Search

(1) Background: Romania has one of the highest cervical cancer incidence rates in Europe. In Cluj County, the first screening program was initiated in 1998. We aimed to investigate the time trends of cervical cancer incidence in women from Cluj County and to evaluate the data quality at the Cancer Registry. (2) Methods: We calculated time trends of standardized incidence rates in the period 1998–2014 and the Annual Percent Change (APC%). To assess data quality, we used the indicators: mortality/incidence ratio (M/I), percentage of cases declared only at death (DOD%), and percentage of cases with pathological confirmation (PC%). (3) Results: The standardized incidence rate increased steadily, from 23.74 cases/100,000 in 1998, to 32/100,000 in 2014, with an APC% of 2.49% (p &lt, 0.05). The rise in incidence affected both squamous cell carcinoma (APC% 2.49%) (p &lt, 0.05) and cervical adenocarcinoma (APC% 10.54%) (p &lt, 0.05). The M/I ratio was 0.29, DOD% 2.66%, and MC% 94.8%. The last two parameters are within the silver standard concerning data quality. (4) Conclusions. Our study revealed an ascending trend of cervical cancer incidence, more consistent for adenocarcinoma, in the context of a newly introduced screening program and partially due to the improvement of the quality of case reporting at the Cancer Registry from Cluj.

This paper introduces a new Romanian speech corpus from the ROBIN project, called ROBIN Technical Acquisition Speech Corpus (ROBINTASC). Its main purpose was to improve the behaviour of a conversational agent, allowing human-machine interaction in the context of purchasing technical equipment. The paper contains a detailed description of the acquisition process, corpus statistics as well as an evaluation of the corpus influence on a low-latency ASR system as well as a dialogue component., Comment: V. P\u{a}i\c{s}, R. Ion, A. -M. Avram, E. Irimia, V. B. Mititelu and M. Mitrofan, "Human-Machine Interaction Speech Corpus from the ROBIN project", Proceedings of the 2021 International Conference on Speech Technology and Human-Computer Dialogue (SpeD), 2021, pp. 91-96

Within the larger project ROBIN, focused on the development of software and services for human-robot interaction, we present here a set of activities focused on the creation and enhancement of language resources necessary for making dialogue possible between humans and the robot Pepper. More precisely, we describe the preparatory activities for turning the robot Pepper into a dialog partner for a human by using the Romanian language. The language resources that have been created are a lexicon, a language model, and an acoustic language model. They have been enhanced by using other language resources, such as the Romanian wordnet and word embeddings extracted from the large Romanian reference corpus CoRoLa. They will ensure oral communication within some envisaged microworlds, thus they are specific to these microworlds. These resources have been carefully curated and evaluated.

In this paper, we present a web platform for volume rendering of MRI data. Our platform are designed for mobile and desktop devices and VR applications. Unlike a classical desktop application which is platform dependent, we developed a series of software services in order to optimize and deliver fast and efficient volumetric data. The proposed platform is the summary of our initial efforts to implement raytracing algorithms in practical situations. We consider our platform and implementation a viable solution for volumetric rendering of MRI data in web environments.

Extracting parallel units (e.g. sentences or phrases) from comparable corpora in order to enrich existing statistical translation models is an avenue that has attracted a lot of research in recent years. There are experiments that convincingly show how parallel sentences extracted from comparable corpora are able to improve statistical machine translation (SMT). Yet, the existing body of research on the subject does not take into account the degree of comparability of the corpus being processed nor the computation time that it takes to extract translational similar pairs from a corpus of a given size. We will show that the performance of a parallel unit extractor crucially depends on the degree of comparability, such that it is more difficult to mine for parallel data in a weakly comparable corpus than a strongly comparable corpus.

This paper describes the Named Entity Recognition system of the Institute for Artificial Intelligence “Mihai Draganescu” of the Romanian Academy (RACAI for short). Our best F1 score of 0.84984 was achieved using an ensemble of two systems: a gazetteer-based baseline and a RNN-based NER system, developed specially for PharmaCoNER 2019. We will describe the individual systems and the ensemble algorithm, compare the final system to the current state of the art, as well as discuss our results with respect to the quality of the training data and its annotation strategy. The resulting NER system is language independent, provided that language-dependent resources and preprocessing tools exist, such as tokenizers and POS taggers.

The availability of parallel corpora is limited, especially for under-resourced languages and narrow domains. On the other hand, the number of comparable documents in these areas that are freely available on the Web is continuously increasing. Algorithmic approaches to identify these documents from the Web are needed for the purpose of automatically building comparable corpora for these under-resourced languages and domains. How do we identify these comparable documents? What approaches should be used in collecting these comparable documents from different Web sources? In this chapter, we firstly present a review of previous techniques that have been developed for collecting comparable documents from the Web. Then we describe in detail three new techniques to gather comparable documents from three different types of Web sources: Wikipedia, news articles, and narrow domains.

To assess the inter-operator concordance and the potential sources of discordance in defining response to sorafenib in hepatocellular carcinoma (HCC). All patients who received sorafenib between September 2008 and February 2015 were scrutinised for this retrospective study. Images were evaluated separately by three radiologists with different expertise in liver imaging (operator 1, >10 years; operator 2, 5 years; operator 3, no specific training in liver imaging), according to: response evaluation radiological criteria in solid tumours (RECIST) 1.1, modified RECIST (mRECIST) and response evaluation criteria in cancer of the liver (RECICL). The overall response concordance between the more expert operators was good, irrespective of the criteria (RECIST 1.1, ĸ = 0.840; mRECIST, ĸ = 0.871; RECICL, ĸ = 0.819). Concordance between the less expert operator and the other colleagues was lower. The most evident discordance was in target lesion response assessment, with expert operators disagreeing mostly on lesion selection and less expert operators on lesion measurement. As a clinical correlate, overall survival was more tightly related with “progressive disease” as assessed by the expert compared to the same assessment performed by operator 3. Decision on whether a patient is a responder or progressor under sorafenib may vary among different operators, especially in case of a non-specifically trained radiologist. Regardless of the adopted criteria, patients should be evaluated by experienced radiologists to minimise variability in this critical instance. • Inter-operator variability in the assessment of response to sorafenib is poorly known. • The concordance between operators with expertise in liver imaging was good. • Target lesions selection was the main source of discordance between expert operators. • Concordance with non-specifically trained operator was lower, independently from the response criteria. • The non-specifically trained operator was mainly discordant in measurements of target lesions.

This paper presents the adaptation of the Hidden Markov Models-based TTL part-of-speech tagger to the biomedical domain. TTL is a text processing platform that performs sentence splitting, tokenization, POS tagging, chunking and Named Entity Recognition (NER) for a number of languages, including Romanian. The POS tagging accuracy obtained by the TTL POS tagger exceeds 97% when TTL’s baseline model is updated with training information from a Romanian biomedical corpus. This corpus is developed in the context of the CoRoLa (a reference corpus for the contemporary Romanian language) project. Informative description and statistics of the Romanian biomedical corpus are also provided.

One of the fundamental tasks in natural-language processing is the morpho-lexical disambiguation of words occurring in text. Over the last twenty years or so, approaches to part-of-speech tagging based on machine learning techniques have been developed or ported to provide high-accuracy morpho-lexical annotation for an increasing number of languages. Due to recent increases in computing power, together with improvements in tagging technology and the extension of language typologies, part-of-speech tags have become significantly more complex. The need to address multilinguality more directly in the web environment has created a demand for interoperable, harmonized morpho-lexical descriptions across languages. Given the large number of morpho-lexical descriptors for a morphologically complex language, one has to consider ways to avoid the data sparseness threat in standard statistical tagging, yet ensure that full lexicon information is available for each word form in the output. The chapter overviews the current major approaches to part-of-speech tagging.

In this paper, we propose a two electrodes configuration for noncontact long-term cardiac monitoring. Our system was designed as an alternative to Holter devices. Unlike a traditional long-term monitoring equipment, our system is stationary and non-contact, the cardiac signal is acquired through a capacitive coupling between the sensor and the subject's body. The proposed system is the first prototype and a stepping stone in the development of a future bio-monitorisation platform dedicated to non-contact cardio-respiratory investigations.

Parallel corpora are essential resources for certain Natural Language Processing tasks such as Statistical Machine Translation. However, the existing publically available parallel corpora are specific to limited genres or domains, mostly juridical (e.g. JRC-Acquis) and medical (e.g. EMEA), and there is a lack of such resources for the general domain. This paper addresses this issue and presents a collection of parallel sentences extracted from the entire Wikipedia collection of documents for the following pairs of languages: English-German, English-Romanian and English-Spanish. Our work began with the processing of the publically available Wikipedia static dumps for the three languages in- volved. The existing text was stripped of the specific mark-up, cleaned of non- textual entries like images or tables and sentence-split. Then, corresponding documents for the above mentioned pairs of languages were identified using the cross-lingual Wikipedia links embedded within the documents themselves. Considering them comparable documents, we further employed a publically available tool named LEXACC, developed during the ACCURAT project, to extract parallel sentences from the preprocessed data. LEXACC assigns a score to each extracted pair, which is a measure of the degree of parallelism between the two sentences in the pair. These scores allow researchers to select only those sentences having a certain degree of parallelism suited for their intended purposes. This resource is publically available at: http://ws.racai.ro:9191/repository/search/?q=Parallel+Wiki

The project on the Romanian wordnet has been under continuous development for more than 10 years now. It has been in constant use in many projects and applications which determined, to a large extent, the content and coverage of various lexical domains. The article presents the most recent developments of the Romanian wordnet and offers quantitative data for its current version.

Our main goal is to ease the design of built-in intelligent diagnose algorithms and to develop performant cardiac monitors. First, the most popular methods of generating an ECG (electrocardiogram) signal, used at the moment for testing and maintenance of cardiac monitors, were discussed and reviewed; the method, which proved to be the most performant among all, was selected and software implemented in a cardiac monitor developed by our research team. We succeeded to develop a portable cardiac monitor that can generate authentic ECG signals. After validation this equipment will be used to provide advanced diagnostic and prognostic information to physicians, medical and bio-medical researchers.

The paper describes our recent developments in automatic extraction of translation equivalents from parallel corpora. We describe three increasingly complex algorithms: a simple baseline iterative method, and two non-iterative more elaborated versions. While the baseline algorithm is mainly described for illustrative purposes, the non-iterative algorithms outline the use of different working hypotheses which may be motivated by different kinds of applications and to some extent by the languages concerned. The first two algorithms rely on cross-lingual POS preservation, while with the third one POS invariance is not an extraction condition. The evaluation of the algorithms was conducted on three different corpora and several pairs of languages.

Given the unrestricted context for text-to-speech (TTS) synthesis and the current multilingual environment, TTS is often hampered by the presence of out-of-vocabulary (OOV) words. There are many precipitating factors for OOV words, from the use of technical terms, proper nouns, rare words that were not covered by the lexicon, and foreign words partially morphologically adapted; the latter, in fact, is a problem often confronted by non-English TTS synthesis systems. Furthermore, in order to derive natural speech from arbitrary text, all words that make up an utterance must undergo a series of complex processes such as: diacritic restoration; part-of-speech tagging; expansion to pronounceable form; syllabification; lexical stress prediction; and letter-to-sound conversion. OOV words require both automatic and trainable methods that can perform such tasks, which are usually based on a limited lexical context. The exception to this rule are those cases where part of speech and surrounding words are used as discriminative features such as in homograph disambiguation and abbreviation expansion. In this chapter we introduce the basic architecture of a generic natural language processing module in TTS synthesis, proposing data-driven solutions to various tasks, comparing our results concerning OOV words and prosody modeling with the current state-of-the-art TTS synthesis systems.

In an industrial plant it is very important to find out the faults in the shortest possible way and time. The paper describes a way of analysing and representing a generic functional component and constructing the complex of components as a skeleton of the facts database on which the fault diagnosis expert system will rely. Then, through intelligent encoding of sensors’ signal, the fault diagnosis is performed, the faults and their possibility degree are displayed. A case study expert system is presented for the diagnosis of a high pressure hydraulic installation in a rolling mill plant.

This paper presents an unsupervised word sense disambiguation. (WSD) algorithm that makes use of lexical chains concept [6] to quantify the degree of semantic relatedness between two words. Essentially, the WSD algorithm will try to maximize this semantic measure over a graph of content words in a given sentence in order to perform the disambiguation.

This paper describes the participation of the Research Institute for Artificial Intelligence of the Romanian Academy (RACAI) to the Multiple Language Question Answering Main Task at the CLEF 2008 competition. We present our Question Answering system answering Romanian questions from Romanian Wikipedia documents focusing on the implementation details. The presentation will also emphasize the fact that question analysis, snippet selection and ranking provide a useful basis of any answer extraction mechanism.

We present a pattern-based question answering system for Romanian that participated in the Romanian monolingual task of the [email protected] 2007 track. We aim to prove that working with a good Boolean searching engine and using question type driven answer extraction heuristics and structural matching between the linkage representations of the question and candidate sentences, one can achieve acceptable results (30% overall accuracy).

This paper describes the development of a Question Answering (QA) system and its evaluation results in the Romanian-English cross-lingual track organized as part of the CLEF 2006 campaign. The development stages of the cross-lingual Question Answering system are described incrementally throughout the paper, at the same time pinpointing the problems that occurred and the way they were addressed. The system adheres to the classical architecture for QA systems, debuting with question processing followed, after term translation, by information retrieval and answer extraction. Besides the common QA difficulties, the track posed some specific problems, such as the lack of a reliable translation engine from Romanian into English, and the need to evaluate each module individually for a better insight into the system's failures.

The paper presents a method for word sense disambiguation based on parallel corpora. The method exploits recent advances in word alignment and word clustering based on automatic extraction of translation equivalents and being supported by available aligned wordnets for the languages in the corpus. The wordnets are aligned to the Princeton Wordnet, according to the principles established by EuroWordNet. The evaluation of the WSD system, implementing the method described herein showed very encouraging results. The same system used in a validation mode, can be used to check and spot alignment errors in multilingually aligned wordnets as BalkaNet and EuroWordNet., Comment: 7 pages in Proc. of COLING2005

We briefly describe a word alignment system that combines two different methods in bitext correspondences identification. The first one is a hypotheses testing approach (Gale and Church, 1991; Melamed, 2001; Tufis 2002) while the second one is closer to a model estimating approach (Brown et al., 1993; Och and Ney, 2000). We show that combining the two aligners the results are significantly improved as compared to each individual aligner.

We provide a rather informal presentation of a prototype system for word alignment based on our previous translation equivalence approach, discuss the problems encountered in the shared-task on word-aligning of a parallel Romanian-English text, present the preliminary evaluation results and suggest further ways of improving the alignment accuracy.

Src homology 2 (SH2) domains have an important role in the regulation of protein activity and intracellular signaling processes. They are geared to bind to specific phosphotyrosine (pY) motifs, with a substrate sequence specificity depending on the three amino acids immediately C-terminal to the pY. Here we report for the first time the 1H, 15N and 13C backbone and side-chain chemical shift assignments for the C-terminal SH2 domain of the human protein tyrosine phosphatase PTPN11, both in its free and bound forms, where the ligand in the latter corresponds to a specific sequence of the human erythropoietin receptor.