Saturnino Luz

Maria Lizarazo Jimenez, Ana Gabriela Claros, Kieran Green et al.

Large Language Models (LLMs) are increasingly demonstrating the potential to reach human-level performance in generating clinical summaries from patient-clinician conversations. However, these summaries often focus on patients' biology rather than their preferences, values, wishes, and concerns. To achieve patient-centered care, we propose a new standard for Artificial Intelligence (AI) clinical summarization tasks: Patient-Centered Summaries (PCS). Our objective was to develop a framework to generate PCS that capture patient values and ensure clinical utility and to assess whether current open-source LLMs can achieve human-level performance in this task. We used a mixed-methods process. Two Patient and Public Involvement groups (10 patients and 8 clinicians) in the United Kingdom participated in semi-structured interviews exploring what personal and contextual information should be included in clinical summaries and how it should be structured for clinical use. Findings informed annotation guidelines used by eight clinicians to create gold-standard PCS from 88 atrial fibrillation consultations. Sixteen consultations were used to refine a prompt aligned with the guidelines. Five open-source LLMs (Llama-3.2-3B, Llama-3.1-8B, Mistral-8B, Gemma-3-4B, and Qwen3-8B) generated summaries for 72 consultations using zero-shot and few-shot prompting, evaluated with ROUGE-L, BERTScore, and qualitative metrics. Patients emphasized lifestyle routines, social support, recent stressors, and care values. Clinicians sought concise functional, psychosocial, and emotional context. The best zero-shot performance was achieved by Mistral-8B (ROUGE-L 0.189) and Llama-3.1-8B (BERTScore 0.673); the best few-shot by Llama-3.1-8B (ROUGE-L 0.206, BERTScore 0.683). Completeness and fluency were similar between experts and models, while correctness and patient-centeredness favored human PCS.

Saturnino Luz, Fasih Haider, Davida Fromm et al.

This Signal Processing Grand Challenge (SPGC) targets a difficult automatic prediction problem of societal and medical relevance, namely, the detection of Alzheimer's Dementia (AD). Participants were invited to employ signal processing and machine learning methods to create predictive models based on spontaneous speech data. The Challenge has been designed to assess the extent to which predictive models built based on speech in one language (English) generalise to another language (Greek). To the best of our knowledge no work has investigated acoustic features of the speech signal in multilingual AD detection. Our baseline system used conventional machine learning algorithms with Active Data Representation of acoustic features, achieving accuracy of 73.91% on AD detection, and 4.95 root mean squared error on cognitive score prediction.

Saturnino Luz

This chapter provides an introduction to computational linguistics methods, with focus on their applications to the practice and study of translation. It covers computational models, methods and tools for collection, storage, indexing and analysis of linguistic data in the context of translation, and discusses the main methodological issues and challenges in this field. While an exhaustive review of existing computational linguistics methods and tools is beyond the scope of this chapter, we describe the most representative approaches, and illustrate them with descriptions of typical applications.

Qingkun Deng, Saturnino Luz, Sofia de la Fuente Garcia

Depression is a common mental disorder. Automatic depression detection tools using speech, enabled by machine learning, help early screening of depression. This paper addresses two limitations that may hinder the clinical implementations of such tools: noise resulting from segment-level labelling and a lack of model interpretability. We propose a bi-modal speech-level transformer to avoid segment-level labelling and introduce a hierarchical interpretation approach to provide both speech-level and sentence-level interpretations, based on gradient-weighted attention maps derived from all attention layers to track interactions between input features. We show that the proposed model outperforms a model that learns at a segment level ($p$=0.854, $r$=0.947, $F1$=0.897 compared to $p$=0.732, $r$=0.808, $F1$=0.768). For model interpretation, using one true positive sample, we show which sentences within a given speech are most relevant to depression detection; and which text tokens and Mel-spectrogram regions within these sentences are most relevant to depression detection. These interpretations allow clinicians to verify the validity of predictions made by depression detection tools, promoting their clinical implementations.

Saturnino Luz, Masood Masoodian

The proliferation of consumer health devices such as smart watches, sleep monitors, smart scales, etc, in many countries, has not only led to growing interest in health monitoring, but also to the development of a countless number of ``smart'' applications to support the exploration of such data by members of the general public, sometimes with integration into professional health services. While a variety of health data streams has been made available by such devices to users, these streams are often presented as separate time-series visualizations, in which the potential relationships between health variables are not explicitly made visible. Furthermore, despite the fact that other aspects of life, such as work and social connectivity, have become increasingly digitised, health and well-being applications make little use of the potentially useful contextual information provided by widely used personal information management tools, such as shared calendar and email systems. This paper presents a framework for the integration of these diverse data sources, analytic and visualization tools, with inference methods and graphical user interfaces to help users by highlighting causal connections among such time-series.

Erfan Loweimi, Sofia de la Fuente Garcia, Saturnino Luz

We investigate the use of Large Language Models (LLMs) for zero-shot prediction of Ryff Psychological Well-Being (PWB) scores from spontaneous speech. Using a few minutes of voice recordings from 111 participants in the PsyVoiD database, we evaluated 12 instruction-tuned LLMs, including Llama-3 (8B, 70B), Ministral, Mistral, Gemma-2-9B, Gemma-3 (1B, 4B, 27B), Phi-4, DeepSeek (Qwen and Llama), and QwQ-Preview. A domain-informed prompt was developed in collaboration with experts in clinical psychology and linguistics. Results show that LLMs can extract semantically meaningful cues from spontaneous speech, achieving Spearman correlations of up to 0.8 on 80\% of the data. Additionally, to enhance explainability, we conducted statistical analyses to characterise prediction variability and systematic biases, alongside keyword-based word cloud analyses to highlight the linguistic features driving the models' predictions.

Erfan Loweimi, Sofia de la Fuente Garcia, Samira Loveymi et al.

LLMs can estimate Hospital Anxiety and Depression Scale (HADS) scores from speech in a zero-shot manner, but clinical deployment requires reliability across three dimensions: intra-model consistency, ASR robustness, and evidence faithfulness. We evaluate three LLMs (Phi-4, Gemma-2-9B, and Llama-3.1-8B) on 111 English-speaking participants using ground-truth transcripts and three Whisper ASR variants (Large, Medium, Small), with three independent runs per model-condition pair. We find that (i) Phi-4 and Gemma-2-9B achieve excellent intra-model consistency (ICC > 0.89) with minimal degradation under ASR; (ii) Llama-3.1-8B shows ASR-fragile consistency, with ICC dropping from 0.82 to 0.36 at 10% WER; (iii) predictive validity is largely preserved under ASR for robust models; and (iv) keyword groundedness exceeds 93% for Phi-4 and Gemma-2-9B but falls to 77-81% for Llama-3.1-8B. Inter-model keyword agreement is far lower than score-level agreement, revealing a score-evidence dissociation with implications for clinical interpretability.

Qingkun Deng, Saturnino Luz, Sofia de la Fuente Garcia

Speech-based depression detection tools could aid early screening. Here, we propose an interpretable speech foundation model approach to enhance the clinical applicability of such tools. We introduce a speech-level Audio Spectrogram Transformer (AST) to detect depression using long-duration speech instead of short segments, along with a novel interpretation method that reveals prediction-relevant acoustic features for clinician interpretation. Our experiments show the proposed model outperforms a segment-level AST, highlighting the impact of segment-level labelling noise and the advantage of leveraging longer speech duration for more reliable depression detection. Through interpretation, we observe our model identifies reduced loudness and F0 as relevant depression signals, aligning with documented clinical findings. This interpretability supports a responsible AI approach for speech-based depression detection, rendering such tools more clinically applicable.

Fuxiang Tao, Bahman Mirheidari, Madhurananda Pahar et al.

Dementia is associated with various cognitive impairments and typically manifests only after significant progression, making intervention at this stage often ineffective. To address this issue, the Prediction and Recognition of Cognitive Decline through Spontaneous Speech (PROCESS) Signal Processing Grand Challenge invites participants to focus on early-stage dementia detection. We provide a new spontaneous speech corpus for this challenge. This corpus includes answers from three prompts designed by neurologists to better capture the cognition of speakers. Our baseline models achieved an F1-score of 55.0% on the classification task and an RMSE of 2.98 on the regression task.

Oscar J. Ponce-Ponte, David Toro-Tobon, Luis F. Figueroa et al.

Shared decision-making (SDM) is necessary to achieve patient-centred care. Currently no methodology exists to automatically measure SDM at scale. This study aimed to develop an automated approach to measure SDM by using language modelling and the conversational alignment (CA) score. A total of 157 video-recorded patient-doctor conversations from a randomized multi-centre trial evaluating SDM decision aids for anticoagulation in atrial fibrillations were transcribed and segmented into 42,559 sentences. Context-response pairs and negative sampling were employed to train deep learning (DL) models and fine-tuned BERT models via the next sentence prediction (NSP) task. Each top-performing model was used to calculate four types of CA scores. A random-effects analysis by clinician, adjusting for age, sex, race, and trial arm, assessed the association between CA scores and SDM outcomes: the Decisional Conflict Scale (DCS) and the Observing Patient Involvement in Decision-Making 12 (OPTION12) scores. p-values were corrected for multiple comparisons with the Benjamini-Hochberg method. Among 157 patients (34% female, mean age 70 SD 10.8), clinicians on average spoke more words than patients (1911 vs 773). The DL model without the stylebook strategy achieved a recall@1 of 0.227, while the fine-tuned BERTbase (110M) achieved the highest recall@1 with 0.640. The AbsMax (18.36 SE7.74 p=0.025) and Max CA (21.02 SE7.63 p=0.012) scores generated with the DL without stylebook were associated with OPTION12. The Max CA score generated with the fine-tuned BERTbase (110M) was associated with the DCS score (-27.61 SE12.63 p=0.037). BERT model sizes did not have an impact the association between CA scores and SDM. This study introduces an automated, scalable methodology to measure SDM in patient-doctor conversations through explainable CA scores, with potential to evaluate SDM strategies at scale.

Saturnino Luz, Sofia De La Fuente Garcia, Fasih Haider et al.

We present a novel benchmark dataset and prediction tasks for investigating approaches to assess cognitive function through analysis of connected speech. The dataset consists of speech samples and clinical information for speakers of Mandarin Chinese and English with different levels of cognitive impairment as well as individuals with normal cognition. These data have been carefully matched by age and sex by propensity score analysis to ensure balance and representativity in model training. The prediction tasks encompass mild cognitive impairment diagnosis and cognitive test score prediction. This framework was designed to encourage the development of approaches to speech-based cognitive assessment which generalise across languages. We illustrate it by presenting baseline prediction models that employ language-agnostic and comparable features for diagnosis and cognitive test score prediction. The models achieved unweighted average recall was 59.2% in diagnosis, and root mean squared error of 2.89 in score prediction.

Saturnino Luz, Fasih Haider, Sofia de la Fuente et al.

Building on the success of the ADReSS Challenge at Interspeech 2020, which attracted the participation of 34 teams from across the world, the ADReSSo Challenge targets three difficult automatic prediction problems of societal and medical relevance, namely: detection of Alzheimer's Dementia, inference of cognitive testing scores, and prediction of cognitive decline. This paper presents these prediction tasks in detail, describes the datasets used, and reports the results of the baseline classification and regression models we developed for each task. A combination of acoustic and linguistic features extracted directly from audio recordings, without human intervention, yielded a baseline accuracy of 78.87% for the AD classification task, an MMSE prediction root mean squared (RMSE) error of 5.28, and 68.75% accuracy for the cognitive decline prediction task.

Sofia de la Fuente Garcia, Craig Ritchie, Saturnino Luz

Language is a valuable source of clinical information in Alzheimer's Disease, as it declines concurrently with neurodegeneration. Consequently, speech and language data have been extensively studied in connection with its diagnosis. This paper summarises current findings on the use of artificial intelligence, speech and language processing to predict cognitive decline in the context of Alzheimer's Disease, detailing current research procedures, highlighting their limitations and suggesting strategies to address them. We conducted a systematic review of original research between 2000 and 2019, registered in PROSPERO (reference CRD42018116606). An interdisciplinary search covered six databases on engineering (ACM and IEEE), psychology (PsycINFO), medicine (PubMed and Embase) and Web of Science. Bibliographies of relevant papers were screened until December 2019. From 3,654 search results 51 articles were selected against the eligibility criteria. Four tables summarise their findings: study details (aim, population, interventions, comparisons, methods and outcomes), data details (size, type, modalities, annotation, balance, availability and language of study), methodology (pre-processing, feature generation, machine learning, evaluation and results) and clinical applicability (research implications, clinical potential, risk of bias and strengths/limitations). While promising results are reported across nearly all 51 studies, very few have been implemented in clinical research or practice. We concluded that the main limitations of the field are poor standardisation, limited comparability of results, and a degree of disconnect between study aims and clinical applications. Attempts to close these gaps should support translation of future research into clinical practice.

Blaž Škrlj, Nika Eržen, Shane Sheehan et al.

Neural language models are becoming the prevailing methodology for the tasks of query answering, text classification, disambiguation, completion and translation. Commonly comprised of hundreds of millions of parameters, these neural network models offer state-of-the-art performance at the cost of interpretability; humans are no longer capable of tracing and understanding how decisions are being made. The attention mechanism, introduced initially for the task of translation, has been successfully adopted for other language-related tasks. We propose AttViz, an online toolkit for exploration of self-attention---real values associated with individual text tokens. We show how existing deep learning pipelines can produce outputs suitable for AttViz, offering novel visualizations of the attention heads and their aggregations with minimal effort, online. We show on examples of news segments how the proposed system can be used to inspect and potentially better understand what a model has learned (or emphasized).

Saturnino Luz, Fasih Haider, Sofia de la Fuente et al.

The ADReSS Challenge at INTERSPEECH 2020 defines a shared task through which different approaches to the automated recognition of Alzheimer's dementia based on spontaneous speech can be compared. ADReSS provides researchers with a benchmark speech dataset which has been acoustically pre-processed and balanced in terms of age and gender, defining two cognitive assessment tasks, namely: the Alzheimer's speech classification task and the neuropsychological score regression task. In the Alzheimer's speech classification task, ADReSS challenge participants create models for classifying speech as dementia or healthy control speech. In the the neuropsychological score regression task, participants create models to predict mini-mental state examination scores. This paper describes the ADReSS Challenge in detail and presents a baseline for both tasks, including feature extraction procedures and results for classification and regression models. ADReSS aims to provide the speech and language Alzheimer's research community with a platform for comprehensive methodological comparisons. This will hopefully contribute to addressing the lack of standardisation that currently affects the field and shed light on avenues for future research and clinical applicability.

Bridget Kane, Jing Su, Saturnino Luz

While machine learning (ML) systems have produced great advances in several domains, their use in support of complex cooperative work remains a research challenge. A particularly challenging setting, and one that may benefit from ML support is the work of multidisciplinary medical teams (MDTs). This paper focuses on the activities performed during the multidisciplinary medical team meeting (MDTM), reviewing their main characteristics in light of a longitudinal analysis of several MDTs in a large teaching hospital over a period of ten years and of our development of ML methods to support MDTMs, and identifying opportunities and possible pitfalls for the use of ML to support MDTMs.

Fasih Haider, Senja Pollak, Pierre Albert et al.

Research in automatic affect recognition has seldom addressed the issue of computational resource utilization. With the advent of ambient intelligence technology which employs a variety of low-power, resource-constrained devices, this issue is increasingly gaining interest. This is especially the case in the context of health and elderly care technologies, where interventions may rely on monitoring of emotional status to provide support or alert carers as appropriate. This paper focuses on emotion recognition from speech data, in settings where it is desirable to minimize memory and computational requirements. Reducing the number of features for inductive inference is a route towards this goal. In this study, we evaluate three different state-of-the-art feature selection methods: Infinite Latent Feature Selection (ILFS), ReliefF and Fisher (generalized Fisher score), and compare them to our recently proposed feature selection method named `Active Feature Selection' (AFS). The evaluation is performed on three emotion recognition data sets (EmoDB, SAVEE and EMOVO) using two standard acoustic paralinguistic feature sets (i.e. eGeMAPs and emobase). The results show that similar or better accuracy can be achieved using subsets of features substantially smaller than the entire feature set. A machine learning model trained on a smaller feature set will reduce the memory and computational resources of an emotion recognition system which can result in lowering the barriers for use of health monitoring technology.

Saturnino Luz, Sofia de la Fuente, Pierre Albert

We present an approach to automatic detection of Alzheimer's type dementia based on characteristics of spontaneous spoken language dialogue consisting of interviews recorded in natural settings. The proposed method employs additive logistic regression (a machine learning boosting method) on content-free features extracted from dialogical interaction to build a predictive model. The model training data consisted of 21 dialogues between patients with Alzheimer's and interviewers, and 17 dialogues between patients with other health conditions and interviewers. Features analysed included speech rate, turn-taking patterns and other speech parameters. Despite relying solely on content-free features, our method obtains overall accuracy of 86.5\%, a result comparable to those of state-of-the-art methods that employ more complex lexical, syntactic and semantic features. While further investigation is needed, the fact that we were able to obtain promising results using only features that can be easily extracted from spontaneous dialogues suggests the possibility of designing non-invasive and low-cost mental health monitoring tools for use at scale.

Saturnino Luz, Masood Masoodian

We present an application of mosaic diagrams to the visualisation of set relations. Venn and Euler diagrams are the best known visual representations of sets and their relationships (intersections, containment or subsets, exclusion or disjointness). In recent years, alternative forms of visualisation have been proposed. Among them, linear diagrams have been shown to compare favourably to Venn and Euler diagrams, in supporting non-interactive assessment of set relationships. Recent studies that compared several variants of linear diagrams have demonstrated that users perform best at tasks involving identification of intersections, disjointness and subsets when using a horizontally drawn linear diagram with thin lines representing sets, and employing vertical lines as guide lines. The essential visual task the user needs to perform in order to interpret this kind of diagram is vertical alignment of parallel lines and detection of overlaps. Space-filling mosaic diagrams which support this same visual task have been used in other applications, such as the visualization of schedules of activities, where they have been shown to be superior to linear Gantt charts. In this paper, we present an application of mosaic diagrams for visualization of set relationships, and compare it to linear diagrams in terms of accuracy, time-to-answer, and subjective ratings of perceived task difficulty. The study participants exhibited similar performance on both visualisations, suggesting that mosaic diagrams are a good alternative to Venn and Euler diagrams, and that the choice between linear diagrams and mosaics may be solely guided by visual design considerations.