Search

This paper presents the implementation of two automated text classification systems for prostate cancer findings based on the PI-RADS criteria. Specifically, a traditional machine learning model using XGBoost and a language model-based approach using RoBERTa were employed. The study focused on Spanish-language radiological MRI prostate reports, which has not been explored before. The results demonstrate that the RoBERTa model outperforms the XGBoost model, although both achieve promising results. Furthermore, the best-performing system was integrated into the radiological company's information systems as an API, operating in a real-world environment., (© 2024. The Author(s).)

Background and Objective: Colorectal cancer is one of the major causes of cancer death worldwide. Essential for prognosis and treatment planning, TNM staging offers critical insights into the advancement of colorectal cancer. However, manual TNM staging from colon magnetic resonance imaging (MRI) is a laborious and error prone process. This study introduces an automated text classification system for TNM staging of colon MRI images in Spanish., Methods: A dataset of 1319 Spanish colon MRI reports was collected and manually labeled with TNM staging. In order to automate the task of TNM staging, a multimodal system was proposed. The system is based on RoBERTa language model pre-trained on a combination of biomedical and clinical Spanish language corpora and uses Natural Language Processing (NLP) techniques to extract relevant categorical and numerical features from MRI reports., Results: The performance of the system was evaluated using different metrics and the results obtained are very promising: the best performance among the proposed systems reached 0.7464, 0.8792 and 0.6776 of macro F1-score for T, N and M respectively., Conclusions: This study demonstrates the feasibility of using a language model for automatic TNM staging based on Spanish clinical reports of colorectal cancer patients. The proposed system can be a useful tool to improve the efficiency and accuracy of colorectal cancer diagnosis., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

This paper presents a new corpus of radiology medical reports written in Spanish and labeled with ICD-10. CARES (Corpus of Anonymised Radiological Evidences in Spanish) is a high-quality corpus manually labeled and reviewed by radiologists that is freely available for the research community on HuggingFace. These types of resources are essential for developing automatic text classification tools as they are necessary for training and tuning computational systems. However, in the medical domain these are very difficult to obtain for different reasons including privacy and data protection issues or the involvement of medical specialists in the generation of these resources. We present a corpus labeled and reviewed by radiologists in their daily practice that is available for research purposes. In addition, after describing the corpus and explaining how it has been generated, a first experimental approach is carried out using several machine learning algorithms based on transformer language models such as BioBERT and RoBERTa to test the validity of this linguistic resource. The best performing classifier achieved 0.8676 micro and 0.8328 macro f1-score and these results encourage us to continue working in this research line., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Background: Natural language processing (NLP) and text mining technologies for the extraction and indexing of chemical and drug entities are key to improving the access and integration of information from unstructured data such as biomedical literature., Methods: In this paper we evaluate two important tasks in NLP: the named entity recognition (NER) and Entity indexing using the SNOMED-CT terminology. For this purpose, we propose a combination of word embeddings in order to improve the results obtained in the PharmaCoNER challenge., Results: For the NER task we present a neural network composed of BiLSTM with a CRF sequential layer where different word embeddings are combined as an input to the architecture. A hybrid method combining supervised and unsupervised models is used for the concept indexing task. In the supervised model, we use the training set to find previously trained concepts, and the unsupervised model is based on a 6-step architecture. This architecture uses a dictionary of synonyms and the Levenshtein distance to assign the correct SNOMED-CT code., Conclusion: On the one hand, the combination of word embeddings helps to improve the recognition of chemicals and drugs in the biomedical literature. We achieved results of 91.41% for precision, 90.14% for recall, and 90.77% for F1-score using micro-averaging. On the other hand, our indexing system achieves a 92.67% F1-score, 92.44% for recall, and 92.91% for precision. With these results in a final ranking, we would be in the first position., (© 2021. The Author(s).)

Background: Controlled vocabularies are fundamental resources for information extraction from clinical texts using natural language processing (NLP). Standard language resources available in the healthcare domain such as the UMLS metathesaurus or SNOMED CT are widely used for this purpose, but with limitations such as lexical ambiguity of clinical terms. However, most of them are unambiguous within text limited to a given clinical specialty. This is one rationale besides others to classify clinical text by the clinical specialty to which they belong., Results: This paper addresses this limitation by proposing and applying a method that automatically extracts Spanish medical terms classified and weighted per sub-domain, using Spanish MEDLINE titles and abstracts as input. The hypothesis is biomedical NLP tasks benefit from collections of domain terms that are specific to clinical subdomains. We use PubMed queries that generate sub-domain specific corpora from Spanish titles and abstracts, from which token n-grams are collected and metrics of relevance, discriminatory power, and broadness per sub-domain are computed. The generated term set, called Spanish core vocabulary about clinical specialties (SCOVACLIS), was made available to the scientific community and used in a text classification problem obtaining improvements of 6 percentage points in the F-measure compared to the baseline using Multilayer Perceptron, thus demonstrating the hypothesis that a specialized term set improves NLP tasks., Conclusion: The creation and validation of SCOVACLIS support the hypothesis that specific term sets reduce the level of ambiguity when compared to a specialty-independent and broad-scope vocabulary.

Background and Objective: Assignment of medical imaging procedure protocols requires extensive knowledge about patient's data, usually included in radiological request forms and radiological reports. Assignment of protocol is required prior to radiological study acquisition, determining procedure for each patient. The automation of this protocol assignment process could improve the efficiency of patient's diagnosis. Artificial intelligence has proven to be of great help in these healthcare-related problems, and specifically the application of Natural Language Processing (NLP) techniques for extracting information from text reports has been successfully used in automatic text classification tasks., Methods: In this paper, machine learning classification models based on NLP have been developed using patient's data present in radiological reports and radiological imaging protocols. We have used a real corpus provided by the private medical center "HT medica" composed of almost 700,000 Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) examinations obtained during routine clinical use. We have compared several models including traditional machine learning methods such as support vector machine and random forest, neural networks and transfer language techniques., Results: The results obtained are encouraging taking into account that the system is performing a complex text multiclass classification task. Specifically, for the best proposed system we obtain 92.2% accuracy in the CT dataset and 86.9% in the MRI dataset., Conclusions: The best machine learning system is potentially efficient, quality and cost effective. For this reason it is currently used in real scenarios by radiologists as decision support tool for assigning protocols of CT and MRI studies., (Copyright © 2021. Published by Elsevier B.V.)

COVID-19 diagnosis is usually based on PCR test using radiological images, mainly chest Computed Tomography (CT) for the assessment of lung involvement by COVID-19. However, textual radiological reports also contain relevant information for determining the likelihood of presenting radiological signs of COVID-19 involving lungs. The development of COVID-19 automatic detection systems based on Natural Language Processing (NLP) techniques could provide a great help in supporting clinicians and detecting COVID-19 related disorders within radiological reports. In this paper we propose a text classification system based on the integration of different information sources. The system can be used to automatically predict whether or not a patient has radiological findings consistent with COVID-19 on the basis of radiological reports of chest CT. To carry out our experiments we use 295 radiological reports from chest CT studies provided by the ''HT médica" clinic. All of them are radiological requests with suspicions of chest involvement by COVID-19. In order to train our text classification system we apply Machine Learning approaches and Named Entity Recognition. The system takes two sources of information as input: the text of the radiological report and COVID-19 related disorders extracted from SNOMED-CT. The best system is trained using SVM and the baseline results achieve 85% accuracy predicting lung involvement by COVID-19, which already offers competitive values that are difficult to overcome. Moreover, we apply mutual information in order to integrate the best quality information extracted from SNOMED-CT. In this way, we achieve around 90% accuracy improving the baseline results by 5 points., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Although the Enders strain of mumps virus (MuV) encodes a functional V protein that acts as an interferon (IFN) antagonist, in multi-cycle growth assays MuV Enders grew poorly in naïve ('IFN-competent' Hep2) cells but grew to high titres in 'IFN-compromised' Hep2 cells. Even so, the growth rate of MuV Enders was significantly slower in 'IFN-compromised' Hep2 cells when compared with its replication rate in Vero cells and with the replication rate of parainfluenza virus type 5 (a closely related paramyxovirus) in both naïve and 'IFN-compromised' Hep2 cells. This suggests that a consequence of slower growth is that the IFN system of naïve Hep2 cells can respond quickly enough to control the growth of MuV Enders. This is supported by the finding that rapidly growing variants of MuV Enders that were selected on 'IFN-compromised' Hep2 cells (i.e. in the absence of any selection pressure exerted by the IFN response) also grew to high titres on naïve Hep2 cells. Sequencing of the complete genome of one of these variants identified a single point mutation that resulted in a substitution of a conserved asparagine by histidine at position 498 of the haemagglutinin-neuraminidase protein, although this mutation was not present in all rapidly growing variants. These results support the concept that there is a race between the ability of a cell to detect and respond to virus infection and the ability of a virus to block the IFN response. Importantly, this emphasizes that factors other than viral IFN antagonists influence the sensitivity of viruses to IFN.

Searching biomedical information in a large collection of medical data is a complex task. The use of tools and biomedical resources could ease the retrieval of the information desired. In this paper, we use the medical ontology MeSH to improve a Multimodal Information Retrieval System by expanding the user's query with medical terms. In order to accomplish our experiments, we have used the dataset provided by ImageCLEFmed task organizers for years 2005 and 2006. This dataset is composed of a multimodal collection (images and text) of clinical cases, a list of queries for each year, and a list of relevance judgments for each query to evaluate the results. The results from the experiments show that the use of a medical ontology to expand the queries greatly improves the results.

The intragroup antigenic diversity of the G glycoprotein of 226 human respiratory syncytial virus (HRSV) strains isolated in Buenos Aires (Argentina) and Santiago (Chile) between 1995 and 2002 was evaluated by ELISA with a panel of 14 anti-G monoclonal antibodies (MAbs). Out of 226 strains characterized, 172 (76%) belonged to group A and 54 (24%) to group B. Strains from both groups cocirculated throughout the study period in both countries, except in 1996, 2000, and 2002 when only group A strains were isolated. Within group A 23 different antigenic patterns were found as defined by the combination of reactivities with eight strain-specific anti-G MAbs. These antigenic patterns showed different behavior regarding their circulation. Some major patterns were observed in most years with variable proportions; other minor patterns were present in low proportions during 1 or 2 years and then were apparently replaced by new patterns. Some antigenic patterns occurred both in Argentina and Chile during the same epidemics. Since no strain-specific MAbs were available for group B, we could not evidence the antigenic variability within group B. These are the first data on antigenic characterization of HRSV strains isolated in Argentina and Chile. It is shown that the ELISA with MAbs directed against the G protein of RSV is a valuable tool. These results will also provide useful information for further studies to evaluate the antigenic variability of HRSV strains in relation with genetic characteristics.

The genetic and antigenic variability of human respiratory syncytial virus (HRSV) strains isolated in Buenos Aires from 1995 to 2001 was evaluated by partial nucleotide sequencing of the G gene and enzyme-linked immunosorbent assay analysis with anti-G monoclonal antibodies. Phylogenetic analyses showed that 37 group A strains clustered into five genotypes, whereas 20 group B strains clustered into three genotypes. Group A showed more genetic variability than group B. A close correlation between genotypes and antigenic patterns was observed. Changes detected in the G protein of viruses from both groups included (i) amino acid substitutions and(ii) differences in protein length due to either changes in stop codon usage or sequence duplications. Three B strains from 1999 exhibited a duplication of 20 amino acids, while one B strain from 2001 had 2 amino acids duplicated. The comparison among Argentinean HRSV strains and viruses isolated in other geographical areas during different epidemics is discussed.

In immunocompromised patients, diagnosis of Cytomegalovirus (CMV) active infection is of utmost importance for the initiation, monitoring and ending of antiviral therapy. Therefore, the presence of viral replication should be demonstrated. Isolation in tissue culture is one of the standard methods. The objective of the present paper was to compare two isolation procedures for CMV: conventional cell culture (CC) and rapid shell vial (SV) assay in human fibroblasts. A total of 584 clinical samples were studied between 1991 and 1998. CMV was isolated in 14.4% of the samples, 11.8% of which were positive by SV and 7.7% by CC. Out of 84 positive samples, concordance between both methods was observed in 36% of the cases. We found that 46% of the samples were positive only by SV, while 18% were positive only by CC. The average time required for obtaining the results by CC was 22.6 +/- 2.3 days. Out of the 69 samples positive by SV, 43% were already positive after 24 hours and the rest after 48 hours. These results indicate that SV was more sensitive and rapid than CC. The main advantage of CC, despite its time-consuming process, is the ability to recover the viral strain for both antiviral susceptibility phenotypical tests and strain characterization. Furthermore, in this study, absence of CC would have resulted in the loss of 18% of the positive diagnoses. In conclusion, simultaneous use of both methods is suggested in order to obtain a rapid result and the highest sensitivity.

The objective of this study was to search for regional cerebral blood flow (rCBF) abnormalities in adolescents with initial-stage schizophrenia by means of brain single-photon emission tomography (SPET) using technetium-99m hexamethylpropylene amine oxime (HMPAO). SPET studies were performed on a homogeneous sample of 15 carefully selected adolescents with a recent diagnosis of schizophrenia, and without previous electroconvulsive or antipsychotic drug treatment. Computed tomography (CT) and electro-encephalographic (EEG) studies were performed in all patients. Qualitative and semiquantitative analysis of 99mTc-HMPAO SPET studies showed an impaired rCBF in 12 patients (80%). The most common pattern was a decreased uptake of 99mTc-HMPAO in the frontal lobes, usually in the left hemisphere. Conventional and quantitative EEG was positive in 12 (80%) and 15 (100%) patients, respectively. CT findings were positive in two patients (13%). There was a high level of concordance between SPET and EEG results and between SPET and clinical features (P>0.05). This study suggests that previously untreated patients in the first stages of schizophrenia present functional abnormalities that are revealed by brain SPET.