrade and refurbishment. In addition to the replacement of the ...beamline's original bending magnet source by a two‐pole wiggler, leading to an increase in brilliance by a factor of 60, the sample environment of the beamline was almost completely refurbished: a vacuum‐compatible Pilatus3 X 2M with a sensitive area of 253.7 mm × 288 mm and frame rates up to 250 Hz was installed, increasing the active area available and thus the q‐scaling of scattering images taken; the sample changer was replaced with an upgraded version, allowing more space for customizable sample environments and the installation of two new sample exposure units; the software associated with the beamline was also renewed. In addition, the layout and functionality of the BSXCuBE3 (BioSAXS Customized Beamline Environment) data acquisition software was redesigned, providing an intuitive `user first' approach for inexperienced users, while at the same time maintaining more powerful options for experienced users and beamline staff. Additional features of BSXCuBE3 are queuing of samples; either consecutive sample changer and/or SEC‐SAXS (size‐exclusion chromatography small‐angle X‐ray scattering) experiments, including column equilibration were also implemented. Automatic data processing and analysis are now managed via Dahu, an online server with upstream data reduction, data scaling and azimuthal integration built around PyFAI (Python Fast Azimuthal Integration), and data analysis performed using the open source FreeSAS. The results of this automated data analysis pipeline are displayed in ISPyB/ExiSAXS. The upgraded BM29 has been in operation since the post‐EBS restart in September 2020, and here a full description of its new hardware and software characteristics together with examples of data obtained are provided.
As part of its Extremely Brilliant Source (EBS) upgrade project, the ESRF's BioSAXS BM29 beamline was subject to a significant upgrade. Its bending magnet source was replaced by a two‐pole wiggler, and a larger hybrid‐pixel detector, a new sample changer and improved sample exposure cells were also installed and commissioned. Additionally, nearly all beamline software, including that for experiment control and data analysis, was renewed. Here the current status of the beamline is described, including new opportunities post‐ESRF–EBS.
The clinical notes in a given patient record contain much redundancy, in large part due to clinicians' documentation habit of copying from previous notes in the record and pasting into a new note. ...Previous work has shown that this redundancy has a negative impact on the quality of text mining and topic modeling in particular. In this paper we describe a novel variant of Latent Dirichlet Allocation (LDA) topic modeling, Red-LDA, which takes into account the inherent redundancy of patient records when modeling content of clinical notes. To assess the value of Red-LDA, we experiment with three baselines and our novel redundancy-aware topic modeling method: given a large collection of patient records, (i) apply vanilla LDA to all documents in all input records; (ii) identify and remove all redundancy by chosing a single representative document for each record as input to LDA; (iii) identify and remove all redundant paragraphs in each record, leaving partial, non-redundant documents as input to LDA; and (iv) apply Red-LDA to all documents in all input records. Both quantitative evaluation carried out through log-likelihood on held-out data and topic coherence of produced topics and qualitative assessment of topics carried out by physicians show that Red-LDA produces superior models to all three baseline strategies. This research contributes to the emerging field of understanding the characteristics of the electronic health record and how to account for them in the framework of data mining. The code for the two redundancy-elimination baselines and Red-LDA is made publicly available to the community.
•Intraocular anti-VEGF drugs dominate the market of macular edema.•In retinal diseases, anti-VEGFs are primarily anti-edematous drugs.•The anti-edematous mechanisms of anti-VEGF are not fully ...understood.•There is evidence that anti-edematous is independent from VEGF neutralization.•Intravitreous proteins regulate genes encoding anti-edematous proteins.
The intravitreous injection of therapeutic proteins that neutralize vascular endothelial growth factor (VEGF) family members is efficient to reduce macular edema associated with wet age-related macular degeneration (AMD), retinal vein occlusion (RVO) and diabetic retinopathy (DR). It has revolutionized the visual prognosis of patients with macular edema. The antiedematous effect is dependent on an intravitreous dose of drug, which varies between patients and requires frequent and repeated injections to maintain its effects. At the time when optimizing the duration of anti-VEGF effects is a major challenge, understanding how anti-VEGF reduces macular edema is crucial. We discuss herein how anti-VEGF exerts antiedematous effects and raise the hypothesis that mechanisms, unrelated to VEGF neutralization, might have been underestimated.
Treatment of secondary hyperparathyroidism with vitamin D and calcium in patients receiving dialysis is often complicated by hypercalcemia and hyperphosphatemia, which may contribute to ...cardiovascular disease and adverse clinical outcomes. Calcimimetics target the calcium-sensing receptor and lower parathyroid hormone levels without increasing calcium and phosphorus levels. We report the results of two identical randomized, double-blind, placebo-controlled trials evaluating the safety and effectiveness of the calcimimetic agent cinacalcet hydrochloride.
Patients who were receiving hemodialysis and who had inadequately controlled secondary hyperparathyroidism despite standard treatment were randomly assigned to receive cinacalcet (371 patients) or placebo (370 patients) for 26 weeks. Once-daily doses were increased from 30 mg to 180 mg to achieve intact parathyroid hormone levels of 250 pg per milliliter or less. The primary end point was the percentage of patients with values in this range during a 14-week efficacy-assessment phase.
Forty-three percent of the cinacalcet group reached the primary end point, as compared with 5 percent of the placebo group (P<0.001). Overall, mean parathyroid hormone values decreased 43 percent in those receiving cinacalcet but increased 9 percent in the placebo group (P<0.001). The serum calcium-phosphorus product declined by 15 percent in the cinacalcet group and remained unchanged in the placebo group (P<0.001). Cinacalcet effectively reduced parathyroid hormone levels independently of disease severity or changes in vitamin D sterol dose.
Cinacalcet lowers parathyroid hormone levels and improves calcium-phosphorus homeostasis in patients receiving hemodialysis who have uncontrolled secondary hyperparathyroidism.
The paper examines the limits of state interference in proscribing cultural norms by considering gender discrimination, right of people to leave their community free of penalties, denying women ...appropriate education, and forced or arranged marriages for girls and young women. The discussion opens by reflecting on the discriminatory practices of the Pueblo tribes against their women and analysing an American court case, Santa Clara v. Martinez. It is argued that the severity of rights violations within the minority group, the insufficient dispute-resolution-mechanisms, and the inability of individuals to leave the community if they so desire without penalty justify state intervention to uphold the dissenters’ basic rights. Next, a Canadian case, Hofer v. Hofer, illustrates the problematics of denying reasonable exit right to members who may wish to leave their community. Subsequently, the discussion turns to the issue of arranged and forced marriages of girls and young women. While the latter is coercive the former is not. While forced marriages should be denounced as unjust, arranged marriages can be accepted. Finally, the paper considers denying education to women, arguing that such a denial is unjust and discriminatory.