Kazumasa Kishimoto

BACKGROUND Brolucizumab is an effective treatment for neovascular age-related macular degeneration (nAMD). However, its use has been closely associated with the occurrence of intraocular inflammation (IOI), which may lead to severe and potentially irreversible visual impairment. These safety concerns have limited the clinical use of brolucizumab. Therefore, a reliable method for detecting IOI following brolucizumab injection is urgently needed. OBJECTIVE To evaluate the feasibility of quantitatively detecting IOI-related features following intravitreal brolucizumab injection in patients with nAMD using optical coherence tomography (OCT)-based deep learning analysis. METHODS This study retrospectively enrolled treatment-naïve nAMD patients who developed IOI after brolucizumab injection from Kyoto University Hospital and Gunma University Hospital between 2020 and 2025. Kyoto University Hospital provided 11 eyes of 11 patients with IOI, and Gunma University Hospital provided 20 eyes of 20 patients with IOI and 10 eyes of 10 patients without IOI. We developed a novel framework for longitudinal assessment of IOI by explicitly detecting and quantifying subtle IOI-related features at the object level. The detection model was trained on 304 OCT images from 11 patients at Kyoto University Hospital using patient-level ten-fold cross-validation to prevent data leakage. Externally validation was subsequently performed on 30 patients at Gunma University Hospital. To improve micro-target detection, high-resolution layers and a lightweight channel purification module (CPM) were integrated. Two quantitative imaging indicators, the vitreous cell score (VCS) and vitreoretinal interface irregularity score (VRIIS), were automatically derived from the detected features. Their changes from baseline to post-injection were subsequently integrated into a longitudinal assessment, with an increase in either feature considered indicative of increased IOI risk. RESULTS The proposed model reduced the number of parameters from 2.59 million to 0.63 million while improving detection performance, with a 4.61 percentage points increase in precision and a 3.68 percentage points increase in recall. In the external validation dataset, the proposed model correctly identified 28 out of 30 cases, achieving an accuracy of 93.3%, compared with 73.3% (22 of 30 cases) for the reference model. CONCLUSIONS This study presents a framework for the objective detection and quantification of IOI-related features in OCT images and the longitudinal assessment during nAMD treatment. The proposed model achieved improved detection performance with approximately one-quarter of the original number of parameters and demonstrated 93.3% accuracy on external validation using data from a different institution and OCT device, supporting its potential clinical applicability. Misclassified cases involved either preretinal hemorrhage or anterior segment inflammation, indicating the limitation of OCT-based assessment alone. Future work will incorporate multimodal imaging to further improve IOI detection performance in nAMD treatment.

Background: A persistent discrepancy exists between patient-reported information and physician documentation. While conversational agents have been developed to collect medical histories prior to consultations, existing evaluations have largely focused on diagnostic accuracy or user satisfaction rather than on the completeness and clinical relevance of the information collected. There remains a need to assess the extent to which clinically relevant information is captured through chatbot-based interviews, and to understand how model configurations and instructional strategies influence this coverage. Objective: This study aimed to evaluate the extent to which a chatbot can obtain clinically useful patient history information, and to examine how prompt detail and internal reasoning influence information coverage during chatbot-based medical interviews. Methods: We developed a medical history-taking chatbot using the Qwen3-14B-Instruct model and evaluated 4 configurations in a 2×2 factorial design: detailed and thinking mode, detailed and nonthinking mode, minimal and thinking mode, and minimal and nonthinking mode. These configurations were compared against a rule-based system baseline (choice mode) using 66 standardized primary care clinical cases, with simulated patients interacting with the chatbot according to predefined case scripts. Information coverage (%) was assessed using a checklist inspired by Objective Structured Clinical Examination (OSCE) frameworks. Three physicians independently evaluated transcript coverage, with interrater agreement assessed using full agreement rates and Fleiss κ. For the 4 LLM configurations, coverage was analyzed using 2-way repeated-measures ANOVA to examine the effects of prompt detail, internal reasoning, and their interaction. All 5 configurations, including the rule-based baseline, were additionally compared using 1-way repeated-measures ANOVA with post hoc 2-tailed paired t tests. Results: Interrater agreement was substantial (Fleiss κ=0.75). Across all 66 simulated cases, information coverage differed significantly among configurations ( P <.001), with the detailed prompt with thinking (detailed and thinking) mode achieving the highest mean coverage (72.3%, SD 14.3%), compared with moderate coverage in configurations using either thinking or detailed prompts alone (approximately 60%) and lower coverage in minimal nonthinking and rule-based configurations (approximately 51%‐54%). In the factorial analysis of the 4 LLM configurations, both prompt detail and internal reasoning were significantly associated with improved information coverage, with a significant interaction between prompt detail and internal reasoning interaction. Mode-related differences were most pronounced for past medical and family history domains. Symptom-level analyses revealed substantial variability, with higher coverage for symptoms associated with well-defined diagnostic frameworks and lower coverage for multisystem presentations. Conclusions: In this controlled, simulated setting, the detailed prompt with a thinking mode achieved the highest overall checklist-based information coverage. The findings suggest that combining structured clinical prompts with internal reasoning may improve the completeness of chatbot-collected patient histories. Further research is needed to evaluate its impact on clinical documentation, workflow integration, and real-world usefulness.

Background: Nurses are required to perform multiple tasks concurrently, which leads to multitasking situations, and they have to continuously determine which tasks should be prioritized. This is particularly challenging for novice nurses. Although IT-based systems supporting prioritization have begun to emerge, research on the types of information required when prioritization is processed computationally is scant. Despite the clear need for a supportive information system to assist nursing task prioritization, such systems are not yet sufficiently developed. Objective: This study aimed to explore the appropriate granularity and structure of information that should be provided to computational systems to support decision-making based on the influencing factors of nursing task prioritization. Methods: Semistructured interviews were conducted with 10 nurses working in general wards to examine the factors they consider when determining task prioritization during clinical practice. Data were analyzed using an inductive, semantic approach based on a thematic analysis framework. Results: Three themes and nine categories including (1) medical condition assessment factors (signs of acute physiological changes and indicators of clinical status and conditions), (2) patient-related nursing care factors (physical status, psychological condition, and personal characteristics; care needs during hospitalization; and treatment goals and care preferences), and (3) organizational and operational work factors (temporally structured tasks, requiring collaboration partners for task execution, environmental factors affecting task performance, and institutional- and ward-level policies) were identified. Conclusions: Analysis of computational tractability of the identified factors indicated that medical condition assessment factors are relatively quantifiable. In contrast, patient-centered care and organizational and operational work factors rely on contextual and experiential judgment, limiting standardization and formalization. Regarding such ambiguous and context-dependent elements, flexible information-processing approaches, such as large language models, in addition to conventional rule-based methods, may be effective. Furthermore, the appropriate level of information granularity should be determined by the nature of the prioritization outputs required in actual nursing practice rather than the degree of abstraction itself.

BACKGROUND Problem-oriented medical record (POMR), which incorporate the SOAP format, are regarded as the de facto standard for physicians’ clinical notes. However, since 2010, variations of POMR (hereafter, modern POMR) that differ from the original POMR proposed in the 1960s (classical POMR) have been reported. In practice, physicians’ clinical notes often differ from both forms of POMRs, and many notes are copied and pasted. Thus, POMRs have characteristics that hinder their effective adoption in clinical practice. OBJECTIVE This study aimed to investigate the characteristics of POMRs that hinder their adoption in clinical practice. METHODS We adopted a qualitive research method as an approach suitable for characterizing the properties of existing concepts in literature. The literatures of medical records were gathered manually. After selection, the papers were grouped, and their characteristics were analyzed from the perspectives of the background of the proposed documentation methods, the authors’ philosophies, the purpose of documentation, contents, and the advantages and disadvantages of its digitization. RESULTS Ten papers (four classical POMR; six modern POMR) were selected and classified into six types. Five features common to all types were identified: (1) documentation takes priority over the provision of medical care, which should be the physician’s primary responsibility; (2) the definitions of problems addressed are inconsistent; (3) the fundamental premises of the clinical care model are unclear; (4) adherence to an early methodology (before 1935) that records only differences in the patient’s course; and (5) failure to utilize the advantages of digital data, which can be duplicated. This study addressed deficiency (3) by identifying the clinical care model underlying POMRs and explicitly naming it the “Updated Clinical Care Model.” CONCLUSIONS This study identified five shortcomings that hinder physicians from documenting clinical notes according to POMRs. Redesigning electronic health record systems based on the underlying clinical model identified in this study may offer an alternative to copying and pasting documentation and facilitate documentation that is consistent with POMRs.

BACKGROUND A persistent discrepancy exists between patient-reported information and physician documentation. While conversational agents have been developed to collect medical histories prior to consultation, existing evaluations have largely focused on diagnostic accuracy or user satisfaction rather than the completeness and clinical usefulness of the information collected. There remains a need to assess the extent of clinically relevant information captured through chatbot-based interviews and to understand how model configurations and instructional strategies influence this coverage. OBJECTIVE This study aimed to evaluate the extent to which a chatbot can obtain clinically useful patient history information and to examine how prompt detail and internal reasoning influence information coverage during chatbot-based medical interviews. METHODS We developed a medical history-taking chatbot using the Qwen3-14B-Instruct model and evaluated four configurations in a 2×2 factorial design: Detailed/Thinking (DT), Detailed/Non-thinking (DN), Minimal/Thinking (MT), and Minimal/Non-thinking (MN). These configurations were compared against a rule-based system baseline (choice-based mode) using 66 standardized primary care clinical cases, with simulated patients interacting with the chatbot according to predefined case scripts. Information coverage (%) was assessed using a checklist inspired by Objective Structured Clinical Examination (OSCE) frameworks. Three physicians independently evaluated transcript coverage, with inter-rater agreement assessed using full agreement rates and Fleiss’ κ. Coverage percentages were compared across configurations using repeated-measures analysis of variance with post hoc testing. RESULTS Inter-rater agreement was substantial (Fleiss’ κ = 0.75). Across all 66 simulated cases, information coverage differed significantly among configurations (p < .001), with the detailed prompt with thinking (DT) mode achieving the highest mean coverage (72.3%), compared with moderate coverage in configurations using either thinking or detailed prompts alone (approximately 60%) and lower coverage in minimal non-thinking and rule-based configurations (approximately 51-54%). Differences were most pronounced for past medical and family history domains. Symptom-level analyses revealed substantial variability, with higher coverage for symptoms associated with well-defined diagnostic frameworks and lower coverage for multi-system presentations. CONCLUSIONS The combination of clinically detailed prompt instructions and internal reasoning significantly enhances the clinical usefulness of AI-driven history-taking by ensuring more comprehensive data collection. This approach allows for a more systematic and robust foundation for automated clinical documentation, facilitating better integration into healthcare workflows.