BMC Medical Informatics and Decision Making最新文献

Background: Biomedical Relation Extraction (RE) is essential for uncovering complex relationships between biomedical entities within text. However, training RE classifiers is challenging in low-resource biomedical applications with few labeled examples.

Methods: We explore the potential of Shortest Dependency Paths (SDPs) to aid biomedical RE, especially in situations with limited labeled examples. In this study, we suggest various approaches to employ SDPs when creating word and sentence representations under supervised, semi-supervised, and in-context-learning settings.

Results: Through experiments on three benchmark biomedical text datasets, we find that incorporating SDP-based representations enhances the performance of RE classifiers. The improvement is especially notable when working with small amounts of labeled data.

Conclusion: SDPs offer valuable insights into the complex sentence structure found in many biomedical text passages. Our study introduces several straightforward techniques that, as demonstrated experimentally, effectively enhance the accuracy of RE classifiers.

Background: Previous studies have shown that shared decision-making (SDM) between a practitioner and a patient strengthens the ideal of treatment adherence. This study employed a multi-method approach to SDM in healthcare to reinforce the theoretical and methodological grounds of this argument. As the study design, self-reported survey items and experimental vignettes were combined in one electronic questionnaire. This technique aimed to analyze the effects of previous experiences and the current preferences regarding SDM on the intentions to follow-through with the medical recommendations.

Method: Using quantitative data collected from the members of the Finnish Pensioners' Federation (N = 1610), this study focused on the important and growing population of older adults as healthcare consumers. Illustrated vignettes were used in the evaluation of expected adherence to both vaccination and the treatment of an illness, depending on the decision-making style varying among the repeated scenarios. In a within-subjects study design, each study subject acted as their own control.

Results: The findings demonstrated that SDM correlates with expected adherence to a treatment and vaccination. Both the retrospective experiences and prospective aspirations of SDM in clinical encounters supported the patients' expected adherence to vaccination and treatment while decreasing the probability of pseudo-compliance. The association between SDM and expected adherence was not affected by the perceived health of the respondents. However, the associations among the expected adherence and decision-making styles were found to differ between the treatment and vaccination scenarios.

Conclusions: SDM enables expected treatment adherence among older adults. Thus, the multi-method study emphasizes the importance of SDM in various healthcare encounters. The findings further imply that SDM research benefits from questionnaires combining self-report methods and experimental study designs. Further cross-validation studies using various types of written and illustrated scenarios are encouraged.

Background: The frequency of hip and knee arthroplasty surgeries has been rising steadily in recent decades. This trend is attributed to an aging population, leading to increased demands on healthcare systems. Fast Track (FT) surgical protocols, perioperative procedures designed to expedite patient recovery and early mobilization, have demonstrated efficacy in reducing hospital stays, convalescence periods, and associated costs. However, the criteria for selecting patients for FT procedures have not fully capitalized on the available patient data, including patient-reported outcome measures (PROMs).

Methods: Our study focused on developing machine learning (ML) models to support decision making in assigning patients to FT procedures, utilizing data from patients' self-reported health status. These models are specifically designed to predict the potential health status improvement in patients initially selected for FT. Our approach focused on techniques inspired by the concept of controllable AI. This includes eXplainable AI (XAI), which aims to make the model's recommendations comprehensible to clinicians, and cautious prediction, a method used to alert clinicians about potential control losses, thereby enhancing the models' trustworthiness and reliability.

Results: Our models were trained and tested using a dataset comprising 899 records from individual patients admitted to the FT program at IRCCS Ospedale Galeazzi-Sant'Ambrogio. After training and selecting hyper-parameters, the models were assessed using a separate internal test set. The interpretable models demonstrated performance on par or even better than the most effective 'black-box' model (Random Forest). These models achieved sensitivity, specificity, and positive predictive value (PPV) exceeding 70%, with an area under the curve (AUC) greater than 80%. The cautious prediction models exhibited enhanced performance while maintaining satisfactory coverage (over 50%). Further, when externally validated on a separate cohort from the same hospital-comprising patients from a subsequent time period-the models showed no pragmatically notable decline in performance.

Conclusions: Our results demonstrate the effectiveness of utilizing PROMs as basis to develop ML models for planning assignments to FT procedures. Notably, the application of controllable AI techniques, particularly those based on XAI and cautious prediction, emerges as a promising approach. These techniques provide reliable and interpretable support, essential for informed decision-making in clinical processes.

Background: The risk assessment for survival in heart failure (HF) remains one of the key focuses of research. This study aims to develop a simple and feasible nomogram model for survival in HF based on the Heart Failure-A Controlled Trial Investigating Outcomes of Exercise TraiNing (HF-ACTION) to support clinical decision-making.

Methods: The HF patients were extracted from the HF-ACTION database and randomly divided into a training cohort and a validation cohort at a ratio of 7:3. Multivariate Cox regression was used to identify and integrate significant prognostic factors to form a nomogram, which was displayed in the form of a static nomogram. Bootstrap resampling (resampling = 1000) and cross-validation was used to internally validate the model. The prognostic performance of the model was measured by the concordance index (C-index), calibration curve, and the decision curve analysis.

Results: There were 1394 patients with HF in the overall analysis. Seven prognostic factors, which included age, body mass index (BMI), sex, diastolic blood pressure (DBP), exercise duration, peak exercise oxygen consumption (peak VO₂), and loop diuretic, were identified and applied to the nomogram construction based on the training cohort. The C-index of this model in the training cohort was 0.715 (95% confidence interval (CI): 0.700, 0.766) and 0.662 (95% CI: 0.646, 0.752) in the validation cohort. The area under the ROC curve (AUC) value of 365- and 730-day survival is (0.731, 0.734) and (0.640, 0.693) respectively in the training cohort and validation cohort. The calibration curve showed good consistency between nomogram-predicted survival and actual observed survival. The decision curve analysis (DCA) revealed net benefit is higher than the reference line in a narrow range of cutoff probabilities and the result of cross-validation indicates that the model performance is relatively robust.

Conclusions: This study created a nomogram prognostic model for survival in HF based on a large American population, which can provide additional decision information for the risk prediction of HF.

Background: Generalized Joint Hyper-mobility (GJH) can aid in the diagnosis of Ehlers-Danlos Syndrome (EDS), a complex genetic connective tissue disorder with clinical features that can mimic other disease processes. Our study focuses on developing a unique image-based goniometry system, the HybridPoseNet, which utilizes a hybrid deep learning model.

Objective: The proposed model is designed to provide the most accurate joint angle measurements in EDS appraisals. Using a hybrid of CNNs and HyperLSTMs in the pose estimation module of HybridPoseNet offers superior generalization and time consistency properties, setting it apart from existing complex libraries.

Methodology: HybridPoseNet integrates the spatial pattern recognition prowess of MobileNet-V2 with the sequential data processing capability of HyperLSTM units. The system captures the dynamic nature of joint motion by creating a model that learns from individual frames and the sequence of movements. The CNN module of HybridPoseNet was trained on a large and diverse data set before the fine-tuning of video data involving 50 individuals visiting the EDS clinic, focusing on joints that can hyperextend. HyperLSTMs have been incorporated in video frames to avoid any time breakage in joint angle estimation in consecutive frames. The model performance was evaluated using Spearman's coefficient correlation versus manual goniometry measurements, as well as by the human labeling of joint position, the second validation step.

Outcome: Preliminary findings demonstrate HybridPoseNet achieving a remarkable correlation with manual Goniometric measurements: thumb (rho = 0.847), elbows (rho = 0.822), knees (rho = 0.839), and fifth fingers (rho = 0.896), indicating that the newest model is considerably better. The model manifested a consistent performance in all joint assessments, hence not requiring selecting a variety of pose-measuring libraries for every joint. The presentation of HybridPoseNet contributes to achieving a combined and normalized approach to reviewing the mobility of joints, which has an overall enhancement of approximately 20% in accuracy compared to the regular pose estimation libraries. This innovation is very valuable to the field of medical diagnostics of connective tissue diseases and a vast improvement to its understanding.

This research study demonstrates an efficient scheme for early detection of cardiorespiratory complications in pandemics by Utilizing Wearable Electrocardiogram (ECG) sensors for pattern generation and Convolution Neural Networks (CNN) for decision analytics. In health-related outbreaks, timely and early diagnosis of such complications is conclusive in reducing mortality rates and alleviating the burden on healthcare facilities. Existing methods rely on clinical assessments, medical history reviews, and hospital-based monitoring, which are valuable but have limitations in terms of accessibility, scalability, and timeliness, particularly during pandemics. The proposed scheme commences by deploying wearable ECG sensors on the patient's body. These sensors collect data by continuously monitoring the cardiac activity and respiratory patterns of the patient. The collected raw data is then transmitted securely in a wireless manner to a centralized server and stored in a database. Subsequently, the stored data is assessed using a preprocessing process which extracts relevant and important features like heart rate variability and respiratory rate. The preprocessed data is then used as input into the CNN model for the classification of normal and abnormal cardiorespiratory patterns. To achieve high accuracy in abnormality detection the CNN model is trained on labeled data with optimized parameters. The performance of the proposed scheme is evaluated and gauged using different scenarios, which shows a robust performance in detecting abnormal cardiorespiratory patterns with a sensitivity of 95% and specificity of 92%. Prominent observations, which highlight the potential for early interventions include subtle changes in heart rate variability and preceding respiratory distress. These findings show the significance of wearable ECG technology in improving pandemic management strategies and informing public health policies, which enhances preparedness and resilience in the face of emerging health threats.