DARTH group mission

Ever-increasing healthcare provision demands strain national health systems globally, which struggle to meet patients' needs. There is a growing unmet clinical need to develop accurate, robust, practical decision support tools to facilitate diagnosis, assessment, and symptom monitoring to improve patient health and care.

We are a dynamic research group based at the Usher Institute, Edinburgh Medical SchoolUniversity of Edinburgh working at the interface of engineering, mathematics, informatics, and medicine. We passionately believe we can design and provide effective solutions which will revolutionize contemporary healthcare delivery through capitalizing technology and harnessing data.

We develop and apply signal processing and statistical machine learning algorithms to explore data and decipher complicated concealed statistical relationships. Our algorithms are directly driven by and validated on complicated real-world problems, aiming to facilitate interpretation of the underlying key mechanisms of the modelled system. Our work is inherently multi-disciplinary and we collaborate with industrial partners and researchers worldwide. We tackle challenging problems in healthcare domains from neurodegenerative disorders and mental health, to asthma, cardiovascular disease, and neonatal monitoring.


12 May 2021

New journal article accepted for publication in Clinical Chemistry 

Cardiac troponin is a widely used biomarker and has been associated with cardiovascular events. It is unclear whether sex-specific troponin thresholds should be considered in clinical guidelines to inform cardiovascular events. We used data from the Generation Scotland, a well-phenotyped family-based cohort with more than 24,000 participants, and demonstrate that cardiac troponin concentrations are stronger predictors of cardiovascular events in women.  

1 May 2021

Andres wins the Edinburgh Branch of Parkinson’s UK (ERIG) talk prize 

ERIG had invited PhD supervisors to nominate students who work in this research area and Andres has won this award for his presentation: “An engineering approach as a support tool for Parkinson’s". In his talk he had highlighted some of his recent PhD findings capitalizing on voice as a robust tool towards providing new insights into Parkinson's disease. The panel commented Andres for making the talk accessible to a lay audience. This engagement work was started by Prof. Ken Bowler with the establishment of the annual Edinburgh Parkinson’s Lecture.

30 March 2021

Evi wins the SPR MD/PhD student research award in the Pediatrics conference

Evi has won this award at the Society for Pediatrics Research conference (one of the biggest international academic meetings in the area of Pediatrics) for her research paper: “Machine Learning for Stratification of children at risk of language delay following Preterm Birth". This is work that Evi publishes as lead author, and is a multi-centre collaboration with colleagues across the University of Edinburgh. She will be presenting her work at the conference in May.

17 February 2021

New journal article accepted for publication in Frontiers in Human Neuroscience

"A neuromotor to acoustical jaw-tongue projection model with application in Parkinson’s disease hypokinetic dysarthria". We explored concurrent speech, surface electromyography, and three-dimensional accelerometry signals to assess the neuromotor activity of the massseter-jaw-tongue articulation in Parkinson's disease (PD). We demonstrate that there are important cross-correlations which may serve as useful acoustic biomarkers of hypokinetic dysarthria in PD.

13 February 2021

Thanasis won the Biosignals 2021 best paper award

"Assessing Parkinson’s disease speech signal generalization of clustering results across three countries: findings in the Parkinson’s voice initiative". We demonstrated that we can identify four robust clusters of Parkinson's disease using sustained vowels which generalize well on three cohorts collected across three countries as part of the Parkinson's Voice Initiative. These findings have important implications for cost-effective group membership assignment.   

1 February 2021

Minhong has successfully defended her PhD thesis 

"New insights into probabilistic pattern formation of embryonic stem cells using agent-based modelling”. Minhong proposed an agent-based modelling approach to identify biologically plausible rules acting at the meso-scale within stem cell collectives that may explain spontaneous patterning. She introduced a new distance-based metric to assess the deviation between probabilistic ground truth and probabilistic simulated outcomes. Her work provides some useful key insights into stem cell understanding, facilitating further investigations towards the development of novel treatments.

27 January 2021

New journal article accepted for publication in IEEE Access

"Smartphone speech testing for symptom assessment in rapid eye movement sleep behavior disorder and Parkinson's disease”. We analyzed 4242 smartphone voice recordings from people diagnosed with REM sleep behavior disorder and Parkinson's disease, and contrasted findings with data from controls. We demonstrate the potential of our approach towards a novel digital biomarker facilitating intervention in the early and prodromal stages of Parkinson's disease.

11 January 2021

New journal article accepted for publication in Biomedical Signal Processing and Control

"Acoustic to kinematic projection in Parkinson’s disease dysarthria”. We investigated the effects of neuromotor activity during muscular exertion that translates formant acoustics into speech articulatory movements affected by hypokinetic dysarthria in Parkinson’s Disease (PD). The study provides new mechanistic insights into the biomechanical system of voice production and the understanding of PD effects.

Previous years in review

25 December 2020

New journal article accepted for publication in IEEE Access

"Remote assessment of Parkinson's disease symptom severity using the simulated cellular mobile telephone network”. We had previously demonstrated the use of high-quality lab-based speech signals to (1) differentiate people with Parkinson's from controls, and (2) replicate UPDRS, a clinical scale quantifying symptom severity. Here, we extend findings demonstrating the use of the standard cellular telephone network towards large-scale Parkinson's telemonitoring.

21 December 2020

New conference articles accepted for oral presentations at BIOSTEC 2021

Andres and Thanasis got their manuscripts accepted as full oral presentations at BIOSTEC. Both papers investigate Parkinson's disease: Andres' paper uses a more mechanistic, first-principles approach of voice production and resulting deficiencies in Parkinson's disease. Thanasis' paper demonstrates new robust clustering results across three English-speaking cohorts and makes a strong case for four distinct subgroups. The paper is shortlisted for the 'best paper award'.

27 November 2020

New journal article accepted for publication in BMC Medical Research Methodology

"Linkage of primary care prescribing records and pharmacy dispensing records in the Salford lung study: application in asthma”. Motivated by challenges in assessing adherence, we developed a novel probabilistic record linkage methodology to match medication pharmacy dispensing records and primary care prescribing records, using semantic (meaning) and syntactic (structure) harmonization, domain knowledge integration, and natural language feature extraction.

2 September 2020

New journal article accepted for publication in Scientific Reports

"Predicting pattern formation in embryonic stem cells using a minimalist, agent-based probabilistic model”. We explored new physiologically plausible rule-based approaches using agent-based modelling to tentatively decipher stem cell pattern formations. This work may have important implications in understanding underlying movement characteristics of stem cells, thus paving the way towards developing novel therapies.  

20 August 2020

New journal article accepted for publication in Scientific Reports

"A data-driven typology of asthma medication adherence using cluster analysis”. Medication non-adherence is strongly associated with poor control of symptoms and increased morbidity and mortality. We study in detail the nuanced patterns of medication taking and propose that monitoring patient adherence behaviour is necessary to assess the impact of interventions, and to determine the effect on clinical outcomes.

28 July 2020

New journal article accepted for publication in IEEE Transactions on Affective Computing

"Beyond Mobile Apps: a Survey of Technologies for Mental Well-being”. Mental disorders are closely associated with fear of stigma, structural barriers such as financial burden, and lack of available services and resources which often prohibit the delivery of frequent clinical advice and monitoring. This review article provides an overview of traditional techniques followed by their technological alternatives, sensing devices, behaviour changing tools, and feedback interfaces. 

24 June 2020

New journal article accepted for publication in the British Journal of Clinical Pharmacology

"Measuring and reporting treatment adherence: what can we learn by comparing two respiratory conditions?". This study investigated medication non-adherence and its longer-term effects in terms of side-effects, morbidity, and mortality. We used as testbeds two respiratory conditions, asthma and tuberculosis, to cross-reference approaches towards assessing adherence, and making recommendations for effective adherence reporting.

24 June 2020

New journal article accepted for publication in the Journal of Child Psychology and Psychiatry

"Eye-tracking for longitudinal assessment of social cognition in children born preterm". We have monitored preterm-born and term-born children to investigate social attentional preference in infancy and at 5 years, its relationship with neurodevelopment, and the influence of socioeconomic deprivation.


27 May 2020

New journal article accepted for publication in Neurocomputing

"Artificial intelligence within the interplay between natural and artificial Computation: advances in data science, trends and applications". This comprehensive study from an inter-disciplinary international team summarizes many of the recent advances in data science and artificial intelligence within the interplay between natural and artificial computation.

8 May 2020

New journal article accepted for publication in Stroke

"Telemedicine cognitive hebavioural therapy for anxiety after stroke: proof of concept randomized controlled trial". Anxiety affects a large proportion of stroke survivors. We developed a telemedicine-based approach for delivering guided self-help cognitive behavioural therapy for anxiety after stroke. The study also explored the use of objective assessment using wrist-worn actigraphy to complement patient reported outcome measures.

2 March 2020

New journal article accepted for publication in JMIR mHealth and uHealth

"Objective characterization of activity, sleep, and circadian variability patterns using a wrist-worn activity sensor: insights into post-traumatic stress disorder". This study provides new approaches towards visualizing, extracting patterns, and interpreting findings to characterize actigraphy data. The application is on post-traumatic stress disorder, but the algorithmic tools can be applied in any setting where we passively record actigraphy signals using wrist-worn wearables.

11 February 2020

New journal article accepted for publication in JMIR Medical Informatics

"Challenges of clustering multimodal clinical data: a review of applications in asthma subtyping" investigates common pitfalls in the application of clustering methodologies in clinical settings, using asthma as an exemplar. The manuscript makes a strong case for careful algorithmic considerations based on data variable type, sample size, and more general issues that appear in multimodal clinical datasets when attempting to infer data properties through clustering.

19 January 2020

Four new conference papers accepted for publication in BioSignals (BIOSTEC 2020)

Thanasis is co-organising a special session 'SERPICO' as part of the 13th Internatonal Joint Conference on Biomedical Engineering Systems and Technologies, and four members of the group (Andres, Evi, Minhong, Thanasis) will be presenting work across different topics. Andres and Thanasis present findings on Parkinson's disease, Evi presents work from her PhD on neonatal monitoring, and Minhong presents her latest findings on stem cell pattern formation.


1 September 2019

Three new conference papers accepted for publication in IEEE BIBE

Holly, Minhong, and Thanasis will be presenting their research findings in the next IEEE BioInformatics and BioEngineering conference in Athens, Greece at the end of October 2019. Thanasis is co-organising a special session at the conference entitled Intelligent Digital Health Interventions towads Prevention, Self-Management, and Treatment of Pathologies.

16 August 2019

New journal article accepted for publication in Circulation

We developed a statistical machine learning algorithm which uses changes in cardiac troponin concentrations, controlling for age and gender, and demonstrated we can very accurately predict myocardial infarction. The findings have been validated on a very large cohort of approximately 8,000 people. The algorithm can be used to identify low-risk and high-risk patients and lead to better-informed clinical decisions.

1 August 2019

New conference papers accepted and keynote lecture - MAVEBA

Andres will be presenting his work on: "A neuromechanical model of jaw-tongue articulation in Parkinson's disease speech", and Thanasis will be presenting results from the Parkinson's Voice Initiative study. Thanasis will also be delivering this year's keynote lecture in MAVEBA, entitled: Developing new speech signal processing algorithms for biomedical and life sciences applications: principles, findings, challenges, and a view to the future

4 June 2019

New journal article accepted for publication in BMJ Open

Asthma is a long-term condition with rapid onset worsening of symptoms (‘attacks’) which can be unpredictable and may prove fatal.  Models predicting asthma attacks require high sensitivity to minimise mortality risk, and high specificity to avoid unnecessary prescribing of preventative medications that carry an associated risk of adverse events.  This protocol paper describes the planned work to create a risk score to predict asthma attacks in primary care using a statistical learning approach trained on routinely collected electronic health record (EHR) data.


14 May 2019

New journal article accepted for publication in Scientific Reports

"Quantifying ultrasonic mouse vocalizations using acoustic analysis in a supervised statistical machine learning framework"  Most mouse communication is produced in ultrasonic frequencies beyond human hearing. These ultrasonic vocalizations (USVs) are typically described by experts using nine call types. In this study, we replicated the expert-defined call types of communicative vocal behavior with 85% accuracy by using acoustic analysis to characterize USVs and a principled supervised learning setup. This has important implications in understanding animal behaviors.


11 April 2019

New conference article accepted for publication in EMBC

"New insights into Parkinson’s disease through statistical analysis of standard clinical scales quantifying symptom severity"  In clinical practice, neurologists do not have time to use the full span of clinical scales and anecdotally rely on their experience to use a subset of the available items in fully established clinical metrics assessing Parkinson's disease symptom severity. This study aimed to investigate the practical implications of standard clinical assessment tools in monitoring Parkinson's disease symptom severity.


10 April 2019

New journal article accepted for publication in JASA

"Developing a large scale population screening tool for the assessment of Parkinson’s disease using telephone-quality speech". We collected the largest speech database in the world through the Parkinson's Voice Initiative and aimed to demonstrate that speech recordings collected under highly uncontrolled conditions may lead to acceptable accuracy in differentiating healthy controls from people diagnosed with Parkinson's disease simply processing sustained vowels collected over the standard telephone network. 


22 March 2019

New article published in Urology News, highlighted in the frontpage of the magazine

"Rise of the machines: will artificial intelligence replace the urologist?" In this article we review the state of the art in the field of Artificial Intelligence (AI) focusing in the field of urology with a view to contemporary developments and future directions. We argued that data scientists and urologists need to work in close collaboration to harness opportunities: "rather than replace urologists, AI will mainly be used to inform, enhance, and complement their practice." 

17 February 2019

New PhD student joining the DARTH group

Congratulations to Andres Gomez-Rodellar for successfully interviewing at the highly competitive DTP in Precision Medicine programme and being offered a position to join our research group. Andres will be starting with us in September 2019.


20 January 2019

New journal article accepted for publication in JMIR

Machine learning has attracted considerable research interest towards developing smart digital health interventions. These interventions have the potential to revolutionize healthcare and lead to substantial outcomes for patients and medical professionals. Unfortunately, research findings are rarely translated in clinical interventions. This study provides a literature review to contextualize the current state of the art in this setting and highlight contemporary limitations.


Setting up the DARTH group.