arXiv subjects

Daniel Ortega

Publications and source records attributed to Daniel Ortega.

9 recordsLinked to original sources

arXiv2026

Unravelling Challenges in Heating Power Measurements for Magnetic Hyperthermia -- the RADIOMAG Round Robin Study Revisited

Non-adiabatic AC calorimetry is the most widely used technique for estimating the heating power of magnetic nanoparticles in magnetic hyperthermia. However, it is prone to systematic errors which lead to a standard deviation in the intrinsic loss power (ILP) of approximately 30-40%, as revealed by the RADIOMAG EU COST Action TD1402 round-robin study involving 21 European laboratories. In this study, we re-examine the RADIOMAG dataset to both uncover previously unreported instrumentation issues, and to explore more deeply some of the reported instrumentation issues. We identify four common sources of error: i) Insufficient temperature resolution, ii) AC-field sensitive thermometers, iii) Non-physical temperature oscillations, and iv) Apparent non-linear heat loss. Based on these findings, we propose criteria for sufficient measurement quality and apply them to re-estimate the ILP values. These results have a standard deviation of 18-30%., demonstrating that addressing instrumentation and analysis issues can improve measurement reliability and decrease the inter-laboratory deviation by up to 38%. When re-estimating ILP, we used the initial slope method, arguing that the corrected slope method, which was previously used to investigate the RADIOMAG data, could introduce misleading interpretations of systematic ILP deviations due to sub-optimal measurement conditions and the unnoticed influence of non-linear heat losses. However, we emphasise that the corrected slope method is preferred, given a linear heat loss. Based on our analysis, we introduce a diagnostic protocol by using slope curves - a simple yet effective plot type - for identifying and solving common instrumentation challenges proactively before the data acquisition phase.

physics.app-ph↗

arXiv2023

Modeling Speaker-Listener Interaction for Backchannel Prediction

We present our latest findings on backchannel modeling novelly motivated by the canonical use of the minimal responses Yeah and Uh-huh in English and their correspondent tokens in German, and the effect of encoding the speaker-listener interaction. Backchanneling theories emphasize the active and continuous role of the listener in the course of the conversation, their effects on the speaker's subsequent talk, and the consequent dynamic speaker-listener interaction. Therefore, we propose a neural-based acoustic backchannel classifier on minimal responses by processing acoustic features from the speaker speech, capturing and imitating listeners' backchanneling behavior, and encoding speaker-listener interaction. Our experimental results on the Switchboard and GECO datasets reveal that in almost all tested scenarios the speaker or listener behavior embeddings help the model make more accurate backchannel predictions. More importantly, a proper interaction encoding strategy, i.e., combining the speaker and listener embeddings, leads to the best performance on both datasets in terms of F1-score.

cs.CL↗

arXiv2023

Oh, Jeez! or Uh-huh? A Listener-aware Backchannel Predictor on ASR Transcriptions

This paper presents our latest investigation on modeling backchannel in conversations. Motivated by a proactive backchanneling theory, we aim at developing a system which acts as a proactive listener by inserting backchannels, such as continuers and assessment, to influence speakers. Our model takes into account not only lexical and acoustic cues, but also introduces the simple and novel idea of using listener embeddings to mimic different backchanneling behaviours. Our experimental results on the Switchboard benchmark dataset reveal that acoustic cues are more important than lexical cues in this task and their combination with listener embeddings works best on both, manual transcriptions and automatically generated transcriptions.

cs.CL↗

arXiv2022

Estimating the heating of complex nanoparticle aggregates for magnetic hyperthermia

Understanding and predicting the heat released by magnetic nanoparticles is central to magnetic hyperthermia treatment planning. These nanoparticles tend to form aggregates when injected in living tissues, which alters their response to the applied alternating magnetic field and prevents predicting the released heat accurately. We performed an in silico analysis to investigate the heat released by nanoparticle aggregates featuring different size and fractal geometry factors. By digitally mirroring aggregates seen in biological tissues, we found that the average heat released per particle stabilizes starting from moderately small aggregates, facilitating the estimates for their larger counterparts. Additionally, we studied the heating performance of particle aggregates over a wide range of fractal parameters. We compared this result with the heat released by non-interacting nanoparticles to quantify the reduction of heating power after being instilled into tissues. This set of results can be used to estimate the expected heating in vivo based on the experimentally determined nanoparticle properties.

cond-mat.mes-hall↗

arXiv2020

ADVISER: A Toolkit for Developing Multi-modal, Multi-domain and Socially-engaged Conversational Agents

We present ADVISER - an open-source, multi-domain dialog system toolkit that enables the development of multi-modal (incorporating speech, text and vision), socially-engaged (e.g. emotion recognition, engagement level prediction and backchanneling) conversational agents. The final Python-based implementation of our toolkit is flexible, easy to use, and easy to extend not only for technically experienced users, such as machine learning researchers, but also for less technically experienced users, such as linguists or cognitive scientists, thereby providing a flexible platform for collaborative research. Link to open-source code: https://github.com/DigitalPhonetics/adviser

cs.CL↗

arXiv2019

Context-aware Neural-based Dialog Act Classification on Automatically Generated Transcriptions

This paper presents our latest investigations on dialog act (DA) classification on automatically generated transcriptions. We propose a novel approach that combines convolutional neural networks (CNNs) and conditional random fields (CRFs) for context modeling in DA classification. We explore the impact of transcriptions generated from different automatic speech recognition systems such as hybrid TDNN/HMM and End-to-End systems on the final performance. Experimental results on two benchmark datasets (MRDA and SwDA) show that the combination CNN and CRF improves consistently the accuracy. Furthermore, they show that although the word error rates are comparable, End-to-End ASR system seems to be more suitable for DA classification.

cs.CL↗

arXiv2018

Lexico-acoustic Neural-based Models for Dialog Act Classification

Recent works have proposed neural models for dialog act classification in spoken dialogs. However, they have not explored the role and the usefulness of acoustic information. We propose a neural model that processes both lexical and acoustic features for classification. Our results on two benchmark datasets reveal that acoustic features are helpful in improving the overall accuracy. Finally, a deeper analysis shows that acoustic features are valuable in three cases: when a dialog act has sufficient data, when lexical information is limited and when strong lexical cues are not present.

cs.CL↗

arXiv2017

Neural-based Context Representation Learning for Dialog Act Classification

We explore context representation learning methods in neural-based models for dialog act classification. We propose and compare extensively different methods which combine recurrent neural network architectures and attention mechanisms (AMs) at different context levels. Our experimental results on two benchmark datasets show consistent improvements compared to the models without contextual information and reveal that the most suitable AM in the architecture depends on the nature of the dataset.

cs.CL↗

arXiv2016

Fundamentals and advances in magnetic hyperthermia

Nowadays, magnetic hyperthermia constitutes a complementary approach to cancer treatment. The use of magnetic particles as heating mediators, proposed in the 1950s, provides a novel strategy for improving tumor treatment and, consequently, patient quality of life. This review reports a broad overview about several aspects of magnetic hyperthermia addressing new perspectives and the progress on relevant features such as the ad hoc preparation of magnetic nanoparticles, physical modeling of magnetic heating, methods to determine the heat dissipation power of magnetic colloids including the development of experimental apparatus and the influence of biological matrices on the heating efficiency.

cond-mat.soft↗