OBJECTIVE: To develop a user-friendly visualization application for the automatic annotation of Human Phenotype Ontology (HPO) terms based on large language models and retrieval-augmented generation (RAG) technology, and to validate its performance in an authoritative case dataset. METHODS: By integrating the domestic open-source large language model DeepSeek-V3 with RAG technology, an interactive web application was deployed on the Streamlit cloud platform. Using only the latest official HPO dataset as the data source, the lightweight sentence-embedding model BAAI/bge-small-en-v1.5 was employed to construct a FAISS vector index. During the online phase, a four-step closed-loop process is automatically completed: multilingual translation, phenotype phrase extraction, RAG candidate retrieval, term mapping, and official database validation. 121 English case reports publicly released by BMJ Case Reports and Oxford Medical Case Reports (with a gold-standard HPO set of 1 794 terms) were selected for application validation. Precision, recall, and F1 score were calculated and compared horizontally with traditional dictionary tools, standalone large language models, and the similar application "RAG-HPO". Finally, replace the model with the more advanced ChatGPT-5 and evaluate its performance on the newly extracted dataset. RESULTS: An HPO term automatic annotation visualization application named PhenoRAG, based on large language models and RAG technology, was successfully developed. Users can access it directly via a web link. Across the 112 cases, a total of 2 150 HPO terms were generated; 2,064 (96.0%) were fully validated by the official database, with a hallucination rate of 1.3% and an HPO ID-name mismatch rate of 2.7%. After deduplication, 1,906 terms remained for testing. The overall precision was 63.65%, recall was 67.34%, and F1 was 65.44%, significantly outperforming traditional annotation tools (F1: 0.45-0.49, P < 0.001). Although PhenoRAG's F1 was lower than that of RAG-HPO (F1 = 0.78, P < 0.001), which relies on a manually constructed synonym database of 54 000 entries plus the HPO dataset, it requires no additional dictionary maintenance and can be used without any background in computer programming. Moreover, after switching to the GPT-5 model, PhenoRAG exhibited no hallucination rate on the new dataset, and its F1 score significantly increased (P = 0.038). CONCLUSION Without constructing a synonym database, the PhenoRAG achieved high-accuracy automatic mapping from clinical text to standard HPO terms. It features a low usage threshold, free access, and a Chinese-language interface, and can directly serve rare disease diagnosis, genetic counseling, and research scenarios in China and worldwide, warranting further clinical promotion and multicenter validation.
Humans ; Phenotype ; Biological Ontologies ; Language ; Software ; Large Language Models

OBJECTIVE: To explore the efficacy of a Thalassemia risk assessment software for the screening of thalassemia mutation carriers and distribution of thalassemia genotypes detected by screening. METHODS: A total of 6 040 individuals were evaluated at Leshan Maternal and Child Health Care Hospital between 2022 and 2024 using the commonly used clinical thalassemia risk assessment method and the thalassemia screening software, respectively, and the performance indicators of the two methods were compared and analyzed against the result of thalassemia gene testing. This study was approved by the Ethics Committee of our hospital (Ethics No.: LfyLL[2022]005). RESULTS: The high-risk rate by the thalassemia screening software was 11.19%, with a sensitivity of 95.12%, specificity of 93.28%, positive predictive value of 43.20%, negative predictive value of 99.72%, and the area under the ROC curve (AUC) was 0.942. The thalassemia gene detection rate of the high-risk samples screened was 4.83%. The high-risk screening rate of the conventional method was 2.50%, with a sensitivity of 51.22%, specificity of 93.28%, positive predictive value of 80.79%, negative predictive value of 97.40%, and the AUC was 0.754. The thalassemia gene detection rate of the high-risk samples was 2.02%. CONCLUSION The software can effectively detect thalassemia carriers and significantly reduce the missed detection compared with conventional method, thereby significantly improve the efficacy of screening.
Humans ; Thalassemia/diagnosis* ; Software ; Female ; Genetic Testing/methods* ; Male ; Mutation ; Adult ; Genotype ; ROC Curve ; Risk Assessment

Computational modeling is an invaluable tool for mechanism analysis, directed engineering, and rational design of biological parts, metabolic networks, and even cellular systems. It can provide new technological solutions to address biological challenges at different levels and has become a central focus of research in biomanufacturing. In the computational modeling of proteins, which are the key parts in biological systems, the traditional physics-based methods (computer software and mathematical model) have been widely used to study the physical and chemical processes in the functioning of proteins, and have thus been recognized as a powerful tool for understanding complex biological systems and guiding experimental designs. As the scale of computational modeling continues to expand, traditional modeling techniques face difficulties in balancing computational accuracy and speed. In recent years, the explosive growth of biological data has made it possible to construct high-performance artificial intelligence (AI) models, which brings new opportunities to the computational modeling of proteins, and the AI-enhanced physics-based computational modeling technologies have emerged. This combined strategy not only incorporates the chemical knowledge and established physical principles but also is powerful in data processing and pattern recognition, which greatly improves the computational efficiency and prediction accuracy, as well as possesses stronger interpretation ability, transferability, and robustness. The AI-enhanced physics-based computational modeling technologies have already shown great potential and value in biocatalysis, paving a new way for the future development of biomanufacturing.
Artificial Intelligence ; Proteins/chemistry* ; Computer Simulation ; Software ; Computational Biology/methods*

BACKGROUND: Long screen time hours may be associated with behavioral problems in children. To better understand the relationship between children's behavioral problems and screen time, it the associated risk factors must be subdivided based on the purpose underlying screen use. This study examined the relationship between screen time based on intended usage and behavioral problems in Japan. METHODS: This study included 3,332 children aged between 7-17 years from the Hokkaido Study on Environment and Children's Health. From October 2020 to October 2021, the children and their parents answered questionnaires on the children's screen use duration (never used, <30 min, ≥30 min & <1 hour, ≥1 h & <2 h, ≥2 h) based on seven intended usage categories: watching television/video, video gaming, reading books/comics, sending/receiving e-mail/messages, browsing/posting on social networking services, studying for classes/homework, drawing/editing pictures/photos/videos, along with the Strengths and Difficulties Questionnaire (SDQ). Logistic regression was used to analyze the association between screen time, purpose of children's screen use, and behavioral problems across the 13 SDQ total scores. RESULTS: The mean ± standard deviation age of the participants was 12.4 ± 2.4-years-old, 487 (14.6%) children were determined to have behavioral problems, and the duration of screen time increased with their age. The children's primary purposes for screen use were watching television/video, video gaming, sending/receiving e-mail/messages, and browsing/posting on social networking services. Children who reported playing video games for ≥2 hours on weekdays had higher odds of problematic total difficulties scores than never user (Odds Ratio: 2.10, 95% confidence interval: 1.45-3.06). CONCLUSION Long video gaming screen time is associated with behavioral issues, hyperactivity/inattention, and prosocial behaviors in children. Conversely, watching television and videos for 30 min-1 h per day, using e-mail or messaging, and using social networking services were significantly association with reduced odds ratio for peer relationship problems as compared to children who never engaged in these activities. Longitudinal follow-up is needed to further examine screen time and problem behaviors.
Humans ; Screen Time ; Child ; Japan/epidemiology* ; Male ; Female ; Adolescent ; Problem Behavior/psychology* ; Surveys and Questionnaires ; Child Behavior ; Television/statistics & numerical data* ; Video Games/statistics & numerical data*

OBJECTIVE: To compare the registration accuracy of cone beam computed tomography (CBCT) images while registering to virtual craniodentofacial patients based on soft tissue and the dentition registration method. METHODS: Virtual dentofacial patients out of 13 selected participants who needed CBCT scanning were established by impression with a registered-block impression (RBI) based on digital dental images, three-dimensional (3D) facial images and maxillofacial CBCT images. CBCT images were processed in the Mimics software program, establishing the craniofacial virtual patients based on CBCT images (CCTs). Registration between virtual patients from RBI and CCT, using the soft tissue in lower half face (STE) and dentition (DTN) as the reference area, respectively, forming two kinds of virtual craniofacial patients based on digital dental images, 3D facial images and skeletal images of CBCT (hiding the soft tissue and dental casts from CBCT). Three-dimensional deviation analysis was performed in the upper half face and lower half face of facial images from CBCT between two kinds of virtual craniodentofacial patients and compared with 3D facial images from RBI and recorded as root mean square error (RMSE). Paired-t test was used to compare the deviations of RMSEs between the upper and lower half of the face and the upper half of the face of facial images from CCT, respectively, between the two kinds of virtual craniodentofacial patients based on STE and DTN methods. RESULTS: Paired-t tests showed that there was no statistically significant difference between the upper and lower half faces of facial images from CCT between STE and DTN (P>0.05), but the deviation of RMSEs of the upper half face of facial images from CCT in STE was smaller than those in DTN [(1.696±0.420) mm vs. (1.752±0.424) mm, P < 0.01]. CONCLUSION The registration accuracy of CBCT registered in virtual craniodentofacial patients using soft tissue as the reference area was higher.
Humans ; Cone-Beam Computed Tomography/methods* ; Imaging, Three-Dimensional/methods* ; Male ; Face/anatomy & histology* ; Female ; Adult ; Image Processing, Computer-Assisted/methods* ; Young Adult ; User-Computer Interface

OBJECTIVE: This study aims to address the configuration and efficiency issues in the use of digital subtraction angiography (DSA) equipment through the practical implementation of a rationalization platform based on the Internet of Things (IoT). METHODS: By employing IoT and data integration technologies, the deep integration of DSA equipment operational data with clinical data was achieved to construct a knowledge base for rational use of DSA equipment. Simultaneously, a knowledge base was developed using software engineering techniques to visually display data analysis results. RESULTS: Through thorough data analysis, an imbalance in DSA usage between the southern and northern hospital campuses was identified. Addressing this issue, optimizations were implemented based on the data analysis results, which ultimately yielded significant effects. These adjustments not only effectively alleviated the pressure on DSA equipment usage in the southern campus, but also increased equipment utilization in the northern district (the average daily working hours have increased from 4.64 h to 7.19 h), shortened patient appointment wait time (the appointment duration in the southern campus decreased by 21.86% year-on-year, while the appointment duration in the northern campus decreased by 20.51% year-on-year). CONCLUSION Through the practical implementation of a DSA rationalization platform based on IoT, this study not only successfully explored methods for rational DSA usage but also provided valuable reference for the rational management of medical equipment.
Internet of Things ; Angiography, Digital Subtraction/instrumentation* ; Humans ; Software

In order to track the running status of the accelerator in real time, discover potential problems in time, reduce the failure rate, and ensure the safety of radiotherapy patients, a linear accelerator quality management monitoring system is designed based on log files. The system adopts B/S architecture, with the server written in Python3.7 language, and is built based on Django2.2.7 framework. The system uses Python3.7 and Pylinac packages to analyze the log files of each plan, obtaining the planned beam quantity, flux gamma pass rate, and position information of multileaf collimator, etc., to realize the quality monitoring of medical linear accelerator, and customize the development of accelerator spare parts and maintenance management modules. According to statistics, after the establishment of the quality management monitoring system, the accelerator has achieved a 16% reduction in failure rate and a 30% reduction in the downtime rate, which ensures its stable operation in clinical settings.
Particle Accelerators ; Quality Control ; Software

From the perspective of performance evaluation, this paper describes briefly the concerns of study on each component module and the entire instrument, clinical items, software, and product testing on droplet digital PCR instruments, including the study methods and quality control requirements. The increase of the research and development efficiency of products and contribute to the promotion of application of digital PCR instruments in clinical laboratories are expected.
Polymerase Chain Reaction/methods* ; Quality Control ; Software ; Humans

OBJECTIVE: Evaluate the clinical application effect of low-field infant MRI. METHODS: Using literature review, expert consultation, and two rounds of Delphi to determine the evaluation index system. Then retrospectively analyze and compare the data of low-field infant MRI and high-field MRI from January 2023 to December 2024. RESULTS: There is a certain gap between low-field infant MRI and high-field MRI in terms of signal-to-noise ratio, image uniformity, software system reliability, scanning time, user interface friendliness and image result consistency. However, there was no difference in terms of spatial resolution and image quality. The noise, hardware system reliability, mean time between failure and the rate of examination completed without sedation are better than that of high-field MRI. CONCLUSION Low-field infant MRI meets needs of clinical diagnostic and has stable performance. It can be used as a routine screening tool for brain diseases near the bed.
Magnetic Resonance Imaging/methods* ; Humans ; Infant ; Retrospective Studies ; Signal-To-Noise Ratio ; Reproducibility of Results ; Brain Diseases/diagnostic imaging* ; Brain/diagnostic imaging* ; Software

BACKGROUND AND OBJECTIVE

First responders must be physically and mentally healthy to ensure effective emergency response. However, literature showed that Filipino first responders continue to have elevated levels of stress and increased risk for post-traumatic stress and other mental health problems months after their deployment. The “Mi Salud” app was created to help Filipino first responders, their team leaders, and their agencies monitor and manage the responders’ real-time stress levels before, during, and after their deployment more effectively.

The “Mi Salud” app was pretested with Filipino first responders (n=30) to establish convergent validity using existing validated scales measuring the same construct. Participants also completed a Likert scale and questionnaire to assess user experience and app recommendations. During the rollout, first responders (n=32) tested the app and completed a survey on user experience and app recommendations. A focus group discussion (n=11; FGD) was conducted to further explore their experiences with the app. Survey data were analyzed using descriptive statistical methods, while FGD data were examined through thematic analysis.

Results from the online survey showed that the app was generally found to be helpful and that the recommendations within the app were useful. The emerging themes from the FGD corroborated many of the themes from the survey, particularly the benefits of using the app and the app’s ease of use. Positive effects were observed both on the responders and on the responders’ team leader and teammates, which further established the value of the “Mi Salud” app.

The findings show that the “Mi Salud” stress check-app may serve as a useful tool for monitoring and managing the stress levels, a critical aspect for Filipino first responders to maintain optimal functioning during deployments and daily activities.