Introduction

Poster HC2019

The main goal of the 3rd Industrial workshop, promoted by the Portuguese Network for Mathematics in Industry and Innovation (PT-MATHS-IN), is to inspire new approaches to address the challenges faced by the health industry nowadays. Bringing together specialists from the health ecosystem and researchers from several areas, this workshop aims to discuss those challenges, to hear perspectives on the latest trends in healthcare and to identify promising opportunities where mathematics plays a significant role to improve that field.

Healthcare challenges are spread across many different areas such as precision medicine, big data and data management, wellness and prevention, or the social and ethical implications of advances in healthcare. Mathematics tools are often required do deal with these problems and are essential to derive effective solutions.

After last year's successes, achieved with the Big Data and Smart Security industrial days, this one-day event aims to keep bringing together Academy and Industry in the discussion of areas of great scientific or industrial relevance and with the highest societal impact.

Gathering industry, research community, start-ups and regulators in an out-of-the-box discussion at the historical and vibrant University of Coimbra - classified by UNESCO as World Heritage site - PT-MATHS-IN aims to help the participants to contact with new ideas and trends on this field, as well as to extend their network among industrials and researchers on this area.

The PT-MATHS-IN consortium of the thirteen main Portuguese mathematics research centres serves as an interface between the Portuguese Industry and the industrial mathematics community. PT-MATHS-IN also acts as the national representative of Portugal in the European Service Network for Mathematics in Industry and Innovation (EU-MATHS-IN).

This conference, organized by PT-MATHS-IN, is sponsored and supported by several institutions. Check out our sponsors and supporters. Contact us at hc2019@spm.pt if you want to become a sponsor/supporter.

Sponsors

Keynote Speakers

	Ana Paula Rocha CMUP and Department of Mathematics, FCUP, University of Porto Challenges of heart rate variability analysis in critical illness and critical care A journey into modelling and analysis of heart rate variability, as a window to the autonomic nervous system characterization, and with the aim of risk assessment. Complex/nonlinear characteristics of HRV are considered, namely long memory and volatility from a parametric point of view. This methodology allowed the design of individualized filters and obtaining clinically interpretable measures, reflecting changes in HRV dynamics related with the stress response to acute brain lesion and medical management. More recently we are also addressing the emerging field of network physiology, with the challenge to document the brain-organ crosstalk.
	André Dias Pereira Centre for Biomedical Law, Faculty of Law University of Coimbra, National Council of Ethics for Life Sciences The great robot doctor or the right to a human (inter)face? Healthcare and big data are key topics of scientific research and financial investment in recent years and is deemed to increase significantly. The risks of transformation of healthcare delivery and the continuance of many healthcare professionals are real; contrariwise, the potential for Artificial Intelligence (AI) to gain ground in this area of society is immense. Ethics and Law are considering several issues: from data protection and privacy, including of genetic information, as well as the right to keep a human interface in such vulnerable situations that arise from disease (UNESCO Declaration of Bioethics and Human Rights 2005 (Art. 8)). To imagine a confrontation between the machine (AI and robotics) and humans is not science fiction (Gerd Leonhard); mathematics is playing a leading role in this 4th industrial revolution that also affects or affects foremost traditional intellectual activities, such as medicine and nursing, but… “Defending humane relationships will be one of the key challenges to bioethics in the coming decades!” (Michael Cook, BioEdge)
	Bart Romeny Department of Biomedical Engineering, Eindhoven University of Technology On the neuro-mathematics of deep learning The success of deep learning neural networks is evident, leading to a revolution in the field. The breakthrough is enabled by the discovery of how to train a multi-layered neural net with backpropagation, by the advent of cheap GPU processing power and the availability of huge amounts of curated training data. The first part of the lecture will focus on examples of important convolutional neural networks today. Then, we zoom in on 'brain-inspired computing', with an important biomedical application: the large-scale screening for diabetes by automatic analysis of retinal fundus images. With the progression of diabetes, blood vessels begin to leak, and this can be detected at high resolution and at low cost in the retina. Especially in Asia diabetes is exploding, with China now close to 12%. It is a major source of blindness. The Sino-Dutch RetinaCheck project aims to screen millions of people in the province Liaoning, China. I will carefully explain the brain-inspired image analysis tools we successfully developed for automated and quantitative biomarker analysis. The second part will focus on the intrinsic mathematics of deep learning. How does the incremental contextual processing actually work? We can learn a lot from modern brain research, where optical and physiological recording techniques shed new light on what the functional circuits in the brain may be computing. Bridging the fields of computational models in deep learning artificial intelligence, and biological models of brain research is highly rewarding. The biomedical engineer, speaking both languages, is just the right professional to benefit from and contribute to these developments. The lecture will be highly visual and is aimed to a broad audience. The speaker is known for his excellent educational skills.
	Inês Sousa CBMA and Department of Mathematics, School of Sciences, University of Minho Forecasting daily admissions for an hospital emergency department This study aims, through patterns and trends, to develop models that allow to predict on a daily basis the number of patients that visit an hospital emergency department. Data on the number of admissions for an hospital emergency department was daily available on four consecutive years and prediction was made for the first trimester of the following year with no data. These data were analysed with a two-stage approach using linear regression and a time series statistical models. In the first step we model the trend of the time series allowing also for all the covariates available, using regression methods. In the second step we model temporal correlations still present in the residuals of the first step model, using time series techniques. Calendar, meteorological, environmental and epidemiological variables were tested as explanatory variables in the statistical model. However, all variables except calendar ones, proved to have small predictive value. This methodology allows emergency departments to manage resources in advance.
	Isabel Narra Figueiredo Department of Mathematics, FCTUC, University of Coimbra Computer assisted methods for automatic analysis of medical images Description of some different computer-assisted methods we have developed for the automatic interpretation of ophthalmological images and also gastroenterological images. In particular, we describe an automatic screening method for diabetic retinopathy, in use in Portuguese public hospitals, different colonic polyp detector methods, and, finally, a method towards wireless capsule endoscope localization.
	João Malva Coimbra Institute for Clinical and Biomedical Research (iCBR), Faculty of Medicine, University of Coimbra A quadruple helix-based regional ecosystem to support healthy living and active ageing Ageing research and innovative practices to support healthy living and active ageing became Demographic Changes priorities in Horizon 2020. The European Innovation Partnership on Active and Healthy Ageing (EIP on AHA) stands on networks supported by quadruple helix-based regional ecosystems, the Reference Sites. Reference Sites show large-scale replication potential of innovative practices supporting independent living of citizens, the “blueprint for digital transformation on health and care in an ageing society” and “boosting innovation on active and healthy ageing in the digital single market”. The concept of the knowledge triangle connecting knowledge production, high education and business activity has settled the architecture of the so called Knowledge Innovation Communities (KIC′s), including the EIT Health that supports Healthy Living, Active Ageing and Improvements in Healthcare in Europe. In the first 4 years of operation, EIT Health has delivered an impressive number of Academia-Business collaborative projects by means of innovative education modalities, innovation projects, entrepreneurship and acceleration of businesses. The intimate collaboration of EIT Health and EIP on AHA partners produces real impact on society, business and policy/decision making, seeding innovation and accelerating go-to-market businesses. Ageing@Coimbra is the Reference Site of the Center Region of Portugal, awarded following the EU call for new Reference Sites in 2012. Ageing@Coimbra, strongly supported by the Regional Authority (CCDRC) has delivered innovative projects with real impact in the region. We highlight a recently EU-funded flagship project, the Multidisciplinary Institute of Ageing (MIA-Portugal; April 2019). MIA-Portugal is a Horizon 2020 Teaming partnership between one academic partner in a low-performer country (UC in Portugal) and two academic partners in high-performer countries (UMCG in the Netherlands and UNEW in the UK). MIA-Portugal, global investment >45 M€, is the first Center of Excellence in ageing research in Southern Europe and a real example of synergies between EIP on AHA Reference Sites, EIT Health and strategic funding from Research Executive Agency (REA) of the European Union and Regional Structural and Cohesion Funds. Healthy Living and Active Ageing outcomes produced by Horizon 2020 has show maturity level for maximal impact in Horizon Europe, to benefit European Citizens. Reference Sites, EIT Health and the Widespread Program (with the flagship example of MIA-Portugal) created a virtuous circle to maximize ageing research impact into the society and support widespread opportunities contributing to science and social cohesion in Europe.
	Ricardo Teresa Ribeiro GoatLabs \| ESTeSL - IPL The Development of Chest X-Ray 2.0 The Chest X-Ray (CXR) represents roughly 40% of all imaging procedures performed worldwide. It′s a preliminary diagnosis tool, enabling the detection of a wide range of lung and cardiovascular diseases. CXR shows great advantages, namely it′s low cost and easiness to perform, and it gathers a large amount of information regarding a patient′s health. As medical images become more detailed, accurate and ultimately more complex, radiologists and medical doctors have an increased pressure to balance thorough analysis and good levels of operational efficiency. Automation and augmentation are therefore severely needed, supporting a high-quality analysis in a reasonable time-to-diagnosis. In this talk I′ll present the development of our computer assisted tool for the analysis of CXR. In particular, I will be focusing on the challenges of translation academic work to product development. Our goal is to develop a solution that consists in the observation of imaging findings, augmenting with machine learning descriptors to provide a medical interpretation of those observations as a holistic analysis.
	Rui Rodrigues Department Mathematics and CMA - FCT University Nova de Lisboa Pattern recognition in continuous biomedical signals Physiological signals like ECG, blood pressure and respiration are closely related. However, it is challenging to develop algorithms that explore these relations to improve diagnostic or patient monitoring. We will present some examples: detection of life threatening arrhythmia alarms in the Intensive Care Unit, noninvasive fetal ECG and non invasive respiration monitoring. We will discuss several mathematical approaches and their limitations.
	Wil Schilders President of EU-MATHS-IN Healthcare - Challenges and opportunities from EU-MATHS-IN point of view

Programme

Venue

University of Coimbra: Sala Pedro Nunes, Department of Mathematics, Praça Dom Dinis; 3001-501 Coimbra; Portugal

University of Coimbra - UNESCO World Heritage from Universidade de Coimbra on Vimeo.

Committees

Scientific Committee: Manuel Cruz (LEMA/ISEP); Paula Amaral (CMA/FCT-UNL); Paulo Vasconcelos (CMUP/FCUP); Rui Bernardes (FMUC); Rui Borges Lopes (CIDMA/UA); Sílvia Barbeiro (CMUC/UC)

Organizing Committee: Daniela Jordão (CMUC/UC); Diogo Lobo (CMUC/UC); Manuel Cruz (LEMA/ISEP); Paula Amaral (CMA/FCT-UNL); Paulo Vasconcelos (CMUP/FCUP); Rui Borges Lopes (CIDMA/UA); Sílvia Barbeiro (CMUC/UC)

Contacts

PT-MATHS-IN: Av. da República, 45 3º esq; 1050-187 Lisboa | Portugal; ☎ +351 217 939 785; hc2019@spm.pt

Supporters