Oğuz Dikenelli

The scarcity of annotated datasets for clinical information extraction in non-English languages hinders the evaluation of large language model (LLM)-based methods developed primarily in English. In this study, we present the first comprehensive bilingual evaluation of LLMs for the clinical Relation Extraction (RE) task in both English and Turkish. To facilitate this evaluation, we introduce the first English-Turkish parallel clinical RE dataset, derived and carefully curated from the 2010 i2b2/VA relation classification corpus. We systematically assess a diverse set of prompting strategies, including multiple in-context learning (ICL) and Chain-of-Thought (CoT) approaches, and compare their performance to fine-tuned baselines such as PURE. Furthermore, we propose Relation-Aware Retrieval (RAR), a novel in-context example selection method based on contrastive learning, that is specifically designed to capture both sentence-level and relation-level semantics. Our results show that prompting-based LLM approaches consistently outperform traditional fine-tuned models. Moreover, evaluations for English performed better than their Turkish counterparts across all evaluated LLMs and prompting techniques. Among ICL methods, RAR achieves the highest performance, with Gemini 1.5 Flash reaching a micro-F1 score of 0.906 in English and 0.888 in Turkish. Performance further improves to 0.918 F1 in English when RAR is combined with a structured reasoning prompt using the DeepSeek-V3 model. These findings highlight the importance of high-quality demonstration retrieval and underscore the potential of advanced retrieval and prompting techniques to bridge resource gaps in clinical natural language processing.

The need to track the origin of shared data/datasets has become apparent, highlighting the necessity of monitoring factors such as trust related to the data/datasets with the widespread use of social media. The concept of Linked Data Space needs to be considered in conjunction with organizations and their provenance with respect to their origin assuming that the shared data is semantic and considering organizations' access to relevant semantic data. In this context, this study elaborates on the concept of Linked Data Space, introducing the terms Internal Data and External Data to the literature. An architecture for Linked Data Space and data management for organizations is defined in addition to these concepts. Furthermore, the study explains how organizations can access External Data in the Linked Data Space and how provenance metadata and ontologies will be created. These developed methods are illustrated in the News Aggregator Scenario, a main scenario for provenance, demonstrating how it can work in a use case.

Pretrained language models augmented with in-domain corpora show impressive results in biomedicine and clinical Natural Language Processing (NLP) tasks in English. However, there has been minimal work in low-resource languages. Although some pioneering works have shown promising results, many scenarios still need to be explored to engineer effective pretrained language models in biomedicine for low-resource settings. This study introduces the BioBERTurk family and four pretrained models in Turkish for biomedicine. To evaluate the models, we also introduced a labeled dataset to classify radiology reports of head CT examinations. Two parts of the reports, impressions and findings, are evaluated separately to observe the performance of models on longer and less informative text. We compared the models with the Turkish BERT (BERTurk) pretrained with general domain text, multilingual BERT (mBERT), and LSTM+attention-based baseline models. The first model initialized from BERTurk and then further pretrained with biomedical corpus performs statistically better than BERTurk, multilingual BERT, and baseline for both datasets. The second model continues to pretrain the BERTurk model by using only radiology Ph.D. theses to test the effect of task-related text. This model slightly outperformed all models on the impression dataset and showed that using only radiology-related data for continual pre-training could be effective. The third model continues to pretrain by adding radiology theses to the biomedical corpus but does not show a statistically meaningful difference for both datasets. The final model combines radiology and biomedicine corpora with the corpus of BERTurk and pretrains a BERT model from scratch. This model is the worst-performing model of the BioBERT family, even worse than BERTurk and multilingual BERT.

In recent years, major advancements in natural language processing (NLP) have been driven by the emergence of large language models (LLMs), which have significantly revolutionized research and development within the field. Building upon this progress, our study delves into the effects of various pre-training methodologies on Turkish clinical language models' performance in a multi-label classification task involving radiology reports, with a focus on addressing the challenges posed by limited language resources. Additionally, we evaluated the simultaneous pretraining approach by utilizing limited clinical task data for the first time. We developed four models, including TurkRadBERT-task v1, TurkRadBERT-task v2, TurkRadBERT-sim v1, and TurkRadBERT-sim v2. Our findings indicate that the general Turkish BERT model (BERTurk) and TurkRadBERT-task v1, both of which utilize knowledge from a substantial general-domain corpus, demonstrate the best overall performance. Although the task-adaptive pre-training approach has the potential to capture domain-specific patterns, it is constrained by the limited task-specific corpus and may be susceptible to overfitting. Furthermore, our results underscore the significance of domain-specific vocabulary during pre-training for enhancing model performance. Ultimately, we observe that the combination of general-domain knowledge and task-specific fine-tuning is essential for achieving optimal performance across a range of categories. This study offers valuable insights for developing effective Turkish clinical language models and can guide future research on pre-training techniques for other low-resource languages within the clinical domain.

Purpose High-risk breast lesions are associated with future risk of breast cancer (1,2). In this study, we aimed to develop a machine learning model to distinguish cases requiring surgical excision from lesions with low risk of accompanying malignancy in high-risk breast lesions. Methods A total of 94 patients who were diagnosed with high-risk lesion by image-guided biopsy between

A large number of data providers publish and connect their structured data on the Web as linked data. Thus, the Web of data becomes a global data space. In this paper, we initially give an overview of query processing approaches used in this interlinked and distributed environment, and then focus on federated query processing on linked data. We provide a detailed and clear insight on data source selection, join methods and query optimization methods of existing query federation engines. Furthermore, we present a qualitative comparison of these engines and give a complementary comparison of the measured metrics of each engine with the idea of pointing out the major strengths of each one. Finally, we discuss the major challenges of federated query processing on linked data.

We present a novel computational model that detects temporal configurations of a given human neuronal pathway and constructs its artificial replication. This poses a great challenge since direct recordings from individual neurons are impossible in the human central nervous system and therefore the underlying neuronal pathway has to be considered as a black box. For tackling this challenge, we used a branch of complex systems modeling called artificial self-organization in which large sets of software entities interacting locally give rise to bottom-up collective behaviors. The result is an emergent model where each software entity represents an integrate-and-fire neuron. We then applied the model to the reflex responses of single motor units obtained from conscious human subjects. Experimental results show that the model recovers functionality of real human neuronal pathways by comparing it to appropriate surrogate data. What makes the model promising is the fact that, to the best of our knowledge, it is the first realistic model to self-wire an artificial neuronal network by efficiently combining neuroscience with artificial self-organization. Although there is no evidence yet of the model’s connectivity mapping onto the human connectivity, we anticipate this model will help neuroscientists to learn much more about human neuronal networks, and could also be used for predicting hypotheses to lead future experiments.