BMC Medical Informatics and Decision Making最新文献

Background: Recently, machine learning (ML), deep learning (DL), and natural language processing (NLP) have provided promising results in the free-form radiological reports' classification in the respective medical domain. In order to classify radiological reports properly, a high-quality annotated and curated dataset is required. Currently, no publicly available breast imaging-based radiological dataset exists for the classification of Breast Imaging Reporting and Data System (BI-RADS) categories and breast density scores, as characterized by the American College of Radiology (ACR). To tackle this problem, we construct and annotate a breast imaging-based radiological reports dataset and its benchmark results. The dataset was originally in Spanish. Board-certified radiologists collected and annotated it according to the BI-RADS lexicon and categories at the Breast Radiology department, TecSalud Hospitals Monterrey, Mexico. Initially, it was translated into English language using Google Translate. Afterwards, it was preprocessed by removing duplicates and missing values. After preprocessing, the final dataset consists of 5046 unique reports from 5046 patients with an average age of 53 years and 100% women. Furthermore, we used word-level NLP-based embedding techniques, term frequency-inverse document frequency (TF-IDF) and word2vec to extract semantic and syntactic information. We also compared the performance of ML, DL and large language models (LLMs) classifiers for BI-RADS category classification.

Results: The final breast imaging-based radiological reports dataset contains 5046 unique reports. We compared K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Naive Bayes (NB), Random Forest (RF), Adaptive Boosting (AdaBoost), Gradient-Boosting (GB), Extreme Gradient Boosting (XGB), Long Short-Term Memory (LSTM), Bidirectional Encoder Representations from Transformers (BERT) and Biomedical Generative Pre-trained Transformer (BioGPT) classifiers. It is observed that the BioGPT classifier with preprocessed data performed 6% better with a mean sensitivity of 0.60 (95% confidence interval (CI), 0.391-0.812) compared to the second best performing classifier BERT, which achieved mean sensitivity of 0.54 (95% CI, 0.477-0.607).

Conclusion: In this work, we propose a curated and annotated benchmark dataset that can be used for BI-RADS and breast density category classification. We also provide baseline results of most ML, DL and LLMs models for BI-RADS classification that can be used as a starting point for future investigation. The main objective of this investigation is to provide a repository for the investigators who wish to enter the field to push the boundaries further.

Background: The mechanism for recording International Classification of Diseases (ICD) and diagnosis related groups (DRG) codes in a patient's chart is through a certified medical coder who manually reviews the medical record at the completion of an admission. High-acuity ICD codes justify DRG modifiers, indicating the need for escalated hospital resources. In this manuscript, we demonstrate that value of rules-based computer algorithms that audit for omission of administrative codes and quantifying the downstream effects with regard to financial impacts and demographic findings did not indicate significant disparities.

Methods: All study data were acquired via the UCLA Department of Anesthesiology and Perioperative Medicine's Perioperative Data Warehouse. The DataMart is a structured reporting schema that contains all the relevant clinical data entered into the EPIC (EPIC Systems, Verona, WI) electronic health record. Computer algorithms were created for eighteen disease states that met criteria for DRG modifiers. Each algorithm was run against all hospital admissions with completed billing from 2019. The algorithms scanned for the existence of disease, appropriate ICD coding, and DRG modifier appropriateness. Secondarily, the potential financial impact of ICD omissions was estimated by payor class and an analysis of ICD miscoding was done by ethnicity, sex, age, and financial class.

Results: Data from 34,104 hospital admissions were analyzed from January 1, 2019, to December 31, 2019. 11,520 (32.9%) hospital admissions were algorithm positive for a disease state with no corresponding ICD code. 1,990 (5.8%) admissions were potentially eligible for DRG modification/upgrade with an estimated lost revenue of $22,680,584.50. ICD code omission rates compared against reference groups (private payors, Caucasians, middle-aged patients) demonstrated significant p-values < 0.05; similarly significant p-value where demonstrated when comparing patients of opposite sexes.

Conclusions: We successfully used rules-based algorithms and raw structured EHR data to identify omitted ICD codes from inpatient medical record claims. These missing ICD codes often had downstream effects such as inaccurate DRG modifiers and missed reimbursement. Embedding augmented intelligence into this problematic workflow has the potential for improvements in administrative data, but more importantly, improvements in administrative data accuracy and financial outcomes.

Introduction: Remote monitoring can strengthen postoperative care in the community and minimise the burden of complications. However, implementation requires a clear understanding of how to sustainably integrate such complex interventions into existing care pathways. This study aimed to explore perceptions of potential facilitators and barriers to the implementation of digital remote postoperative monitoring from key stakeholders and derive recommendations for an implementable service.

Methods: A qualitative implementation study was conducted of digital remote postoperative wound monitoring across two UK tertiary care hospitals. All enrolled patients undergoing general surgery, and all staff involved in postoperative care were eligible. Criterion-based purposeful sampling was used to select stakeholders for semi-structured interviews on their perspectives and experiences of digital remote postoperative monitoring. A theory-informed deductive-inductive qualitative analysis was conducted; drawing on normalisation process theory (NPT) to determine facilitators for and barriers to implementation within routine care.

Results: There were 28 semi-structured interviews conducted with patients (n = 14) and healthcare professionals (n = 14). Remote postoperative monitoring was perceived to fulfil an unmet need in facilitating the diagnosis and treatment of postoperative complications. Participants perceived clear benefit to both the delivery of health services, and patient outcomes and experience, but some were concerned that this may not be equally shared due to potential issues with accessibility. The COVID-19 pandemic demonstrated telemedicine services are feasible to deliver and acceptable to participants, with examples of nurse-led remote postoperative monitoring currently supported within local care pathways. However, there was a discrepancy between patients' expectations regarding digital health to provide more personalised care, and the capacity of healthcare staff to deliver on these. Without further investment into IT infrastructure and allocation of staff, healthcare staff felt remote postoperative monitoring should be prioritised only for patients at the highest risk of complications.

Conclusion: The COVID-19 pandemic has sparked the digital transformation of international health systems, yet the potential of digital health interventions has yet to be realised. The benefits to stakeholders are clear, and if health systems seek to meet governmental policy and patient expectations, there needs to be greater organisational strategy and investment to ensure appropriate deployment and adoption into routine care.

Trial registration: NCT05069103.

Background: Familiar cardiopathies are genetic disorders that affect the heart. Cardiologists face a significant problem when treating patients suffering from these disorders: most DNA variations are novel (i.e., they have not been classified before). To facilitate the analysis of novel variations, we present CardioGraph, a platform specially designed to support the analysis of novel variations and help determine whether they are relevant for diagnosis. To do this, CardioGraph identifies and annotates the consequence of variations and provides contextual information regarding which heart structures, pathways, and biological processes are potentially affected by those variations.

Methods: We conducted our work through three steps. First, we define a data model to support the representation of the heterogeneous information. Second, we instantiate this data model to integrate and represent all the genomics knowledge available for familiar cardiopathies. In this step, we consider genomic data sources and the scientific literature. Third, the design and implementation of the CardioGraph platform. A three-tier structure was used: the database, the backend, and the frontend.

Results: Three main results were obtained: the data model, the knowledge base generated with the instantiation of the data model, and the platform itself. The platform code has been included as supplemental material in this manuscript. Besides, an instance is publicly available in the following link: https://genomics-hub.pros.dsic.upv.es:3090 .

Conclusion: CardioGraph is a platform that supports the analysis of novel variations. Future work will expand the body of knowledge about familiar cardiopathies and include new information about hotspots, functional studies, and previously reported variations.

Background: People who use drugs (PWUD) often face restricted healthcare access despite their heightened healthcare needs. Factors such as stigma, mistrust of the healthcare system, competing priorities, and geographical barriers pose significant healthcare access challenges. Telehealth offers an innovative solution to expand healthcare access for better inclusion of underserved populations in healthcare. We aimed to explore PWUDs' perceptions of telehealth as a healthcare delivery modality.

Methods: We utilized purposive sampling to recruit participants (N = 57) for nine focus group discussions (FGDs) in Athens, Greece. Eligibility criteria required participants to be at least 18 years, with current or prior injection drug use, and current internet access. The FGDs followed a semi-structured interview guide, were audio recorded, transcribed verbatim, translated into English, and de-identified. We applied thematic analysis to analyze FGD transcripts.

Results: Participants' mean (standard deviation) age was 47.9 (8.9) years, 89.5% (51/57) were male, 91.2% (52/57) were of Greek origin, and 61.4% (35/57) had attended at least 10 years of school. Three main themes emerged from the FGDs: (1) high internet utilization for healthcare-related purposes among PWUD, (2) highlighting telehealth benefits despite access obstacles and PWUDs' concerns about diagnostic accuracy, and (3) approaches to overcome access obstacles and build digital trust. Participants extensively used the internet for healthcare-related processes, such as accessing healthcare information and scheduling provider appointments. Despite being telehealth-inexperienced, most participants expressed a strong willingness to embrace telehealth due to its perceived convenience, time-saving nature, and trusted digital environment. Some participants recognized that the inability to conduct physical examinations through telehealth reduces its diagnostic accuracy, while others expressed concerns about digital literacy and technological infrastructure accessibility. Most participants expressed a preference for video-based telehealth encounters over audio-only encounters. To build trust in telehealth and promote patient-centeredness, participants recommended an initial in-person visit, virtual eye contact during telehealth encounters, patient education, and partnerships with PWUD-supportive community organizations equipped with appropriate infrastructure.

Conclusions: PWUD frequently use the internet for health-related purposes and suggested several approaches to enhance virtual trust. Their insights and suggestions are practical guidance for policymakers seeking to enhance healthcare access for underserved populations through telehealth.

Trial registration: NCT05794984.

Background: Determining the optimal timing of surgical intervention for Neonatal necrotizing enterocolitis (NEC) poses significant challenges. This study develops a predictive model using the long short-term memory network (LSTM) with a focal loss (FL) to identify infants at risk of developing Bell IIB + NEC early and issue timely surgical warnings.

Methods: Data from 791 neonates diagnosed with NEC are gathered from the Neonatal Intensive Care Unit (NICU), encompassing 35 selected features. Infants are categorized into those requiring surgical intervention (n = 257) and those managed medically (n = 534) based on the Mod-Bell criteria. A fivefold cross-validation approach is employed for training and testing. The LSTM algorithm is utilized to capture and utilize temporal relationships in the dataset, with FL employed as a loss function to address class imbalance. Model performance metrics include precision, recall, F1 score, and average precision (AP).

Results: The model tested on a real dataset demonstrated high performance. Predicting surgical risk 1 day in advance achieved precision (0.913 ± 0.034), recall (0.841 ± 0.053), F1 score (0.874 ± 0.029), and AP (0.917 ± 0.025). The 2-days-in-advance predictions yielded (0.905 ± 0.036), recall (0.815 ± 0.057), F1 score (0.857 ± 0.035), and AP (0.905 ± 0.029).

Conclusion: The LSTM model with FL exhibits high precision and recall in forecasting the need for surgical intervention 1 or 2 days ahead. This predictive capability holds promise for enhancing infants' outcomes by facilitating timely clinical decisions.

Background: As digital healthcare services handle increasingly more sensitive health data, robust access control methods are required. Especially in emergency conditions, where the patient's health situation is in peril, different healthcare providers associated with critical cases may need to be granted permission to acquire access to Electronic Health Records (EHRs) of patients. The research objective of this work is to develop a proactive access control method that can grant emergency clinicians access to sensitive health data, guaranteeing the integrity and security of the data, and generating trust without the need for a trusted third party.

Methods: A contextual and blockchain-based mechanism is proposed that allows access to sensitive EHRs by applying prognostic procedures where information based on context, is utilized to identify critical situations and grant access to medical data. Specifically, to enable proactivity, Long Short Term Memory (LSTM) Neural Networks (NNs) are applied that utilize patient's recent health history to prognose the next two-hour health metrics values. Fuzzy logic is used to evaluate the severity of the patient's health state. These techniques are incorporated in a private and permissioned Hyperledger-Fabric blockchain network, capable of securing patient's sensitive information in the blockchain network.

Results: The developed access control method provides secure access for emergency clinicians to sensitive information and simultaneously safeguards the patient's well-being. Integrating this predictive mechanism within the blockchain network proved to be a robust tool to enhance the performance of the access control mechanism. Furthermore, the blockchain network of this work can record the history of who and when had access to a specific patient's sensitive EHRs, guaranteeing the integrity and security of the data, as well as recording the latency of this mechanism, where three different access control cases are evaluated. This access control mechanism is to be enforced in a real-life scenario in hospitals.

Conclusions: The proposed mechanism informs proactively the emergency team of professional clinicians about patients' critical situations by combining fuzzy and predictive machine learning techniques incorporated in the private and permissioned blockchain network, and it exploits the distributed data of the blockchain architecture, guaranteeing the integrity and security of the data, and thus, enhancing the users' trust to the access control mechanism.

Background: Outcome measures are crucial to support a treat-to-target approach to rheumatoid arthritis (RA) care, yet their integration into clinical practice remains inconsistent. We developed an Electronic Heath Record-integrated, patient-facing side-car application to display RA outcomes (disease activity, functional status, pain scores), medications, and lab results during clinical visits ("RA PRO Dashboard"). The study aimed to evaluate patient perceptions and attitudes towards the implementation of a novel patient-facing dashboard during clinical visits using a mixed-methods approach.

Methods: RA patients whose clinicians used the dashboard at least once during their clinical visit were invited to complete a survey regarding its usefulness in care. We also conducted semi-structured interviews with a subset of patients to assess their perceptions of the dashboard. The interviews were transcribed verbatim and analyzed thematically using deductive and inductive techniques. Emerging themes and subthemes were organized into four domains of the Ecological Model of Health.

Results: Out of 173 survey respondents, 79% were interested in seeing the dashboard again at a future visit, 71% felt it improved their understanding of their disease, and 65% believed it helped with decision-making about their RA care. Many patients reported that the dashboard helped them discuss their RA symptoms (76%) and medications (72%) with their clinician. Interviews with 29 RA patients revealed 10 key themes: the dashboard was perceived as a valuable visual tool that improved patients' understanding of RA outcome measures, enhanced their involvement in care, and increased their trust in clinicians and the clinic. Common reported limitations included concerns about reliability of RA outcome questionnaires for some RA patients and inconsistent collection and explanation of these measures by clinicians.

Conclusions: In both the quantitative and qualitative components of the study, patients reported that the dashboard improved their understanding of their RA, enhanced patient-clinician communication, supported shared decision-making, and increased patient engagement in care. These findings support the use of dashboards or similar data visualization tools in RA care and can be used in future interventions to address challenges in data collection and patient education.