It is often discussed that the fourth methodology for science research is "informatics". The first methodology is a theoretic approach, the second one is observation and/or experiment, and the third one is computer simulation. Informatics is a new methodology for data intensive science, which is a new concept based on the fact that most scientific data are digitalized and the amount of data is huge. The facilities to support informatics are cloud systems. Herein we propose a cloud system especially designed for science. The basic concepts, design, resources, implementation, and applications of the NICT science cloud are discussed
10 October 2002 in Lecture Notes in Computer Science
Objectives: We evaluated the feasibility of using deep learning with a convolutional neural network for predicting bone mineral density (BMD) and bone microarchitecture from conventional computed tomography (CT) images acquired by multivendor scanners. Methods: We enrolled 402 patients who underwent noncontrast CT examinations, including L1–L4 vertebrae, and dual-energy x-ray absorptiometry (DXA) examination. Among these, 280 patients (3360 sagittal vertebral images), 70 patients (280 sagittal vertebral images), and 52 patients (208 sagittal vertebral images) were assigned to the training data set for deep learning model development, the validation, and the test data set, respectively. Bone mineral density and the trabecular bone score (TBS), an index of bone microarchitecture, were assessed by DXA. BMDDL and TBSDL were predicted by deep learning with a convolutional neural network (ResNet50). Pearson correlation tests assessed the correlation between BMDDL and BMD, and TBSDL and TBS. The diagnostic performance of BMDDL for osteopenia/osteoporosis and that of TBSDL for bone microarchitecture impairment were evaluated using receiver operating characteristic curve analysis. Results: BMDDL and BMD correlated strongly (r = 0.81, P < 0.01), whereas TBSDL and TBS correlated moderately (r = 0.54, P < 0.01). The sensitivity and specificity of BMDDL for identifying osteopenia or osteoporosis were 93% and 90%, and 100% and 94%, respectively. The sensitivity and specificity of TBSDL for identifying patients with bone microarchitecture impairment were 73% for all values. Conclusions: The BMDDL and TBSDL derived from conventional CT images could identify patients who should undergo DXA, which could be a gatekeeper tool for detecting latent osteoporosis/osteopenia or bone microarchitecture impairment.
31 October 2024 in Healthcare Informatics Research
Objectives: This study evaluated the efficacy of integrating a retrieval-augmented generation (RAG) model and a large language model (LLM) to improve the accuracy of drug name mapping across international vocabularies.Methods: Drug ingredient names were translated into English using the Japanese Accepted Names for Pharmaceuticals. Drug concepts were extracted from the standard vocabulary of OHDSI, and the accuracy of mappings between translated terms and RxNorm was assessed by vector similarity, using the BioBERT-generated embedded vectors as the baseline. Subsequently, we developed LLMs with RAG that distinguished the final candidates from the baseline. We assessed the efficacy of the LLM with RAG in candidate selection by comparing it with conventional methods based on vector similarity.Results: The evaluation metrics demonstrated the superior performance of the combined LLM + RAG over traditional vector similarity methods. Notably, the hit rates of the Mixtral 8x7b and GPT-3.5 models exceeded 90%, significantly outperforming the baseline rate of 64% across stratified groups of PO drugs, injections, and all interventions. Furthermore, the r-precision metric, which measures the alignment between model judgment and human evaluation, revealed a notable improvement in LLM performance, ranging from 41% to 50% compared to the baseline of 23%.Conclusions: Integrating an RAG and an LLM outperformed conventional string comparison and embedding vector similarity techniques, offering a more refined approach to global drug information mapping.
Bar code- or radio frequency identification (RFID)-based medical instrument management systems have gradually been introduced in the field of surgical medicine for the individual management and identification of instruments. We hypothesized that individual management of instruments using RFID tags can provide previously unavailable information, particularly the precise service life of an instrument. Such information can be used to prevent medical accidents caused by surgical instrument failure. This study aimed to predict the precise service life of instruments by analyzing the data available in instrument management systems. We evaluated the repair history of instruments and the usage count until failure and then analyzed the data by the following three methods: the distribution of the instrument usage count was determined, an instrument failure probability model was generated through logistic regression analysis, and survival analysis was performed to predict instrument failure. The usage count followed a normal distribution. Analysis showed that instruments were not used uniformly during surgery. In addition, the Kaplan–Meier curves plotted for five types of instruments showed significant differences in the cumulative survival rate of different instruments. The usage history of instruments obtained with RFID tags or bar codes can be used to predict the probability of instrument failure. This prediction is significant for determining the service life of an instrument. Implementation of the developed model in instrument management systems can help prevent accidents due to instrument failure. Knowledge of the instrument service life will also help in developing a purchase plan for instruments to minimize wastage.