BMC Medical Informatics and Decision Making最新文献

{"title":"Gold-standard-guided direct preference optimisation for ICD-10-AM and ACHI code assignment from discharge summaries.","authors":"Rajvir Kaur, Jeewani Anupama Ginige, Oliver Obst","doi":"10.1186/s12911-026-03631-w","DOIUrl":"https://doi.org/10.1186/s12911-026-03631-w","url":null,"abstract":"<p><p>Clinical codes assigned from discharge summaries support clinical research, reimbursement, health resource allocation, and service planning. In Australia, ICD-10-AM and ACHI are used to code diagnoses and interventions for admitted episodes of care, but manual coding is time-consuming, labour-intensive, and subject to variability. This study investigates large language models for automated ICD-10-AM and ACHI code assignment from discharge summaries. We first normalised a public ICD-9-CM coded dataset into the Australian coding system using General Equivalence Mappings and semantic similarity over code descriptions, with validation by a domain expert clinical coder. We then fine-tuned Meta-Llama-3-8B and compared full-parameter fine-tuning with a parameter-efficient QLoRA baseline. Finally, we applied Direct Preference Optimisation (DPO) using gold-standard-guided preference pairs, where the gold-standard ICD-10-AM/ACHI code list was treated as the preferred response and the supervised fine-tuned model output as the non-preferred response. Experiments were conducted on a mapped MIMIC-III subset and a de-identified Australian hospital dataset. Among the evaluated model variants, the DPO-aligned full-parameter model achieved the highest score, with micro-F1 scores of 0.7397 on the mapped dataset and 0.7834 on the Australian hospital dataset. These findings suggest that gold-standard-guided DPO can improve agreement with reference code sets compared with supervised fine-tuning alone. However, the results are limited to high-frequency diagnosis and procedure codes, short-document evaluation settings, and DPO preference pairs reviewed by one annotator.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"26 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149227325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-powered spectral CT analysis for clinical decision support in carotid vulnerable plaque detection: a deep learning approach.","authors":"Yunzhe Ni, Tianyu Zhang, Zonghui Huang, Yuguang Wang, Guochao Han, Yuan Lin, Bo Liu, Wennan Lin, Hui Li","doi":"10.1186/s12911-026-03611-0","DOIUrl":"https://doi.org/10.1186/s12911-026-03611-0","url":null,"abstract":"<p><strong>Background: </strong>Carotid vulnerable plaques (CVPs) represent a major cause of ischemic stroke, yet current diagnostic methods lack sufficient precision for early detection. Spectral computed tomography (CT) enables detailed plaque characterization, but its clinical utility depends on advanced analytical techniques.</p><p><strong>Methods: </strong>A Fourier Transform-enhanced U-Net model was developed using spectral CT data from 214 patients at Qiqihar Medical College. Transfer learning was applied for carotid segmentation, supported by extensive data augmentation and ten-fold cross-validation. Model performance was assessed using sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC). A patient-level independent test set was reserved to evaluate model performance.</p><p><strong>Results: </strong>The FTC-enhanced U-Net achieved a Dice similarity coefficient (DSC) of 0.89 and an intersection over union (IoU) of 0.80, accurately delineating arterial and plaque boundaries. Stable classification performance was maintained across plaque risk levels, with accuracies of 86% for low-risk, 83% for medium-risk, and 87% for high-risk plaques. Ten-fold cross-validation yielded an AUC of 0.88 ± 0.03, an accuracy of 85.3% ± 2.1%, and an F1-score of 0.84 ± 0.02, confirming strong robustness and generalizability.</p><p><strong>Conclusions: </strong>Artificial intelligence (AI)-based predictive models derived from spectral CT imaging show substantial potential for early diagnosis and prognostic assessment of CVPs. The FTC-enhanced U-Net provides an efficient and reliable tool for clinical decision support, facilitating broader adoption of AI in vascular imaging and improving diagnostic precision and patient management.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"26 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149223390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and temporal validation of a machine learning-based model for predicting renal impairment in multiple myeloma: a single-center retrospective cohort study.","authors":"Manli Zhou, Sisi Feng","doi":"10.1186/s12911-026-03689-6","DOIUrl":"https://doi.org/10.1186/s12911-026-03689-6","url":null,"abstract":"<p><strong>Background: </strong>Multiple myeloma (MM) is a hematopoietic system malignancy characterized by clonal proliferation of abnormal plasma cells, commonly presenting with renal impairment (RI) that significantly impacts patient's quality of life. The objective of this study was to develop a predictive model for assessing the risk of MM with RI.</p><p><strong>Methods: </strong>This retrospective study included 1029 patients with MM, who were categorized into two groups: MM without RI group and MM with RI group. Specifically, the modeling dataset (Cohort 1) consisted of 792 patients (401 without RI and 391 with RI), and the validation dataset (Cohort 2) consisted of 237 patients (119 without RI and 118 with RI). Eight machine learning (ML) algorithms were assessed using the modeling dataset (n = 792), and model validation was conducted on the validation dataset (n = 237). The predictive performance of ML models was evaluated through the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis.</p><p><strong>Results: </strong>The optimal model among the eight ML algorithms was XGBoost. The predictive model constructed based on the XGBoost algorithm, RI-Lab6, incorporated six features: NLR, LDH, Hb, ACa, AGR and β2 microglobulin. In the modeling dataset (cohort 1), RI-Lab6 demonstrated an AUC of 0.99 (95% CI, 0.98-0.99), while in the validation dataset (cohort 2), it achieved an AUC value of 0.98 (95% CI, 0.96-0.99).</p><p><strong>Conclusions: </strong>These results indicated that the RI-Lab6 model possessed reliable predictive value for RI risk in MM patients and facilitated early identification of RI in these individuals.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"26 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13625446/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149104452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Measuring what matters- patient and clinician priorities for clinical decision support interactions: a modified Delphi study.","authors":"Rachel Kurtzman, Mustafa Ozkaynak, Frances Jimenez, Polina Kukhareva, Ruth Nwefo, Priyanka Desai, Prashila Dullabh","doi":"10.1186/s12911-026-03715-7","DOIUrl":"https://doi.org/10.1186/s12911-026-03715-7","url":null,"abstract":"<p><strong>Background: </strong>Patient-centered clinical decision support (PC CDS) is a spectrum of digital tools intended to support patient and clinician decision making. Patient-clinician interaction is key to PC CDS, and patient-clinician interaction measurement areas assess how PC CDS impacts the care process and experiences a patient has accessing and receiving healthcare. Identifying measures that assess if and how PC CDS influences outcomes is essential to measuring impact, but it is unclear what outcomes patients and physicians consider important. This study aimed to develop a prioritized list of measurement areas informed by patients and primary care physicians (PCPs) to inform considerations for measuring patient-clinician interaction for PC CDS.</p><p><strong>Methods: </strong>We conducted a modified Delphi study to understand patients' and PCPs' perspectives on the importance of fifteen previously identified patient-clinician interaction measurement areas for PC CDS. We conducted a three-round process with nine patients and nine PCP participants: (1) rating importance of measurement areas, (2) online discussion of ratings, and (3) forced-choice ranking activity of measurement areas.</p><p><strong>Results: </strong>Patients and PCPs generally agreed about the importance of communication quality and trust in clinicians, and the interconnectedness of these areas. They additionally agreed that the relative importance of measurement areas depends on context and described contextual factors such as care expectations and patient complexity. Patients and PCPs also agreed that satisfaction, decision regret and discharge preparedness are of lower importance for PC CDS measurement. However, patients and PCPs had differing views on the importance of measuring access to information, decision quality and self-management. Patients ranked access to information higher than PCPs, while PCPs ranked decision quality and self-management higher than patients.</p><p><strong>Conclusions: </strong>Assessments of PC CDS effectiveness should consider what outcomes are most important to CDS end-users. Measurement of PC CDS is limited by logistics and feasibility considerations, and researchers should examine the outcomes assessed in PC CDS studies and identify opportunities to assess measurement areas important to key stakeholders. The results of this activity highlight areas for future work, including the importance of understanding how measurement of these areas may be impacted by context (e.g. patient complexity, disease status).</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"26 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149223345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Diagnostic performance of machine learning and deep learning algorithms for thyroid cancer metastasis: a systematic review and meta-analysis.","authors":"Mohammad Amouzadeh Lichahi, Saeid Anvari, Hossein Hemmati, Ervin Zadgari, Maryam Jafari, Seyedeh Mohadeseh Mosavi Mirkalaie, Mohaya Farzin, Amirhossein Larijani","doi":"10.1186/s12911-026-03708-6","DOIUrl":"10.1186/s12911-026-03708-6","url":null,"abstract":"","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"26 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13393856/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148560482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A predictive ant colony optimization routing protocol for correlated link stability in wireless body area networks.","authors":"Ben Othman Soufiane, Naima Saeed","doi":"10.1186/s12911-026-03706-8","DOIUrl":"10.1186/s12911-026-03706-8","url":null,"abstract":"<p><strong>Background: </strong>Wireless Body Area Networks (WBANs) are essential for continuous health monitoring but face critical challenges due to dynamic link instability caused by body movements and strict bio-safety requirements. The fluctuating wireless channel leads to high packet loss, while radio frequency (RF) energy absorption, measured as Specific Absorption Rate (SAR), must remain below 1.6 W/kg to prevent tissue damage. Existing routing protocols often react to failures rather than predict them and rarely co-optimize reliability with thermal safety.</p><p><strong>Methods: </strong>This paper proposes PACO-CLS (Predictive Ant Colony Optimization with Correlated Link Stability), a novel routing protocol that proactively addresses both network dynamics and bio-safety. It leverages tri-axial accelerometer data from sensor nodes to classify body posture using a Gaussian Mixture Model (GMM). This posture state conditions a Vector Autoregressive (VAR) model, which predicts future link quality metrics (SNR, distance, SAR) over a 0.5-second horizon. These predictions drive an enhanced Ant Colony Optimization (ACO) algorithm, where path selection is guided by a multi-objective cost function. Crucially, the pheromone evaporation rate is dynamically modulated by the cumulative SAR, calculated via the Pennes bio-heat equation, to enforce thermal safety.</p><p><strong>Results: </strong>Extensive simulations in NS-3.40, calibrated with empirical data from the MySignals HW V3 platform, demonstrate that PACO-CLS significantly outperforms QC-TriL, LEACH-WBAN, and Q-Learning-based routing. It achieves a network lifetime of [Formula: see text] hours (a 40.3% improvement), an energy efficiency of [Formula: see text] Mb/J, a packet delivery ratio of [Formula: see text], and end-to-end latency of [Formula: see text] ms. Critically, it maintains an average SAR of [Formula: see text] W/kg, well below the safety threshold across all tested scenarios, including high interference ([Formula: see text] sources/m²) and mobility ([Formula: see text] m/s).</p><p><strong>Conclusions: </strong>By integrating predictive analytics with thermal-aware mechanisms, PACO-CLS establishes a new benchmark for intelligent WBAN routing. Its ability to proactively avoid unstable links based on correlated motion analytics and simultaneously enforce bio-safety constraints makes it a robust and clinically viable solution. PACO-CLS consistently meets stringent clinical requirements for real-time monitoring (PDR > 95%, [Formula: see text][Formula: see text] ms, SAR [Formula: see text] W/kg, [Formula: see text] h), demonstrating its potential for next-generation, safe, and reliable healthcare applications.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"26 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149219818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Based on the middle ear negative pressure and multimodal data to construct and externally validate the predictive model for pediatric obstructive sleep apnea.","authors":"Simin Zhu, Yewen Shi, Yanuo Zhou, Lina Ma, Yonglong Su, Zitong Wang, Chendi Lu, Zine Cao, Xiaoxin Niu, Yushan Xie, Zihan Xia, Huanhuan Chang, Yuqi Yuan, Jiayi Yang, Rui Lu, Haiqin Liu, Xiaoyong Ren, Wei Hou","doi":"10.1186/s12911-026-03696-7","DOIUrl":"10.1186/s12911-026-03696-7","url":null,"abstract":"<p><strong>Purpose: </strong>The aim of this study was to analyze negative middle ear pressure in children with obstructive sleep apnea (OSA) and establish and evaluate predictive models according to these findings.</p><p><strong>Methods: </strong>This retrospective study involved 931 children: 715 with OSA and 216 controls. Demographic, clinical, lateral head radiograph, and tympanometry data were collected. These characteristics of children with OSA were analyzed, with a particular focus on exploring the value of middle ear-related parameters for the diagnosis of pediatric OSA. Additionally, a logistic regression model incorporating optimal indicators was developed to predict pediatric OSA. The model was visualized via a nomogram and evaluated for discrimination, calibration, clinical effectiveness and external validation.</p><p><strong>Results: </strong>Children with OSA were younger and exhibited longer soft palates, larger tonsils and adenoids than non-OSA children. Additionally, children with OSA presented higher acoustic admittance (AC) and resonance frequency (RF), lower middle ear pressure (MEP), and narrower pressure gradient (PG) than non-OSA children. The external auditory canal volume (ECV), MEP, and PG were identified as independent predictors of childhood OSA. We constructed a foundational prediction model for childhood OSA (Model 0, AUC = 0.845, 95% CI: 0.813-0.878), and then added each tympanometric indicator to the model individually. After incorporating MEP into the model (Model 4), the AUC increased by 0.022 (p < 0.05).</p><p><strong>Conclusions: </strong>Based on a large sample size and multivariate analysis of factors associated with pediatric OSA, we developed a predictive model incorporating middle ear negative pressure for pediatric OSA, which may assist clinicians in diagnosing pediatric OSA in complex clinical settings.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":" ","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13673895/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148547897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Provider perspectives of clinical decision support for chronic kidney disease management: a cross-sectional study.","authors":"Robert Goldberg, Henry Li, Cory E Goldstein, Kevin Yau, Shannon M Ruzycki, Maoliosa Donald, Meghan J Elliott, Brenda R Hemmelgarn, Matthew T James, Pietro Ravani, Neesh Pannu, Derek Chew, Dilaram Acharya, Tyrone G Harrison","doi":"10.1186/s12911-026-03721-9","DOIUrl":"10.1186/s12911-026-03721-9","url":null,"abstract":"<p><strong>Background: </strong>Chronic kidney disease (CKD) affects approximately one in ten adults globally. Multiple studies have identified underuse of effective therapies in eligible patients with CKD despite guideline endorsement. Clinical decision support (CDS) may increase confidence in prescribing recommended medications, but how end-users perceive it is unknown.</p><p><strong>Objectives: </strong>To evaluate clinician-reported perspectives and acceptability of CDS elements for ambulatory CKD management and inform the development of a CKD-specific CDS intervention.</p><p><strong>Methods: </strong>Eligible participants for this electronic cross-sectional study were prescribing clinicians from Alberta, Canada, that practiced in ambulatory care settings, including physicians, nurse practitioners, and pharmacists. The survey presented examples of a proposed CDS' content and format, and collected information on clinician demographics, CKD workflows, perceptions of CDS design, perceived acceptability and clinical utility. Knowledge, comfort, and prescribing patterns were examined for renin-angiotensin system inhibitors (RASi), statins, sodium-glucose cotransporter-2 inhibitors (SGLT2i), glucagon-like peptide-1 receptor agonists (GLP-1RA), and non-steroidal mineralocorticoid receptor agonists (nsMRA). Participant comfort prescribing each medication class was compared using mean paired differences before and after viewing the proposed CDS.</p><p><strong>Results: </strong>Between May and June 2025, 137 clinicians engaged with the survey, 131 completed demographics and 111 completed all survey elements. Most were nephrologists (n = 42; 32%) or general internal medicine specialists (n = 36; 27%). Overall, 88 (79%) supported CDS integration. RASi had the highest self-reported comfort with prescribing followed by statins, SGLT2i, GLP-1RA, and nsMRA. Clinicians reported higher comfort prescribing GLP-1RA and nsMRA with CDS support (median +1 on 9-point scale; p < 0.001), but no change for RASi, statins, and SGLT2i. Participants with low baseline comfort (≤6 on 9-point Likert scale) had statistically significant improvements in comfort with CDS for GLP-1RA and nsMRA (p < 0.05), but no change was observed for those with high baseline comfort.</p><p><strong>Conclusions: </strong>Clinicians support CDS integration to enhance CKD care in ambulatory settings. The proposed CDS increased perceived prescribing confidence, particularly for GLP-1RA and nsMRA.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":" ","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13673755/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148547890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-assisted decision support for sickle cell disease severity stratification using routine blood tests: a systematic review and meta-analysis.","authors":"Naif Taleb Ali, Mansour Abdulnabi H Mehdi, Radfan Saleh Abdullah, Gamila Saleh Ali, Hana Mohsen Ali, Ali N M Gubran, Nazeh Mohammed Al-Abd","doi":"10.1186/s12911-026-03679-8","DOIUrl":"10.1186/s12911-026-03679-8","url":null,"abstract":"<p><strong>Background: </strong>Stratifying sickle cell disease (SCD) severity remains challenging, particularly in resource-limited settings. Artificial intelligence (AI) models using routine complete blood count (CBC) parameters have been proposed as accessible tools for risk stratification; however, their overall performance and clinical applicability remain uncertain.</p><p><strong>Methods: </strong>We conducted a PRISMA-DTA 2020-compliant systematic review and meta-analysis (PROSPERO: CRD420251078389) including 60 studies (25,354 patients across 15 countries, 2010-2025). AI models based on CBC parameters were evaluated against heterogeneous reference standards, including clinical severity classifications and event-based outcomes. Risk of bias was assessed using QUADAS-2 and PROBAST-AI. Pooled estimates were generated using hierarchical summary receiver operating characteristic (HSROC) models.</p><p><strong>Results: </strong>The pooled area under the curve (AUC) was 0.87 (95% CI: 0.84-0.90), with sensitivity 0.83 (0.79-0.86) and specificity 0.85 (0.81-0.88). Performance varied by setting, with higher accuracy in high-income countries (AUC 0.90) compared with low- and middle-income settings (AUC 0.82; p < 0.001). Red cell distribution width and platelet count were consistently identified as important predictors. However, substantial heterogeneity in outcome definitions and reference standards limits interpretability.</p><p><strong>Conclusions: </strong>AI models using routine CBC parameters demonstrate promising analytical performance for SCD risk stratification. However, the absence of a universally accepted severity gold standard, variability in outcome definitions, and differences across healthcare settings limit direct clinical applicability. These findings support cautious prospective validation and clinical utility assessment before consideration of broader real-world implementation.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":" ","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13662964/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148535322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical decision support system for child and adolescent mental health services: a formative usability study.","authors":"Carolyn E Clausen, Dipendra Pant, Roman Koposov, Øystein Nytrø, Bennett Leventhal, Stian Lydersen, Odd Sverre Westbye, Thomas Brox Røst, Norbert Skokauskas","doi":"10.1186/s12911-026-03720-w","DOIUrl":"10.1186/s12911-026-03720-w","url":null,"abstract":"<p><strong>Background: </strong>Child and adolescent mental disorders are creating significant demands on Norway's general and specialized health services. Addressing these demands requires innovative approaches to increase workforce numbers and effectiveness. The Individualized Digital Decision Assist System (IDDEAS) is a Clinical Decision Support System (CDSS) for child and adolescent mental health services in Norway. IDDEAS 1.0 is a guidelines-based prototype with attention-deficit/hyperactivity disorder (ADHD) as the first model clinical paradigm. IDDEAS 1.0 is the focus of a formative usability study to examine clinicians' information processing needs and their perceptions of the acceptability, functionality and ease of interactions with user interface (UI) and user-experience (UX).</p><p><strong>Methods: </strong>In collaboration with the Norwegian Association for Child and Adolescent Mental Health (N-BUP) a provisional list of practitioners in child and adolescent mental health services (CAMHS) was compiled for recruitment of potential participants. Child and adolescent psychologists and psychiatrists (n = 68) completed IDDEAS formative usability testing. Study participants also completed a think-aloud protocol in which they verbalized their thoughts about the experience in real time while assessing hypothetical patient case vignettes. By observing these representative, potential IDDEAS users, developers gain insight from their interactions with the prototype. A self-developed display and functional ease-of-use questionnaire, as well as the IDDEAS modified system usability scale and user engagement scales, were completed. All qualitative data were analyzed using both protocol analysis and content analysis.</p><p><strong>Results: </strong>Examination of the formative usability identified a participant appreciation of the side-by-side guideline/patient view, simplistic UI presentation, and alignment with their diagnostic routines. 89% of usability comments informed next-step improvements pinpointing forced linear navigation, ADHD focused prototype, and ambiguous prompts. This resulted in proposed enhancements centered on dynamic criteria selection, optional guideline choice, individualized content, and seamless electronic health records integration.</p><p><strong>Conclusion: </strong>The scenario-based exploration of the IDDEAS 1.0 prototype allowed CAMHS clinicians to offer honest reflections about receiving decision-support, and what they might potentially need for such support to enhance their clinical decision-making and information processing at the point of care. Functional adjustments in IDDEAS were recommended, with a focus on workflow cohesion and individualized adaptability for optimal ease of use and personalized patient care.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":" ","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13670641/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148535375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}