Search Results (177)

Respiratory diseases represent a persistent global health challenge, underscoring the need for intelligent, accurate, and personalized diagnostic and therapeutic systems. Existing methods frequently suffer from limitations in diagnostic precision, lack of individualized treatment, and constrained adaptability to complex clinical scenarios. To address these challenges, our study introduces a modular AI-powered framework that integrates an audio-based disease classification model with simulated molecular biomarker profiles to evaluate the feasibility of future multimodal diagnostic extensions, alongside a synthetic-data-driven prescription recommendation engine. The disease classification model analyzes respiratory sound recordings and accurately distinguishes among eight clinical classes: bronchiectasis, pneumonia, upper respiratory tract infection (URTI), lower respiratory tract infection (LRTI), asthma, chronic obstructive pulmonary disease (COPD), bronchiolitis, and healthy respiratory state. The proposed model achieved a classification accuracy of 99.99% on a holdout test set, including 94.2% accuracy on pediatric samples. In parallel, the prescription module provides individualized treatment recommendations comprising drug, dosage, and frequency trained on a carefully constructed synthetic dataset designed to emulate real-world prescribing logic.The model achieved over 99% accuracy in medication prediction tasks, outperforming baseline models such as those discussed in research. Minimal misclassification in the confusion matrix and strong clinician agreement on 200 prescriptions (Cohen’s $\kappa$ = 0.91 [0.87–0.94] for drug selection, 0.78 [0.74–0.81] for dosage, 0.96 [0.93–0.98] for frequency) further affirm the system’s reliability. Adjusted clinician disagreement rates were 2.7% (drug), 6.4% (dosage), and 1.5% (frequency). SHAP analysis identified age and smoking as key predictors, enhancing model explainability. Dosage accuracy was 91.3%, and most disagreements occurred in renal-impaired and pediatric cases. However, our study is presented strictly as a proof-of-concept. The use of synthetic data and the absence of access to real patient records constitute key limitations. A trialed clinical deployment was conducted under a controlled environment with a positive rate of satisfaction from experts and users, but the proposed system must undergo extensive validation with de-identified electronic medical records (EMRs) and regulatory scrutiny before it can be considered for practical application. Nonetheless, the findings offer a promising foundation for the future development of clinically viable AI-assisted respiratory care tools. Full article

Background: Semaglutide and Liraglutide are medications in the Glucagon-like peptide-1 agonists (GLP-1 RAs) class used to manage type 2 diabetes mellitus and obesity in Saudi Arabia. Although the 1.0 mg once weekly dosage of Semaglutide does not have a labeled indication for the management of obesity, many believe that this dosage is more effective than the 3.0 mg once daily Liraglutide dosage for the management of both diabetes and obesity. Objective: To compare the effectiveness of the dosage of 1.0 mg of Semaglutide administered once weekly versus 3.0 mg of Liraglutide administered once daily in controlling HbA1c levels, promoting weight loss, and evaluating their financial implications among obese patients in Saudi Arabia using real-world data. Methods: A retrospective review of Electronic Medical Records (EMRs) from January 2021 to June 2024 was conducted on patients prescribed Semaglutide or Liraglutide for at least 12 months. Exclusion criteria included pre-existing severe conditions (e.g., cardiovascular disease, stroke, or cancer) and missing baseline data. The primary outcomes assessed were changes in HbA1c, weight, and direct medical costs. Results: Two hundred patients (100 patients on the 1.0 mg once weekly dose of Semaglutide and 100 patients on the 3.0 mg once daily dose of Liraglutide) of those randomly selected from the EMRs met the inclusion criteria and were included in the analysis. Of the 200 eligible patients (65.5% female, mean age 48.54 years), weight loss was greater with Semaglutide (−8.09 kg) than Liraglutide (−5.884 kg). HbA1c reduction was also greater with Semaglutide (−1.073%) than Liraglutide (−0.298%). The use of Semaglutide resulted in lower costs of USD −1264.76 (95% CI: −1826.82 to 33.76) and greater reductions in weight of −2.22 KG (95% CI: −7.68 to −2.784), as well as lower costs of USD −1264.76 (95% CI: (−2368.16 to −239.686) and greater reductions in HbA1c of −0.77% (95% CI: −0.923 to −0.0971) in more than 95% of the cost effectiveness bootstrap distributions. Conclusions: Semaglutide 1.0 mg weekly seems to be more effective and cost-saving in managing prediabetes, diabetes, and obesity compared to Liraglutide 3.0 mg daily. Future studies should examine these findings using a more representative sample and a robust study design. Full article

Pharmacogenomics (PGx) has emerged as a powerful tool to personalize drug selection and dosing based on a patient’s genetic profile. However, there are a range of challenges that impede uptake in current clinical practice. For example, clinicians often express frustration with commercially available PGx panel tests, which fail to consistently include all key actionable PGx genes (according to the Clinical Pharmacogenetics Implementation Consortium (CPIC), Food and Drug Administration (FDA) PGx guidelines, or The Dutch Pharmacogenetics Working Group (DPWG) guidelines) and instead are too long with clinically unimportant information (unvalidated genotypes). Additionally, the lack of EMR integration, clinician education and awareness of the benefits of PGx impedes uptake. This paper examines key challenges identified in clinical practice and proposes future directions, focusing on limiting PGx reports to essential data, providing point-of-prescription alerts, and establishing reimbursement pathways that encourage adoption. Future directions include leveraging large language models, integrating point-of-prescription alerts and phenoconversion calculators into the electronic medical record, increasing the genomic diversity of PGx study populations, and streamlining coverage by payers. Full article

In recent years, Electronic Medical Record (EMR) sharing has played an indispensable role in optimizing clinical treatment plans, advancing medical research in biomedical science. However, existing EMR management schemes often face security risks and suffer from inefficient search performance. To address these issues, this paper proposes a secure EMR sharing scheme based on blockchain and searchable encryption. This scheme implements a decentralized management system with enhanced security and operational efficiency. Considering the scenario of EMRs requiring confirmation of multiple doctors to improve safety, the proposed solution leverages Shamir’s Secret Sharing to enable multi-party authorization, thereby enhancing privacy protection. Meanwhile, the scheme utilizes Bloom filter and vector operation to achieve efficient data search. The proposed method maintains rigorous EMR protection while improving the search efficiency of EMRs. Experimental results demonstrate that, compared to existing methodologies, the proposed scheme enhances security during EMR sharing processes. It achieves higher efficiency in index generation and trapdoor generation while reducing keyword search time. This scheme provides reliable technical support for the development of intelligent healthcare systems. Full article

Background: Clostridioides difficile (CD) is the main cause of nosocomial diarrhea, resulting in increased morbidity and mortality, and is thought to be greatly affected by strict hygiene. In this study, we assessed changes in CD infection prevalence and outcomes pre- and during the COVID-19 pandemic (CP). Methods: This was an observational cohort performed at a tertiary medical center (MC) and a geriatric hospital (GH). Patients from both hospitals diagnosed with CD were included, and the period of one year prior to the pandemic to one year after was compared. Data was extracted from electronic medical records (EMR). Results: A total of 145 CD-associated diarrhea (CDAD) cases were diagnosed in the MC and 54 in the GH. There was no change in CDAD prevalence or mortality between the study periods in either hospital. Disease duration, measured as days with diarrhea (DWD), was shorter during the CP in the GH (10.6 days vs. 8.1 days, p < 0.01). CDAD was more prevalent in the GH during both periods; however, the disease was milder, with only three mortality cases and a significantly shorter disease duration (3.19 DWD vs. 10.67 in the MC before CP; 3.11 vs. 8.1 during CP, p < 0.01). In a survival analysis for MC patients, no significant differences were found between periods before and after adjustment for age, gender and period. Conclusions: The CP affected the duration but not the prevalence of CDAD. The milder course of CDAD in the GH may have been due to the quality of treatment provided in an academic GH and the subsequent faster diagnosis and treatment. Full article

Background/Objectives: Tramadol is an opioid frequently prescribed for moderate to severe pain and has seen a global increase in use. This presents numerous challenges in clinical management. This study aims to elucidate metabolic signatures associated with tramadol consumption, enhancing predictive capabilities for therapeutic outcomes and optimizing patient-specific treatment plans. Methods: Data were obtained from the Qatar Biobank (QBB), focusing on pharmacogenomic variants associated with tramadol use and prescription trends. A cohort of 27 individuals who were administered daily tramadol doses between 100 and 400 mg with available metabolomic profiles were selected. The pharmacokinetics of tramadol were evaluated in relation to specific CYP2D6 genetic variants. Comparative pharmacometabolomic profiles were generated for tramadol users versus a control group of 54 non-users. Additionally, prescription data encompassing tramadol formulations were collected from the electronic medical records (EMR) system of the major public hospital network in Qatar (Hamad Medical Corporation) to discern prescribing patterns. Results: From January 2019 to December 2022, tramadol prescriptions varied, with chronic pain as the primary indication, followed by acute pain. Pharmacogenomic analysis indicated that CYP2D6 allele variations significantly impacted tramadol and O-desmethyltramadol glucuronide levels, notably in ‘normal metabolizers’. Metabolomic analysis revealed distinct metabolic profiles in tramadol users, with significant variations in phosphatidylcholine, histidine, and lysine pathways compared to controls, highlighting tramadol’s unique biochemical impacts. Conclusions: This study underscores the importance of integrating genetic and omics-based approaches to enhance tramadol’s efficacy and safety. These findings support personalized pain management strategies, enhancing treatment outcomes for both chronic and acute pain. Full article

Despite having the world’s largest HIV burden, South Africa has yet to attain the 95-95-95 targets. Accurate, complete, and timely data are critical for monitoring a country’s HIV progress. The integration of unique identifier codes (UICs) for key populations (KPs) into routine health information management systems (RHIMS) strengthens data accuracy and completeness, facilitating more targeted HIV interventions and greater accountability. This systematic review assessed how Sub-Saharan African (SSA) countries have integrated KPs’ UICs into RHIMS, highlighting key enablers, challenges, and opportunities. A comprehensive search was conducted across PubMed, Scopus, Google Scholar, MEDLINE, PLOS ONE, and various government and non-government websites to identify the published and grey literature relevant to the study objective from June 2013 to December 2024. References were managed using Zotero version 6.0.36. Two authors independently screened studies using Covidence software. The review was done in accordance with the “Preferred Reporting Items for Systematic Reviews and Meta-Analyses” (PRISMA) guidelines and was registered with the “International Prospective Register of PROSPERO) Systematic Reviews” with the registration number CRD42023440656. Out of 1735 studies screened, 361 duplicates were removed. The review found that only nine of the fifty-three SSA countries have incorporated UICs for KPs into their RHIMS through alphanumeric codes. They include Burundi, Burkina Faso, Ghana, Mali, Kenya, Uganda, Togo, Malawi, and Liberia. Facilitators for KPs’ UIC adoption included strong data security and political will, whereas barriers encompassed compromised privacy, stigma and discrimination. In South Africa, the UIC for KPs can be integrated into the new electronic medical record (EMR) system. Full article

Background/Objectives: Diagnostic reasoning is essential in clinical practice and medical education, yet it often becomes an automated process, making its cognitive mechanisms less visible. Despite the widespread use of electronic medical records, few studies have quantitatively evaluated how clinicians’ reasoning is documented in real-world electronic medical records. This study aimed to investigate whether initial electronic medical records contain valuable information for diagnostic reasoning and assess the feasibility of using text analysis and logistic regression to make this reasoning process visible. Methods: We conducted a retrospective analysis of initial electronic medical records at Kochi University Hospital between 2008 and 2022. Two patient cohorts presenting with dizziness and headaches were analysed. Text analysis was performed using GiNZA, a Japanese natural language processing library, and logistic regression analyses were conducted to identify associations with final diagnoses. Results: We identified 1277 dizziness cases, of which 248 were analysed, revealing 48 significant diagnostic terms. Moreover, we identified 1904 headache cases, of which 616 were analysed, revealing 46 significant diagnostic terms. The logistic regression analysis demonstrated that the presence of specific terms, as well as whether they were expressed affirmatively or negatively, was significantly associated with diagnostic outcomes. Conclusions: Initial EMRs contain quantifiable linguistic cues relevant to diagnostic reasoning. Even simple analytical methods can reveal reasoning patterns, offering valuable insights for medical education and supporting the development of explainable diagnostic support systems. Full article

Despite remarkable advances in neural language models, a substantial gap remains in precisely interpreting the complex semantics of Electronic Medical Records (EMR). We propose Contrastive Representations Pre-Training (CRPT) to address this gap, replacing the conventional Next Sentence Prediction task’s cross-entropy loss with contrastive loss and incorporating whole-word masking to capture multi-token domain-specific terms better. We also introduce a carefully designed negative sampling strategy that balances intra-document and cross-document sentences, enhancing the model’s discriminative power. Implemented atop a BERT-based architecture and evaluated on the Biomedical Language Understanding Evaluation (BLUE) benchmark, our Discharge Summary CRPT model achieves significant performance gains, including a natural language inference precision of 0.825 and a sentence similarity score of 0.775. We further extend our approach through Bio+Discharge Summary CRPT, combining biomedical and clinical corpora to boost downstream performance across tasks. Our framework demonstrates robust interpretive capacity in clinical texts by emphasizing sentence-level semantics and domain-aware masking. These findings underscore CRPT’s potential for advancing semantic accuracy in healthcare applications and open new avenues for integrating larger negative sample sets, domain-specific masking techniques, and multi-task learning paradigms. Full article

Introduction: Vertebral Compression Fractures (VCF) are the most common vertebral fractures, usually osteoporotic, with rising incidence. The natural history of VCFs-related pain remains unclear, and treatment protocols are still being evaluated, ranging from conservative to surgical. Patient-reported measures have been proven inaccurate and carry significant biases. This study examines maximal tenderness location (MTL) to palpation and percussion on physical examination during VCF healing and the postoperative period. Methods: A prospective study included 40 patients treated for VCFs per the NICE guidelines (2013) from 2019 to 2021. Treatment was either conservative (n = 12) or surgical (n − 28), Balloon Kyphoplasty (BKP). All patients’ MTL were recorded in EMR (Electronic Medical Record) on every visit. BKP was offered for severe ongoing pain after a recent, unhealed vertebral fracture despite optimal pain management, progressive fracture collapse, or lack of union. Follow-up was six months. Pain evolution was analyzed using Kaplan–Meier survival curves, Log-Rank tests, Mann–Whitney U tests, t-tests, and logistic regression models. A p-value < 0.05 was considered statistically significant. Results: 12 patients were treated conservatively, and 28 underwent BKP for T12-L2 VCFs, accounting for 75% of fractures, mostly single-level fractures. All initially suffered MTL over the VCF; BKP patients showed local VCF pain resolution after 3.5 weeks following surgery while lasting seven weeks under conservative treatment. Lumbosacral pain was more prevalent following BKP (OR = 4, p = 0.05) and developed earlier. Conclusions: This study is novel in relating physical examination findings to fracture age and treatment provided, suggesting that VCFs-related pain is a time-related shift from local fracture pain to lumbosacral pain. Patient-reported pain scales may not reliably distinguish between these varying pain patterns. These findings suggest that only local VCF pain should be considered for surgical treatment. Future studies evaluating VCF outcomes should address physical examination and not rely solely on patient-reported metrics. Full article

Tuberculosis (TB) is an infectious disease that has been declared a global emergency by the World Health Organization and remains one of the top ten causes of death worldwide. TB diagnosis is particularly challenging in developing countries, where limited infrastructure for detection and treatment complicates efforts to control the disease. These resource constraints are especially critical in remote areas with few mechanisms for timely diagnosis, which is essential for effective patient management. Artificial intelligence (AI) has emerged as a valuable tool in supporting health professionals by enhancing diagnostic processes. This paper explores the use of natural language processing (NLP) techniques and machine learning (ML) models to facilitate TB diagnosis in settings where robust data infrastructure is unavailable. Two distinct data sources were analyzed: text extracted from electronic medical records (EMRs) and patient clinical data (CD). Four different ML-based approaches were implemented: two models using each data source independently and two data fusion models combining both sources. The relevance of these strategies was assessed in collaboration with physicians to ensure their practical applicability in clinical decision-making. The results of the data fusion models were compared to determine which source provided more valuable diagnostic information. The best-performing model, which relied solely on CD, achieved a sensitivity of $73\%$, outperforming smear microscopy, which typically ranges from $40\%$ to $60\%$. These findings underscore the importance of analyzing physicians’ reports and assessing the availability of such information alongside structured clinical data. This approach is particularly beneficial in resource-limited settings, where access to comprehensive clinical data may be restricted. Full article

Electronic medical records (EMRs) are fundamental to clinical decision support systems (CDSS). Conventional EMR structures still fail to capture the cyclical nature of treatment plans, leading to fragmented data representation and reduced decision accuracy. This study addresses this gap by proposing a cycle-based EMR framework that systematically integrates treatment cycles, enabling structured, sequential data organization. Treatment cycles categorize patient data into iterative phases, reflecting disease progression and repeated interventions, ensuring data continuity and analytical precision. A dataset inspired by MIMIC-III was developed to empirically evaluate this approach, incorporating treatment cycle fields to enhance data continuity and analytical precision. The results indicate that cycle-based structuring preserves critical variations in patient responses, improves treatment outcome predictions, and strengthens CDSS recommendations. While this approach offers substantial benefits, challenges such as workflow adaptation, usability, and interoperability must be addressed to facilitate seamless integration into clinical practice. Despite these challenges, this study establishes a scientifically validated foundation for cycle-based EMRs, aligning data structures with real-world clinical workflows. By rectifying data organization, this approach elevates diagnostic accuracy, optimizes treatment planning, and enhances patient outcomes, contributing to the future of precision medicine. Full article

Recent advancements in electronic medical records (EMRs) have transformed clinical trials and healthcare systems by improving data accuracy, regulatory compliance, and integration with decision support tools. These innovations enhance trial efficiency, streamline patient recruitment, and enable large-scale data analysis while bridging clinical practice with research. Despite these benefits, challenges such as data standardization, privacy concerns, and usability issues persist. Overcoming these barriers through policy reforms, technological innovations, and robust methodologies is essential to maximizing the potential of EMRs. We examine current developments, challenges, and future directions for optimizing EMRs in clinical trials and healthcare delivery. Full article

This systematic study seeks to evaluate the use and impact of transformer models in the healthcare domain, with a particular emphasis on their usefulness in tackling key medical difficulties and performing critical natural language processing (NLP) functions. The research questions focus on how these models can improve clinical decision-making through information extraction and predictive analytics. Our findings show that transformer models, especially in applications like named entity recognition (NER) and clinical data analysis, greatly increase the accuracy and efficiency of processing unstructured data. Notably, case studies demonstrated a 30% boost in entity recognition accuracy in clinical notes and a 90% detection rate for malignancies in medical imaging. These contributions emphasize the revolutionary potential of transformer models in healthcare, and therefore their importance in enhancing resource management and patient outcomes. Furthermore, this paper emphasizes significant obstacles, such as the reliance on restricted datasets and the need for data format standardization, and provides a road map for future research to improve the applicability and performance of these models in real-world clinical settings. Full article

Thyroid Cancer (TC) is one of the most prevalent endocrine malignancies, with early detection being critical for patient management. The motivation for integrating Machine Learning (ML) in thyroid cancer research stems from the limitations of conventional diagnostic and monitoring approaches, as ML offers transformative potential for reducing human errors and improving prediction outcomes for diagnostic accuracy, risk stratification, treatment options, recurrence prognosis, and patient quality of life. This scoping review maps existing literature on ML applications in TC, particularly those leveraging clinical data, Electronic Medical Records (EMRs), and synthesized findings. This study analyzed 1231 papers, evaluated 203 full-text articles, selected 21 articles, and detailed three themes: (1) malignancy prediction and nodule classification; (2) other metastases derived from TC prediction; and (3) recurrence and survival prediction. This work examined the case studies’ characteristics and objectives and identified key trends and challenges in ML-driven TC research. Finally, this scoping review addressed the limitations of related and highlighted directions to enhance the clinical potential of ML in this domain while emphasizing its capability to transform TC patient care into advanced precision medicine. Full article