1
|
Tsujimura T, Yamada K, Ida R, Miwa M, Sasaki Y. Contextualized medication event extraction with striding NER and multi-turn QA. J Biomed Inform 2023; 144:104416. [PMID: 37321443 DOI: 10.1016/j.jbi.2023.104416] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/07/2023] [Revised: 05/24/2023] [Accepted: 05/31/2023] [Indexed: 06/17/2023]
Abstract
This paper describes contextualized medication event extraction for automatically identifying medication change events with their contexts from clinical notes. The striding named entity recognition (NER) model extracts medication name spans from an input text sequence using a sliding-window approach. Specifically, the striding NER model separates the input sequence into a set of overlapping subsequences of 512 tokens with 128 tokens of stride, processing each subsequence using a large pre-trained language model and aggregating the outputs from the subsequences. The event and context classification has been done with multi-turn question-answering (QA) and span-based models. The span-based model classifies the span of each medication name using the span representation of the language model. In the QA model, event classification is augmented with questions in classifying the change events of each medication name and the context of the change events, while the model architecture is a classification style that is the same as the span-based model. We evaluated our extraction system on the n2c2 2022 Track 1 dataset, which is annotated for medication extraction (ME), event classification (EC), and context classification (CC) from clinical notes. Our system is a pipeline of the striding NER model for ME and the ensemble of the span-based and QA-based models for EC and CC. Our system achieved a combined F-score of 66.47% for the end-to-end contextualized medication event extraction (Release 1), which is the highest score among the participants of the n2c2 2022 Track 1.
Collapse
Affiliation(s)
- Tomoki Tsujimura
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Aichi, Japan
| | - Koshi Yamada
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Aichi, Japan
| | - Ryuki Ida
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Aichi, Japan
| | - Makoto Miwa
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Aichi, Japan.
| | - Yutaka Sasaki
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, 468-8511, Aichi, Japan
| |
Collapse
|
2
|
Leaman R, Islamaj R, Adams V, Alliheedi MA, Almeida JR, Antunes R, Bevan R, Chang YC, Erdengasileng A, Hodgskiss M, Ida R, Kim H, Li K, Mercer RE, Mertová L, Mobasher G, Shin HC, Sung M, Tsujimura T, Yeh WC, Lu Z. Chemical identification and indexing in full-text articles: an overview of the NLM-Chem track at BioCreative VII. Database (Oxford) 2023; 2023:7071696. [PMID: 36882099 PMCID: PMC9991492 DOI: 10.1093/database/baad005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/19/2022] [Revised: 01/06/2023] [Accepted: 02/15/2023] [Indexed: 03/09/2023]
Abstract
The BioCreative National Library of Medicine (NLM)-Chem track calls for a community effort to fine-tune automated recognition of chemical names in the biomedical literature. Chemicals are one of the most searched biomedical entities in PubMed, and-as highlighted during the coronavirus disease 2019 pandemic-their identification may significantly advance research in multiple biomedical subfields. While previous community challenges focused on identifying chemical names mentioned in titles and abstracts, the full text contains valuable additional detail. We, therefore, organized the BioCreative NLM-Chem track as a community effort to address automated chemical entity recognition in full-text articles. The track consisted of two tasks: (i) chemical identification and (ii) chemical indexing. The chemical identification task required predicting all chemicals mentioned in recently published full-text articles, both span [i.e. named entity recognition (NER)] and normalization (i.e. entity linking), using Medical Subject Headings (MeSH). The chemical indexing task required identifying which chemicals reflect topics for each article and should therefore appear in the listing of MeSH terms for the document in the MEDLINE article indexing. This manuscript summarizes the BioCreative NLM-Chem track and post-challenge experiments. We received a total of 85 submissions from 17 teams worldwide. The highest performance achieved for the chemical identification task was 0.8672 F-score (0.8759 precision and 0.8587 recall) for strict NER performance and 0.8136 F-score (0.8621 precision and 0.7702 recall) for strict normalization performance. The highest performance achieved for the chemical indexing task was 0.6073 F-score (0.7417 precision and 0.5141 recall). This community challenge demonstrated that (i) the current substantial achievements in deep learning technologies can be utilized to improve automated prediction accuracy further and (ii) the chemical indexing task is substantially more challenging. We look forward to further developing biomedical text-mining methods to respond to the rapid growth of biomedical literature. The NLM-Chem track dataset and other challenge materials are publicly available at https://ftp.ncbi.nlm.nih.gov/pub/lu/BC7-NLM-Chem-track/. Database URL https://ftp.ncbi.nlm.nih.gov/pub/lu/BC7-NLM-Chem-track/.
Collapse
Affiliation(s)
| | | | - Virginia Adams
- NVIDIA, 2788 San Tomas Expressway, Santa Clara, CA 95051, USA
| | - Mohammed A Alliheedi
- Department of Computer Science, Al Baha University, 4781 King Fahd Rd, Al Aqiq 65779, Saudi Arabia
| | - João Rafael Almeida
- Department of Electronics, Telecommunications and Informatics (DETI), Institute of Electronics and Informatics Engineering of Aveiro (IEETA), University of Aveiro, Campus Universitário de Santiago, Aveiro 3810-193, Portugal
- Department of Information and Communications Technologies, University of A Coruña, Camiño do Lagar de Castro, A Coruña 15008, Spain
| | - Rui Antunes
- Department of Electronics, Telecommunications and Informatics (DETI), Institute of Electronics and Informatics Engineering of Aveiro (IEETA), University of Aveiro, Campus Universitário de Santiago, Aveiro 3810-193, Portugal
| | - Robert Bevan
- Informatics Department, Medicines Discovery Catapult, Alderley Park, Block 35, Mereside, Macclesfield SK10 4ZF, UK
| | - Yung-Chun Chang
- Graduate Institute of Data Science, Taipei Medical University, No. 172-1, Section 2, Keelung Rd, Da’an District, Taipei City , Taipei 106, Taiwan
| | - Arslan Erdengasileng
- Department of Statistics, Florida State University, 117 N. Woodward Ave, Tallahassee, FL 32306, USA
| | - Matthew Hodgskiss
- Informatics Department, Medicines Discovery Catapult, Alderley Park, Block 35, Mereside, Macclesfield SK10 4ZF, UK
| | - Ryuki Ida
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, Aichi 468-8511, Japan
| | - Hyunjae Kim
- Department of Computer Science and Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, South Korea
| | - Keqiao Li
- Department of Statistics, Florida State University, 117 N. Woodward Ave, Tallahassee, FL 32306, USA
| | - Robert E Mercer
- Department of Computer Science, The University of Western Ontario, Room 355, Middlesex College, Ontario , London N6A 5B7, Canada
| | - Lukrécia Mertová
- Scientific Databases and Visualization Group, Heidelberg Institute for Theoretical Studies (HITS gGmbH), Schloss-Wolfsbrunnenweg 35, Heidelberg 69118, Germany
| | - Ghadeer Mobasher
- Scientific Databases and Visualization Group, Heidelberg Institute for Theoretical Studies (HITS gGmbH), Schloss-Wolfsbrunnenweg 35, Heidelberg 69118, Germany
- Institute of Computer Science, Heidelberg University, Im Neuenheimer Feld 205, Heidelberg 69120, Germany
| | - Hoo-Chang Shin
- NVIDIA, 2788 San Tomas Expressway, Santa Clara, CA 95051, USA
| | - Mujeen Sung
- Department of Computer Science and Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, South Korea
| | - Tomoki Tsujimura
- Computational Intelligence Laboratory, Toyota Technological Institute, 2-12-1 Hisakata, Tempaku-ku, Nagoya, Aichi 468-8511, Japan
| | - Wen-Chao Yeh
- Institute of Information Systems and Applications, National Tsing Hua University, No. 101, Section 2, Kuang-Fu Road, Hsinchu 30013, Taiwan
| | - Zhiyong Lu
- *Corresponding author: Tel: +1-301-594-7089; Fax: +1-301-480-2290;
| |
Collapse
|
3
|
Natsuhori M, Okada M, Ida R, Sasaki K, Shimoda M, Kokue E. Binding characteristics of folate to high affinity folate binding protein purified from porcine serum. J Vet Med Sci 1999; 61:743-8. [PMID: 10458095 DOI: 10.1292/jvms.61.743] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/22/2022] Open
Abstract
High affinity folate binding protein (HFBP) in porcine serum was purified 2,000-fold to a specific activity of 1.4 nmol of tetrahydrofolic acid (THF) bound per mg of protein, using folic acid (FA) coupled EAH-Sepharose gel affinity chromatography. Binding activity of purified HFBP to folate was examined by ultrafiltration method linked to high-performance liquid chromatography with electrochemical detection or to liquid scintillation counting. Binding affinity of HFBP to folate was characterized by dissociation constants (Kd): 13, 17, and 31 pM for tritiated FA (3HFA), THF, and N5-methyltetrahydrofolic acid (5MF), respectively. FA, THF, and 5MF significantly inhibited binding of HFBP to 3HFA, and according to the magnitude of intensity of the binding inhibition, the order of affinity of each folate was confirmed to be FA > THF > 5MF. Binding activity was rather high and stable for THF and 5MF at pH ranging from 6.0 to 10.0. The binding activity, however, rapidly decreased at pH below 6.0 and over 10.0. No binding activity was observed pH below 3.0 and over 12. Gel filtration analysis showed that the prepared HFBP solution had specific binding activity at around 77 kDa of apparent molecular weight, which was 82 kDa by SDS-PAGE. It is considered that this specific and stable binding enables THF to distribute in porcine plasma.
Collapse
Affiliation(s)
- M Natsuhori
- Laboratory of Veterinary Radiology and Radiation Biology, Kitasato University School of Veterinary Medicine and Animal Sciences, Towada, Aomori, Japan
| | | | | | | | | | | |
Collapse
|
4
|
Kapur KK, Garrett NR, Hamada MO, Roumanas ED, Freymiller E, Han T, Diener RM, Levin S, Ida R. A randomized clinical trial comparing the efficacy of mandibular implant-supported overdentures and conventional dentures in diabetic patients. Part I: Methodology and clinical outcomes. J Prosthet Dent 1998; 79:555-69. [PMID: 9597609 DOI: 10.1016/s0022-3913(98)70177-5] [Citation(s) in RCA: 50] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/07/2023]
Abstract
STATEMENT OF PROBLEM Scientific evidence is lacking to support the general application of implant-supported mandibular overdentures. PURPOSE This randomized clinical trial was undertaken to compare the efficacy of conventional mandibular and implant-supported overdentures in diabetic edentulous patients with clinically acceptable metabolic control. METHOD A total of 102 diabetic patients, treated with or without insulin, were randomized to receive a new maxillary denture and either a conventional or an implant-supported removable mandibular overdenture. Treatment was completed for 89 patients, 37 with the conventional and 52 with implant-supported dentures. Detailed examinations, tests, and questionnaires were given before and at 6- and 24-months after treatment completion. Comparisons between the two treatment groups were made for treatment failures based on prespecifed criteria and the type and amount of maintenance care provided. RESULTS The insulin and noninsulin treated groups were collapsed because of the lack of significant differences at entry. The conventional denture and implant-supported overdenture groups were similar in terms of general demographics, medical status, quality of their original dentures and denture support, several functional measures, and patient satisfaction. Treatment was judged to be successful in 56.9% of patients with conventional dentures and 72.1% with overdentures. This difference in success rate was not statistically significant (p > 0.05). Patients with treatment failures in both groups required excessive maintenance care. Those with conventional dentures needed frequent denture base adjustments and relines, whereas those with overdentures required frequent clip replacements and repairs. Although significant improvements were seen with both treatment modalities, a higher percentage of patients with implant-supported overdentures than those with conventional dentures reported improvements in chewing comfort and moderate-to-complete overall satisfaction.
Collapse
Affiliation(s)
- K K Kapur
- University of California, Los Angeles, School of Dentistry, USA
| | | | | | | | | | | | | | | | | |
Collapse
|
5
|
Ida R, Lee A, Huang J, Brandi ML, Yamaguchi DT. Prostaglandin-stimulated second messenger signaling in bone-derived endothelial cells is dependent on confluency in culture. J Cell Physiol 1994; 160:585-95. [PMID: 8077296 DOI: 10.1002/jcp.1041600322] [Citation(s) in RCA: 10] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/28/2023]
Abstract
New bone formation is associated with an increase in blood flow by the invasion of capillaries. Endothelial cells that line the capillaries can produce paracrine factors that affect bone growth and development, and in turn, could be affected by products produced by bone cells, in particular the osteoblasts. Since osteoblasts produce prostaglandins E2 and F2 alpha (PGE2, PGF2 alpha), it was investigated if these PGs were agonists to bone-derived endothelial cells (BBE) by assessing changes in cAMP and free cytosolic calcium concentration ([Ca2+]i) second messenger generation. We found that confluent cultures of BBE cells, a clonal endothelial cell line derived from bovine sternal bone, responded to 1 microM PGE2 by an increase in cAMP. PGF2 alpha at the same concentration was less potent in stimulating an increase in cAMP production in confluent BBE cells. Subconfluent cells with a morphology similar to that of fibroblastic cells were not as sensitive to PGE2-stimulated cAMP generation. PGF2 alpha failed to elicit any cAMP production in subconfluent cultures. PGE2 and PGF2 alpha both stimulated an increase in [Ca2+]i concentration in a dose-dependent manner. The potency of PGE2 was similar to that of PGF2 alpha in stimulating an increase in [Ca2+]i. The Ca2+ response was mostly independent of extracellular Ca+, was unchanged even with prior indomethacin treatment, was unaffected by caffeine pretreatment, but was abolished subsequent to thapsigargin pretreatment. The PG-induced increase in [Ca2+]i was also dependent on the confluency of the cells. In a subconfluent state, the responses to PGE2, or PGF2 alpha were either negligible, or only small increases in [Ca2+]i were noted with high concentrations of these two PGs. Consistent, dose-dependent increases in [Ca2+]i were stimulated by these PGs only when the cells were confluent and had a cobblestoned appearance. Since it was previously demonstrated that BBE cells respond to parathyroid hormone (PTH) by the production of cAMP, we tested if bovine PTH(1-34) amide ]bPTH(1-34) also increased [Ca2+]i in these cells. No change in [Ca2+]i was found in response to bPTH (1-34), although bPTH (1-34) stimulated a nine to tenfold increase in cAMP. We conclude that BBE cells respond to PGE2 and PGF2 alpha but not to bPTH(1-34) by an increase in [Ca2+]i probably secondary to stimulation of phospholipase C and that the cAMP and [Ca2+]i second messenger responses in BBE cells are dependent on the state of confluency of the cells.
Collapse
Affiliation(s)
- R Ida
- Dental Service, VAMC, West Los Angeles, California 90073
| | | | | | | | | |
Collapse
|