A Syntactic Information-Based Classification Model for Medical Literature: Algorithm Development and Validation Study.

BACKGROUND: The ever-increasing volume of medical literature necessitates the classification of medical literature. Medical relation extraction is a typical method of classifying a large volume of medical literature. With the development of arithmetic power, medical relation extraction models have evolved from rule-based models to neural network models. The single neural network model discards the shallow syntactic information while discarding the traditional rules. Therefore, we propose a syntactic information-based classification model that complements and equalizes syntactic information to enhance the model. OBJECTIVE: We aim to complete a syntactic information-based relation extraction model for more efficient medical literature classification. METHODS: We devised 2 methods for enhancing syntactic information in the model. First, we introduced shallow syntactic information into the convolutional neural network to enhance nonlocal syntactic interactions. Second, we devise a cross-domain pruning method to equalize local and nonlocal syntactic interactions. RESULTS: We experimented with 3 data sets related to the classification of medical literature. The F1 values were 65.5% and 91.5% on the BioCreative ViCPR (CPR) and Phenotype-Gene Relationship data sets, respectively, and the accuracy was 88.7% on the PubMed data set. Our model outperforms the current state-of-the-art baseline model in the experiments. CONCLUSIONS: Our model based on syntactic information effectively enhances medical relation extraction. Furthermore, the results of the experiments show that shallow syntactic information helps obtain nonlocal interaction in sentences and effectively reinforces syntactic features. It also provides new ideas for future research directions.

Multi-label classification for biomedical literature: an overview of the BioCreative VII LitCovid Track for COVID-19 literature topic annotations.

The coronavirus disease 2019 (COVID-19) pandemic has been severely impacting global society since December 2019. The related findings such as vaccine and drug development have been reported in biomedical literature-at a rate of about 10 000 articles on COVID-19 per month. Such rapid growth significantly challenges manual curation and interpretation. For instance, LitCovid is a literature database of COVID-19-related articles in PubMed, which has accumulated more than 200 000 articles with millions of accesses each month by users worldwide. One primary curation task is to assign up to eight topics (e.g. Diagnosis and Treatment) to the articles in LitCovid. The annotated topics have been widely used for navigating the COVID literature, rapidly locating articles of interest and other downstream studies. However, annotating the topics has been the bottleneck of manual curation. Despite the continuing advances in biomedical text-mining methods, few have been dedicated to topic annotations in COVID-19 literature. To close the gap, we organized the BioCreative LitCovid track to call for a community effort to tackle automated topic annotation for COVID-19 literature. The BioCreative LitCovid dataset-consisting of over 30 000 articles with manually reviewed topics-was created for training and testing. It is one of the largest multi-label classification datasets in biomedical scientific literature. Nineteen teams worldwide participated and made 80 submissions in total. Most teams used hybrid systems based on transformers. The highest performing submissions achieved 0.8875, 0.9181 and 0.9394 for macro-F1-score, micro-F1-score and instance-based F1-score, respectively. Notably, these scores are substantially higher (e.g. 12%, higher for macro F1-score) than the corresponding scores of the state-of-art multi-label classification method. The level of participation and results demonstrate a successful track and help close the gap between dataset curation and method development. The dataset is publicly available via https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/ for benchmarking and further development. Database URL https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/.

[Hemophagocytic lymphohistiocytosis: tortuous experience in the diagnosis and treatment of one lymphoma].

OBJECTIVE: The etiology of hemophagocytic lymphohistiocytosis (HLH) is complicated and difficult to diagnose, unexplained HLH often with hematological malignancies. Invasive biopsy can help to find etiology, the results may be affected by the technique and the location of the puncture site. Multiangle puncture can improve the success rate, but the corresponding risk increases. A patient with HLH was admitted to Affiliated Hospital of Zunyi Medical College. The etiology was unknown. Active symptomatic support treatment was conducted, at the same time, finding the evidence of viral infection, autoimmune disease related detection, blood culture, bone marrow puncture smear and spleen biopsy were performed respectively to find the pathogen basis. Spleen hemorrhage was not being controlled after spleen biopsy in patients, and emergency splenectomy was adopted to stop bleeding for saving lives. Finally, the patients died of low protein, pulmonary edema and respiratory failure. The bone marrow puncture and spleen biopsy failed to provide the basis for tumor invasion, while the spleen pathological slices plus immunohistochemical indicate diffuse large B cell lymphoma (DLBCL) after splenectomy, which was identified as malignant tumor-associated hemophagocytic syndrome. Underscoring the high risk of bleeding after tumor-associated splenomegaly puncture and the importance of having emergency plans. Through analyzing the clinical characteristics, diagnosis and treatment of this patient, we hope to improve the clinicians' understanding of HLH and lymphoma.

Primary exploration of CDR3 spectratyping and molecular features of TCR ß chain in the peripheral blood and tissue of patients with colorectal carcinoma.

BACKGROUND: The immunoscope spectratyping to TRBV CDR3 had been applied in infectious diseases, tumors, autoimmune diseases and so on, this study aimed to primarily explore CDR3 spectratyping and molecular features of TCR ß chain in the peripheral blood and tissue of patients with colorectal carcinoma. METHODS: Blood and tissue samples were collected from seven patients with colorectal carcinoma (CRC), while blood samples were also collected from two healthy controls as control. Using the real-time fluorescence quantitative reverse transcription polymerase chain reaction (RQ-PCR) and DNA melting curve analysis techniques, the features of T-cell receptor beta chain variable region (TRBV) were determined. RESULTS: The gene melting spectral pattern (GMSP) of 24 TRBV gene families exhibited a highly diverse multimodal shape for most of the TRBV gene families, compared to healthy controls, the more GMSP of patients with CRC showed either a single peak, or several prominent melting peaks (skewed) for certain TRBV gene families. Different patients have different skewed patterns. In the analysis results of sequence, some TRBV CDR3 gene families showed sharing the same motif, such as, TRBV6 of P5, TRBV13.1 of P6 and TRBV21 of P7 (tissue sample) shared the same motif 'GT'; TRBV1 of P1 and TRBV21 of P7 shared the same motif 'AGG', TRBV11 of P1, TRBV21 of P5 and TRBV21 of P7 (tissue sample) shared the same motif 'TDTQY', and even TRBV21 of P5 and P7 (tissue sample) shared the large motif 'SGTDTQY'. As a whole, most of TRBV gene families have the similar motifs 'X-Q', the nucleotide 'X' mainly was 'E', it also was possible be 'T', 'G' or 'K' in some CDR3 gene families of patients with CRC. CONCLUSIONS: There were different GMSPs in different patients with CRC, CDR3 spectratyping and the molecular features of TCR ß chain in the peripheral blood and tissue of patients with CRC were not same or similar, this information would provide ideas for individualized therapy to CRC.