Intelligence-based medicine最新文献

Introduction

Methods

Results

Remarks and significance

Introduction

Methods

Conclusion

Emergency department (ED) overcrowding is a significant problem in the US. This paper develops a decision model to mitigate ED overcrowding by helping hospitals proactively plan patient boarding processes. The information obtained after the initial assessment of COVID-19 patients in the ED, including patient demographics and medical history, is utilized to predict ICU admission earlier. The predicted information can be communicated with the inpatient unit to prepare an ICU bed for the patients who need ICU care. As a result, the boarding time when patients wait for an ICU bed to be ready can be reduced. The data used in this study included 100 features and 19,155 COVID-19 patients from an academic medical center located in the Southeast United States. Multiple feature selection methods along with Extreme Gradient Boosting (XGBoost) were utilized to develop the models. The parameters of the XGBoost models are optimized using simulated annealing (SA). Among the proposed models, the best model included ten features and resulted in an area under the curve (AUC) of 89.2%, which is the highest among the models proposed in the literature. The proposed prediction model allows hospital administrators to allocate ICU beds more efficiently, enhance patient flow, and mitigate ED overcrowding.

Background

Polycystic Ovary Syndrome (PCOS) is reported to affect between 4% and 21% of reproductive aged people with ovaries. It is a heterogeneous condition with a lack of established phenotypes that address the range of reproductive and metabolic features present in PCOS. These reproductive and metabolic features may result in patients undergoing a variety of relevant laboratory tests. Previous work has led to the gathering of laboratory test results from a PCOS specific forum, hosted on a website called reddit.

Objectives

In this paper, laboratory results and body mass index (BMI) posted on the PCOS reddit forum were clustered to show the usefulness of the PCOS forum for PCOS research and validate existing PCOS phenotypes or discover other appropriate phenotypes.

Methods and results

Over 1500 sets of PCOS-related reddit laboratory test results and BMIs were clustered using nearest neighbour imputation and K-means clustering. However, only non-imputed data was included in the final clusters. Kernel Density Estimation plots were used to display the distinct clusters. The clustered test results suggested the existence of distinct metabolic and reproductive phenotypes, as well as a group displaying mild features of both types of dysregulations and a group skewed towards normal results. It was also possible to separate the groups further into distinct hypothyroid groups within the mixed dysregulation group and to separate insulin resistant and diabetes-like groups within the metabolic group.

Conclusions

This research further validates the usefulness of exploring alternate data sources in the age of the internet and machine learning. The reddit clusters reinforced the existing notion that people with PCOS can be separated into a primarily metabolic pathology group, a primarily reproductive pathology group and an in between group with pathology in both domains.

Introduction

Deformable image registration (DIR) plays a vital role in adaptive radiotherapy (ART). For the clinical implementation of DIR, evaluation of deformation accuracy is a critical step. While contour-based metrics, for example Dice similarity coefficient (DSC), are widely implemented for DIR validation, they require delineation of contours which is time-consuming and would cause hold-ups in an ART workflow. Therefore, this work aims to accomplish the prediction of DSC using various metrics based on deformation vector field (DVF) by applying machine learning (ML), in order to provide an efficient means of DIR validation with minimised human intervention.

Methods

Planning CT image was deformed to the cone-beam CT images for 20 prostate cancer patients. Various DVF-based metrics and DSC were calculated, and the former was used as input features to predict the latter using three ML models, namely linear regression (LR), Nu Support Vector Regression (NuSVR) and Random Forest Regressor (RFR). Four datasets were used for analysis: 1) prostate, 2) bladder, 3) rectum and 4) all the organs combined. Average mean absolute error (MAE) was computed to evaluate the model performance. The classification performance of the best-performing model was further evaluated, and the prediction interval and feature importance were calculated.

Results

Overall, RFR achieved the lowest average MAE, ranging between 0.045 and 0.069 for the four datasets, while LR and NuSVR had slightly poorer performances. Analysis on the results of best-performing model showed that sensitivity and specificity of 0.86 and 0.51, respectively, were obtained when a prediction threshold of 0.85 was used to classify the fourth dataset. Jacobian determinant was found to be a significant contributor to the predictions of all four datasets using this model.

Conclusion

This study demonstrated the potential of several ML models, especially RFR, to be applied for prediction of DSC to speed up the DIR validation process.

Introduction

Stroke remains the second leading cause of death worldwide, with many survivors facing significant disabilities. In acute stroke care, the timeless adage 'Time is brain' underscores the vital need for quick action. Innovative Artificial Intelligence (AI) technology potentially enables swift detection and management of acute ischemic strokes, revolutionizing acute stroke care towards enhanced automation.

Methods

The study is registered with Prospero under CRD42024496716 and adheres to the Problem, Intervention, Comparison, and Outcomes framework (PICO). The analysis used Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We searched Embase, PubMed, DBLP, Google Scholar, IEEE Xplore, Cochrane database, IEEE, Web of Science, ArXiv, MedRxiv, and Semantic Scholar. The articles included were published between 2019 and 2023. Out of 1528 articles identified, thirty-seven met the inclusion criteria.

Results

We compared AI-augmented Large Vessel Occlusion (LVO) detection and non-AI LVO detection in various patient processing times related to emergent endovascular therapy in acute ischemic strokes. Triage Time, Door-to-Intervention Notification Time (INR), and Door-to -Arterial Puncture Time revealed an odds ratio (OR) of 0.39 (95 % CI: 0.29–0.54, p < 0.001), 0.30 (95 % CI: 0.21–0.42, p < 0.001), and 0.50 (95 % CI: 0detection 0.30–0.82, p = 0.007), respectively -- all of which had negligible heterogeneity (I^2 = 0). CT-to-Puncture-Time and Door-to-CTA-Time yielded an OR of 0.57 (95 % CI: 0.31–1.04, p = 0.065) and 0.77 (95 % CI: 0.37–1.60, p = 0.489), respectively -- both of which had negligible heterogeneity (I^2 = 0). The Last Known Well (LWK) to Time of Arrival resulted in an OR of 1.15 (95 % CI: 0.83–1.59, p = 0.409, I^2 = 0). AI stroke detection sensitivity OR of 0.91 (95 % CI: 0.88–0.95, p < 0.001) should be interpreted with potential heterogeneity in mind (I^2 = 69.3). National Institute of Health score (NIHSS) mean of 16.20 (95 % CI: 14.96–17.45, p = 0.001, I^2 = 0). Patient Transfer-Times between primary and comprehensive stroke centers generated an OR of 0.98 (95 % CI: 0.73–1.32, p = 893, I^2 = 0). Similarly, Door-in-Door-Out Time (DIDO) had an OR of 1.19 (95 % CI: 0.21–6.88, p = 0.848) and low heterogeneity (I^2 = 5.1). The results indicated significant differences across several parameters between the AI augmentation and non-AI groups.

Conclusion

Our findings highlight how AI augments healthcare providers' ability to detect and manage strokes swiftly and accurately within acute care settings. As these technologies progress, healthcare organizations mature, and AI becomes more integrated into healthcare systems, longitudinal studies are critical in evaluating its impact on workflow efficiency, cost-effectiveness, and clinical outcomes.

Humanity faces various types of viral infections, such as COVID-19, annually. In this paper, we propose a Geospatial SEIR(D) model based on a multi-agent approach with continuous-discrete states. This model accounts for key parameters of viral infections, daily human activities, and geodata. Our developed algorithms enable the simulation of statistical parameters such as the number of infected, recovered, deceased, and susceptible individuals, along with the spatial distribution of the pandemic on a geographical map. The model was validated by simulating the COVID-19 spread in Lviv, Ukraine. Several preventive strategies were analyzed: implementing a 50 % reduction in infection probability through mask mandates delayed the peak to 150 days with a 25 % reduction in the maximum number of patients, while a 75 % reduction delayed the peak to 240 days with a 60 % reduction in the maximum number of patients. Prohibiting public transport and public places resulted in the epidemic peaking on day 165 with 2854 patients, significantly reducing the spread rate compared to the base model. Simulating 50 %, 75 %, and 100 % vaccination rates showed a reduction in the peak number of infections by 34 %, 57 %, and 94 %, respectively, also extending the duration of the epidemic. Enforcing weekend quarantine delayed the epidemic onset by one month but had minimal impact on the overall number of infections and duration. Combining mask mandates, transport restrictions, and vaccination led to the most effective mitigation, with the average number of sick agents around 8 and never exceeding 15 over four years. This comprehensive approach highlights the effectiveness of combining various preventive measures to control the spread of viral infections. The proposed model provides a valuable tool for policymakers to evaluate and implement effective strategies against pandemics.

Objective

The study of the epidemiology of delirium in hospitalized patients is challenging. We aimed to identify the presence or absence of delirium from clinical text notes using natural language processing (NLP) techniques and machine learning (ML) models.

Materials and methods

We developed a delirium predictive model using 942 clinical notes from hospitalized patients with an ICD-10 delirium hospital discharge code. Moreover, we implemented ML models using a) delirium-suggestive words from an expert-defined dictionary or b) free text in clinical notes. Both strategies considered positive and negative delirium-associated words.

Results

At the note level, for the dictionary method, the logistic regression model achieved an area under the receiver-operating curve (AUROC) of 0.917 for positive words and 0.914 for combined positive and negative words. The areas under the precision-recall curve (AUPR) were 0.893 and 0.897, respectively. For the free-text method, the model achieved an AUROC of 0.826 and 0.830 and AUPR of 0.852 and 0.856, respectively.

Discussion

NLP-based ML models accurately identified the presence of delirium in clinical notes. The dictionary-based method was superior to the free-text method. The use of negative features improved performance in both methods.

Conclusion

Our proposed NLP-based ML model identified delirium in clinical notes. This model could automatically screen millions of notes and facilitate the study of the epidemiology of in-hospital delirium.