A5024070959- Biomedical Text Mining and Ontologies
- Topic Modeling
- Cardiac Arrhythmias and Treatments
- Cardiac pacing and defibrillation studies
- Natural Language Processing Techniques
- Cardiac electrophysiology and arrhythmias
- Cardiomyopathy and Myosin Studies
- Cardiovascular Effects of Exercise
- Atrial Fibrillation Management and Outcomes
- Cardiovascular Function and Risk Factors
- Electronic Health Records Systems
- Radiology practices and education
- Radiomics and Machine Learning in Medical Imaging
- Cardiac Imaging and Diagnostics
- Clinical practice guidelines implementation
- Machine Learning in Healthcare
- Liver Disease Diagnosis and Treatment
- Cardiac Structural Anomalies and Repair
- Viral Infections and Immunology Research
- Cardiovascular Disease and Adiposity
- Nursing Diagnosis and Documentation
- Surgical Simulation and Training
- Pharmacy and Medical Practices
- ECG Monitoring and Analysis
- Gastrointestinal motility and disorders
Osaka University
2015-2025
Osaka Police Hospital
2011
Kyushu University
2006
The University of Tokyo
2006
Pretraining large-scale neural language models on raw texts has made a significant contribution to improving transfer learning in natural processing. With the introduction of transformer-based models, such as bidirectional encoder representations from transformers (BERT), performance information extraction free text improved significantly both general and medical domains. However, it is difficult train specific BERT perform well domains for which few databases high quality large size are...
Research suggests that heart failure with reduced ejection fraction (HFrEF) is a state of systemic inflammation may be triggered by microbial products passing into the bloodstream through compromised intestinal barrier. However, whether microbiota exhibits dysbiosis in HFrEF patients largely unknown.Methods and Results:Twenty eight non-ischemic 19 healthy controls were assessed 16S rRNA analysis bacterial DNA extracted from stool samples. After processing sequencing data, bacteria...
Extracting clinical terms from free-text format radiology reports is a first important step toward their secondary use. However, there no general consensus on the kind of to be extracted. In this paper, we propose an information model comprising three types entities: observations, findings, and modifiers. Furthermore, determine its applicability for in-house reports, extracted with state-of-the-art deep learning models compared results. We trained evaluated using 540 chest computed...
Abstract Missed critical imaging findings, particularly those indicating cancer, are a common issue that can result in delays patient follow-up and treatment. To address this, we developed rule-based natural language processing (NLP) algorithm to detect cancer-suspicious findings from Japanese radiology reports. The dataset used consisted of chest abdomen CT reports six institutions. Reports our institution were for development internal evaluation, while the other five institutions external...
Abstract Aims Electron microscopy reveals microstructural alterations in cardiomyocyte nuclei and myofilaments non‐ischaemic cardiomyopathy (NICM), particularly dilated (DCM). Nevertheless, the correlation between such observations clinical outcomes, including prognosis left ventricular reverse remodelling (LVRR), remains unclear. This study aimed to examine association electron microscopic findings outcomes patients with NICM. Methods In this multicentre, prospective, observational study,...
A radiology report is prepared for communicating clinical information about observed abnormal structures and clinically important findings with referring clinicians. However, such observations are often accompanied by ambiguous expressions, which can prevent clinicians from accurately interpreting the content of reports. To systematically assess degree diagnostic certainty each observation finding in a report, we defined an ordinal scale comprising five classes: definite, likely, may...
Radiology reports are usually written in a free-text format, which makes it challenging to reuse the reports.For secondary use, we developed 2-stage deep learning system for extracting clinical information and converting into structured format.Our mainly consists of 2 modules: entity extraction relation extraction. For each module, state-of-the-art models were applied. We trained evaluated using 1040 in-house Japanese computed tomography (CT) annotated by medical experts. also performance...
Pre-training large-scale neural language models on raw texts has made a significant contribution to improving transfer learning in natural processing (NLP). With the introduction of transformer-based models, such as bidirectional encoder representations from transformers (BERT), performance information extraction free text by NLP significantly improved for both general domain and medical domain; however, it is difficult train specific BERT that perform well domains which there are few...
Abstract Background : Pre-training large-scale neural language models on raw texts has been shown to make a significant contribution strategy for transfer learning in natural processing (NLP). With the introduction of transformer-based models, such as Bidirectional Encoder Representations from Transformers (BERT), performance information extraction free text by NLP significantly improved both general domain and medical domain; however, it is difficult languages which there are few publicly...
Documentation tasks comprise a large percentage of nurses' workloads. Nursing records were partially based on report from the patient. However, it is not verbatim transcription patient's complaints but type medical record. Therefore, to reduce time spent nursing documentation, necessary assist in appropriate conversion or citation patient reports professional records. few studies have been conducted systems for capturing electronic In addition, there no whether such system reduces...
Background: Left ventricular reverse remodeling (LVRR) is a favorable response in non-ischemic, non-valvular cardiomyopathy (NICM) patients. Recently, 18-lead body surface electrocardiography (ECG), the standard 12-lead ECG with synthesized right-sided/posterior chest leads, has been developed, but its predictive value for LVRR not evaluated.
In this study, we tried to create a machine-learning method that detects disease lesions from chest X-ray (CXR) images using data set annotated with extracted CXR reports information. We the nodule as target lesion. Manually annotating nodules is costly in terms of time. Therefore, used report information automatically produce training for object detection task. First, use semantic segmentation model PSP-Net recognize lung fields described reports. Next, classification ResNeSt-50...
Early repolarization syndrome (ERS) and Brugada (BrS) share many electrocardiographic clinical features, recently have been collectively grouped as J wave syndrome. However, the effects of sodium channel blockers on waves differ greatly between ERS BrS.
We implemented a multilingual medical questionnaire system, which allows patients to answer questionnaires both in and out of the hospital. The response data are sent stored as structured on server hospital information could be converted Japanese quoted part progress notes electronic record.
Some multicenter clinical studies require the acquisition of specimens from patients, and centralized management analysis at a research institution. In such cases, it is necessary to manage with anonymized patient information. addition, need be managed in connection information studies. this study, we have developed specimen system that works electronic data capture for efficient workflow has verified Osaka University Hospital. by combining medical image collection previously, integrated...
Abstract Background Dilated cardiomyopathy (DCM) presents with diverse clinical courses, hardly predictable solely by the left ventricular (LV) ejection fraction (EF). Longitudinal strain (LS) offers distinct information from LVEF and exhibits various distribution patterns. This study aimed to evaluate significance of LS patterns in DCM. Methods We studied 139 patients DCM (LVEF ≤ 35%) who were admitted for heart failure (HF). was assessed using a bull’s eye map relative apical index...
Radiology reports are an essential communication method for ensuring smooth workflow in healthcare. However, many of these described free text, and findings documented by radiologists may not be adequately addressed. In this study, focusing on pulmonary nodules, we evaluated whether cases which follow-up as recommended were receiving appropriate treatment. Reports recommending nodules automatically extracted using natural language processing. our evaluation, out 10,507 reports, 1,501 (14.3%)...
The acquisition of medical images from multiple medial institutions has become important for high-quality clinical studies. In recent years, electronic data submission enabled the transmission image to independent more quickly and easily than before. However, selection, anonymization, still require human resources in form research collaborators. this study, we developed an collection system that works with capture (EDC) system. system, are selected based on EDC input information, patient ID...