School of Computing Science

Speaker: Dr Caroline Hogarth*, Dr Louise Mason**, Dr Hao***

The talk will focus on healthcare technologies funding opportunities both external to and within Glasgow University. It will highlight the dynamic landscape of funding available, including broader translational and commercialisation options like MRC Catalyst Funding, and Innovate UK which supports transformative research endeavours. Additionally, the talk will provide an overview of funding opportunities within the different Impact Accelerator Accounts (IAA) available here at Glasgow. The presentation will include and IAA case study from Dr Hao Gao (Maths and Stats) will illustrate how these funding avenues have driven innovation in healthcare technologies.

Zoom details: 

Meeting ID: 890 8494 8981

Passcode: 500251

Speaker: Patrizia Di Campi San Vito

The RadioMe project set out to create an adaptive live radio system for  people with dementia with reminder capabilities and agitation detection and mitigation with calming music. It is an interdisciplinary project between the University of Plymouth, the Anglia Ruskin University in Cambridge and the University of Glasgow and includes experts in music computation, dementia, music therapy, machine learning and human computer interaction. In this talk we will set out the motivation and goals of the project and discuss the steps towards implementing our current prototype.

Zoom Meeting ID: 810 8234 8522
Passcode: 237685

Speaker: Narinder Kaur - PhD Student

We will be discussing the following paper: 'Shapley Values for Feature Selection: The Good, the Bad, and the Axioms' 

Abstract:

The Shapley value has become popular in the Explainable AI (XAI) literature, thanks, to a large extent, to a solid theoretical foundation, including four “favourable and fair” axioms for attribution in transferable utility games. The Shapley value is probably the only solution concept satisfying these axioms. In this paper, we introduce the Shapley value and draw attention to its recent uses as a feature selection tool. We call into question this use of the Shapley value, using simple, abstract “toy” counterexamples to illustrate that the axioms may work against the goals of feature selection. From this, we develop a number of insights that are then investigated in concrete simulation settings, with a variety of Shapley value formulations, including SHapley Additive exPlanations (SHAP) and Shapley Additive Global importancE (SAGE). The aim is not to encourage any use of the Shapley value for feature selection, but we aim to clarify various limitations around their current use in the literature. In so doing, we hope to help demystify certain aspects of the Shapley value axioms that are viewed as “favourable”. In particular, we wish to highlight that the favourability of the axioms depends non-trivially on the way in which the Shapley value is appropriated in the XAI application.

This event is organised by the Healthcare Informatics Lab.

Speaker: Workshop

Over the past decade, paradigm shifts have been made towards the way healthcare is delivered and managed. Emerging pervasive sensing technologies, coupled with advanced data analytics, have enabled real-time personalised and environmental monitoring. These technologies have the potential to transform clinical practice in terms of diagnostics, prognosis, preventive healthcare, and rehabilitative/assistive technologies. To enable these systems, we need to understand the challenges in developing AI algorithms to handle heterogeneous, varying quality and imbalanced data. Academics, industry and NHS should work together to develop reliable and trustworthy technologies as well as Deep Learning brain-based algorithms that are able to handle and interpret continuous streams of heterogeneous, multi-modal data.

This workshop aims to bring together academics, industry and NHS related organisations with the goal to highlight opportunities for collaboration. We plan to:

Speaker: Gael Varoquaux

Abstract:

Machine-learning, which can learn to predict given labeled data, bares many promises for medical applications. And yet, experience shows that predictors that looked promising most often fail to bring the expected medical benefits. One reason is that they are evaluated detached from actual usage and medical outcomes.
And yet, test runing predictive models on actual medical decisions can be costly and dangerous. How do we bridge the gap? By improving machine-learning model evaluation. First, the metrics used to measure prediction error must capture as well as possible the cost-benefit tradeoffs of the final usage. Second, the evaluation procedure must really put models to the test: on a representative data sample, and accounting for uncertainty in model evaluation. I will discuss advanced topic on these questions.
For medical applications, predictions should come with associated confidence. It is important to evaluate these confidence with adequate metrics. Here, the difficulty is to control individual probabilities, as each individual is observed only once. I will explain a procedure to measure how far a predictor is from outputing the ideal individual probabilities, due to intrinsic uncertainty [1].
Predictors can be used to reason about possible interventions: for a given individual, what is the potential outcome of an intervention versus no intervention? However, the corresponding inferences require a particular type of control on the error of the predictors [2].
Last but not least, a numerical experiment to benchmark predictors comes with arbritrary sources of variation. Understanding and accounting for this uncontroled variance is important to make well-grounded decisions on which predictive model to use. This is possible with simple procedures [3].


[1] Beyond calibration: estimating the grouping loss of modern neural networks
    Alexandre Perez-Lebel, Marine Le Morvan, Gaël Varoquaux
    ICLR 2023 – The Eleventh International Conference on Learning Representations, May 2023, Kigali, Rwanda
    https://proceedings.mlsys.org/paper/2021/hash/cfecdb276f634854f3ef915e2e980c31-Abstract.html

Biography:

Gael is a research director at the National Institute for Research in Digital Science and Technology (INRIA) at France. He is also the team leader of Soda - Computational and Mathematical Methods to understand health and society with data (https://team.inria.fr/soda/). His research interests encompasses three areas: 

• Machine learning and public health, which involves analytics on health databases for personalized medicine and treatment development, biomedical natural language processing and information extration and causal inference.
• Democratizing machine learning which encompasses machine learning on dirty data Missing data in machine learning, machine-learning model evaluation and learning on relational databases.
• Machine learning for mental health, cognition, and brain activity, which encompasses learning models of brain function and its pathologies from brain imaging, biomarkers of mental traits and disorders which encompasses resting-state and functional connectivity Encoding and decoding models of cognition.

 He is also the director of scikit-learn operations at Inria foundation and core contributor of several open source projects in scientific computing with python.  

Registration is required: https://uofglasgow.zoom.us/meeting/register/tZwrf-ygqjguE9NlUx9-uLqR0IEutLzPDe0K

Speaker: Prof. Crispin Miller

Prof. Crispin Miller is the head of bioinformatics at Cancer Research UK Beatson Insitute. He leads the Computational Biology group, which is focused on using data-driven approaches from machine learning to develop a better understanding of the processes that underpin tumour growth and development. 

A major aspect of his work is the use of cancer ‘omics data generated by large-scale tumour sequencing projects. These datasets are large enough to use machine learning algorithms that seek to correlate patterns with phenotype. This is allowing the team to explore aspects of tumour evolution, and to ask how the regulatory systems that control gene expression are perturbed in tumour cells.

His group is particularly interested in the regulatory pathways that act downstream of transcription, including the processes that govern how alternative splicing is coordinated across different pathways. Other projects in the group focus on uncovering novel regulatory sequences within the genome, and in making use of comparative genomics to help interpret the genome rearrangements that occur in tumour cells.

Registration is required: https://uofglasgow.zoom.us/meeting/register/tZUtf-msqjIoEtQhY63wGchI-Vaj42GueFsS

Speaker: Dr. Tanaya Guha

Abstract: Individuals with Autism spectrum disorder (ASD) have significantly impaired social communication abilities characterised by atypicality in perceiving and producing non-verbal behavioural cues, such as facial expressions and gaze. Such atypicality has been studied mainly through observations by human experts. However, subtle patterns in the behavioural cues may not be easy to capture by visual inspection alone, which underscores the need for objective data analysis and computational methods. This talk will discuss our work on developing computational methodologies to better understand behavioural phenotyping in ASD, particularly looking at facial expressions and gaze. Our goal is to objectively understand and quantify behavioural atypicality in ASD that can inform diagnosis and intervention.

Bio: Tanaya Guha is a Senior Lecturer in the School of Computing Science, University of Glasgow, and an Honorary Associate Professor in the Department of Computer Science, University of Warwick. Her research focuses on developing machine intelligence capabilities to understand human activities and behaviour combining Machine Learning, Computer Vision, and Signal/Speech Processing. She has published over 60 peer-reviewed articles in leading venues in these areas. She serves in the Editorial Board of Scientific Reports, and in the Organizing/Program Committees of several major conferences such as WACV, ICME, ICMI and Interspeech. More information: https://www.tanayag.com

Registration: This event is hybrid and we encourage to attend in person at 422/423 Sir Alwyn Williams Building, School of Computing Science.
Either way, please register your interest here: https://uofglasgow.zoom.us/meeting/register/tZAscuCgrDgvGNEB9piiUZf_mCbyPsEBu4c2
(copy and paste the link text in the browser does not redirect you correctly)

Speaker: George Amarantidis Koronaios

Abstract: 

Semantic segmentation is a common problem in medical imaging. In this seminar, we will examine how MATLAB Tools (medical imaging toolbox and deep learning toolbox) can streamline this process. We will look into the tools available at your disposal and do a deeper dive into an example of semantic segmentation cardiac MRI data. This example will cover streamlining data labelling, transfer learning, and dealing with label imbalance.

Venue: 422 Sir Alywn Williams Building (Join us in person to discuss internship/job opportunities at Mathworks)
(Hybrid Format- https://uofglasgow.zoom.us/meeting/register/tZAsfumpqjkqHdBuQcvARMF5brkCgoR_cLjx)

Biography:

George Koronaios works with mathworks team to support academics with research and teaching. Recently MathWorks released the Medical Imaging Toolbox to visualise, register, segment and label 2D and 3D medical images. This presentation will provide few examples with a focus on semantic segmentation and labeling. Medical Imaging Toolbox™ provides apps, functions, and workflows for designing and testing diagnostic imaging applications. You can perform 3D rendering and visualization, multimodal registration, and segmentation and labeling of radiology images. The toolbox also lets you train predefined deep learning networks (with Deep Learning Toolbox™).

Speaker: Prof. Nasir M. Rajpoot

Abstract:

• Large collections of pathology image data offer a potential goldmine of invaluable information, ripe for deep learning of known and deep mining of novel digital histological biomarkers of cancer diagnosis, prognosis, clinical outcome and response to therapy.
• This talk will cover some of the major challenges faced and opportunities offered by the nascent discipline of computational pathology.

Biography:

Nasir Rajpoot is Professor of Computational Pathology at the University of Warwick and Honorary Scientist at the Department of Pathology, University Hospitals Coventry & Warwickshire (UHCW) NHS Trust. Prior to completing his PhD in Computer Science from Warwick in 2001, he was a Postgraduate Research Fellow in the Applied Math program (partially based at the School of Medicine) at Yale University (USA) during 1998-2000 and a Systems Engineering Fellow at PIEAS (Pakistan) during 1994-1996.

Prof Rajpoot is the founding Director of Tissue Image Analytics (TIA) Centre (previously the TIA lab) at Warwick since 2012 and also co-Director of the recently funded £15m PathLAKE centre of excellence on AI in pathology since Jan 2019. The focus of current research in TIA Centre led by Prof Rajpoot is on AI and machine learning algorithms for the study of histological and multi-omic markers of cancer biology, with applications to early detection of cancer and stratification of cancer patients in terms of recurrence, progression and response to therapy. He has been active in the digital pathology community for almost two decades now and has delivered over 80 invited and keynote talks since 2015 at various national and international events and institutions.

Prof Rajpoot recently served as President of the European Congress on Digital Pathology (ECDP), which took place at Warwick in April 2019. Previously, he served as the General Chair of the UK Medical Image Understanding and Analysis (MIUA) conference in 2010 and as the Technical Chair of the British Machine Vision Conference (BMVC) in 2007. He co-chaired several meetings in the histology image analysis (HIMA) series since 2008 and served as a founding PC member of the SPIE Digital Pathology meeting since 2012. He is a Senior Member of IEEE and member of the Association of the Computing Machinery (ACM), the British Association of Cancer Research (BACR), the European Association of Cancer Research (EACR) and the American Society of Clinical Oncology (ASCO).

Prof Rajpoot was recently awarded the Turing Fellowship by the Alan Turing Institute, the UK's national data science institute.

Speaker: Dr Simon Rogers

NHS NSS is a specialist health board, providing services to the other health boards and across the public sector. Within NSS, the Business Intelligence (and more specifically, the Artificial Intelligence Centre of Excellence) provide assistance, expertise, and a technology platform to support projects with an AI component. In this talk, I will introduce NSS BI (including our Seer platform), and discuss some of the work that we are undertaking, including work on the new version of the SPARRA model (predicting admission / re-admission for individuals in Scotland), work on how to safely export trained models from safe-haven environments, and some ongoing work on fairness / inequalities.

Biography:

Dr Simon Rogers is an AI data scientist at NHS National Services Scotland and Honorary Lecturer in Computer Science at University of Glasgow. His research involves the development of Machine Learning and Statistical techniques to help with the analysis of complex datasets. Among his long publication list he has also published a textbook on 'A First Course in Machine Learning', which is an introduction to modern (fairly probabilistic) Machine Learning.

This is a hybrid seminar: Join us at 423 Sir Alwyn Williams Building, School of Computing Science.

Registration is required:  https://uofglasgow.zoom.us/meeting/register/tJUkcOqsqz8vHdboNnP6P-_L3F5na54lHlsx

Speaker: Dr. Simone Stumpf

Abstract:

AI is touted as the panacea that will cure all the ills of the current healthcare system. In this talk I will provide an overview of my work in technologies for healthcare, well-being and accessibility that will draw out lessons learned and challenges for designing and developing effective healthcare technologies that use AI.

Biography:

Dr. Simone Stumpf recently joined University of Glasgow, UK, as a Reader in Responsible and Interactive AI. She has a long-standing research focus on user interactions with machine learning systems. Her research includes self-management systems for people living with long-term conditions, developing teachable object recognisers for people who are blind or low vision, and investigating AI fairness. Her work has contributed to shaping the field of Explainable AI (XAI) through the Explanatory Debugging approach for interactive machine learning, providing design principles for enabling better human-computer interaction and investigating the effects of greater transparency. The prime aim of her work is to empower everyone to use AI systems effectively.

Speaker: Xiao Guo

Abstract:

In this talk Xiao Gu will talk about his recently ECCV accepted paper on 'Tackling long-tailed category distribution under domain shifts' and the related CVPR 2022 challenge. 

Machine learning models fail to perform well on real-world applications when 1) the category distribution P(Y) of the training dataset suffers from long-tailed distribution and 2) the test data is drawn from different conditional distributions P(X|Y). Existing approaches cannot handle the scenario where both issues exist, which however is common for real-world applications. In this study, we took a step forward and looked into the problem of long-tailed classification under domain shifts. By taking both the categorical distribution bias and conditional distribution shifts into account, we designed three novel core functional blocks including Distribution Calibrated Classification Loss, Visual-Semantic Mapping and Semantic-Similarity Guided Augmentation. Furthermore, we adopted a meta-learning framework which integrates the three blocks to improve domain generalization on unseen target domains. Two new datasets were proposed for long-tailed classification under domain shifts, named AWA2-LTS and ImageNet-LTS. We evaluated our method on the two datasets and extensive experimental results demonstrate that our proposed method can achieve superior performance over state-of-the-art long-tailed/domain generalization approaches and the combinations.

Biography: 

Xiao is a PhD student at the Hamlyn Centre, Imperial College London, UK. He obtained a BEng degree (Honor Student) in Electronic Engineering from Fudan University in 2018 and MRes degree in Medical Robotics and Image Guided Intervention (Distinction & Dean's Prize) from Imperial College London in 2019. My research interests focus on integrating pervasive sensing and deep learning for healthcare. More details can be found at his website: https://xiaogu.site/ 

Speaker: Dr Xianghua Ding & Dr Marwa Mahmoud

This is a special hybrid event that hosts Dr Marwa Mahmoud that they are going to talk about their research on 'Advance Technology, Empower People: Exploring Social-Technical Approaches for Everyday Health' and 'Vision-focused multimodal behaviour modelling for mental health applications'

In person location: 422-423 Sir Alwyn Williams Building, School of Computing Science, UoG

Online location: https://uofglasgow.zoom.us/meeting/register/tJUucOCrpjovHdLqvVrzDE4XWTHqH9BtT5uP 

Title: Vision-focused multimodal behaviour modelling for mental health applications

Abstract:

There is a growing interest from healthcare organisations, academia and industry on automatic prediction, prevention and intervention of mental health issues, but most of the current work depends  on non-visual input, such as wearables and mobile phone data or basic analysis of video focussing on facial expressions analysis only. These models do not capture the full picture and overlook complex non-verbal behaviour analysis, which is the basis of many mental disorder diagnoses. Computer vision techniques have not been fully explored in this domain because of the sensitive nature of the data and limited availability of public datasets for training. Moreover, the high dimensionality of the signals collected from video complicates processing.

In this talk, I will present my work on automatic analysis and detection of body gestures and self-adaptors that are related to psychological distress. I will also discuss multimodal behaviour modelling techniques that I adapted to overcome the challenges of data scarcity and noisy complex signals in mental health datasets. Finally, I will also present some of my recent work on modelling gestures and body expressions in infants shedding light on their relationship with their neurodevelopment assessment.

Biography:

Dr Marwa Mahmoud is a Lecturer in Socially Intelligent Technologies in the School of Computing Science at University of Glasgow, and a Visiting Fellow in the Department of Computer Science and Technology at University of Cambridge, UK. Before joining University of Glasgow, she spent 10 years at University of Cambridge, where she obtained her PhD in 2015, then worked as a postdoc for a year before managing to secure the prestigious King’s College Junior Research Fellowship in 2016, which allowed her to start independent research. Her research interests focus on computer vision for social signal processing and multimodal signal processing, especially within the context of affective computing, behaviour analytics, human behaviour understanding and animal behaviour understanding. She applied her research in the areas of automotive applications, mental healthcare, and animal welfare. She is interested in ‘AI for Social Good’, combining computer vision research with health for human well-being and animal welfare applications. She is an elected Executive Committee (EC) Member of the Association for the Advancement of Affective Computing (AAAC), Network Member of Cambridge Trust & Technology Initiative and a member of Cambridge Neuroscience.

Title: Advance Technology, Empower People: Exploring Social-Technical Approaches for Everyday Health

Abstract:

Biography:
Dr. Xianghua (Sharon) Ding is a Senior Lecturer in Healthcare Technologies in the School of Computer Science at the University of Glasgow. Before moving to Glasgow, she was an Associate Professor at Fudan University, Shanghai, China. She received her Ph.D. from University of California, Irvine, USA in 2010. Her research falls in the areas of Human Computer Interaction(HCI), Computer Supported Cooperative Work (CSCW), and Ubiquitous Computing (Ubicomp). She is interested in social-technical approaches to empower people to manage their health in daily lives, with particular interest in intelligent health sensing and collaborative technologies for preventive health, health literacy, and mental wellbeing. Her work has been published at flagship venues in HCI such as CHI, CSCW, and Ubicomp, and has received Best Paper (CSCW2015) and Honorable Mention (CHI2020) awards. She has also been serving leadership roles in these venues, including Editor for CSCW2021-2022, Subcommittee Chair for CHI2021’s Health and CHI2019’s Understanding People, and Associate Chair for CHI and CSCW. She also led the publication on a special issue on human-centered cooperative computing as a lead guest editor for CCF’s Transactions on Pervasive Computing and Interaction.

Speaker: Prof. Julian Fierrez

 In the last decade we have witnessed extraordinary advances in technologies aimed at managing and securing our identity for online and mobile applications including e-commerce, e-health, e-banking, e-learning, and others. This has been enabled by a rapidly evolving mobile market, with smartphones now having impressive computation capabilities and many audiovisual and biometric-specific sensors capable of acquiring high-quality face images, voice, fingerprints, and other biometric information. This context has nurtured the development of biometric technology that can represent in a very distinctive way our individual identity. These advances in biometric identity have come, at the same time, with growing risks. The information secured by biometric models in our smartphones and other computing platforms are nowadays more valuable than ever, and therefore a growing number of attacks are being conducted seeking big returns by faking our biometric identities. This keynote will summarize the main advances in biometric security conducted in the last decade aimed at evaluating the security of biometric systems against presentation attacks (aka spoofing attacks). Additionally, and as a natural evolution of the mentioned security aspects of biometric identities, we will discuss a topic of growing interest nowadays: facial image manipulation techniques, including DeepFakes, and ways to detect such manipulations. These manipulation techniques, boosted by recent advances in deep learning, can nowadays be used for biometric impersonation in very harmful ways. Both lines of research (biometric anti-spoofing and facial manipulation detection) in a sense have the same purpose: in an interconnected world where we interact more than ever via digital representations of ourselves including in many cases biometric information, there is a growing need for securing our digital identities against a growing army of attackers and attacking methods.

Biography: Julian FIERREZ received the MSc and the PhD degrees in telecommunications engineering from Universidad Politecnica de Madrid, Spain, in 2001 and 2006, respectively. Since 2002 he was affiliated as a PhD candidate with the Universidad Politecnica de Madrid, and since 2004 at Universidad Autonoma de Madrid, where he is currently an Associate Professor since 2010. From 2007 to 2009 he was a visiting researcher at Michigan State University in USA under a Marie Curie fellowship. ELLIS Society.

Prof. Fierrez has received best papers research distinctions, including: EBF European Biometric Industry Award 2006, EURASIP Best PhD Award 2012, Medal in the Young Researcher Awards 2015 by the Spanish Royal Academy of Engineering, and the Miguel Catalan Award to the Best Researcher under 40 in the Community of Madrid in the general area of Science and Technology. In 2017 he has been also awarded the IAPR Young Biometrics Investigator Award, given to a single researcher worldwide every two years under the age of 40, whose research work has had a major impact in biometrics.

Registration is required: https://uofglasgow.zoom.us/meeting/register/tJIpcu2pqDgoEtfbxu-X9VKJ-GMZ-8EXD93x

Subscribe to the computing technologies healthcare seminar emailing list here: https://samoa.dcs.gla.ac.uk/events/series.jsp?series=179

Speaker: Prof. Fahim Kawsar

Abstract: Do you know that someone has a heart attack by the time you finish reading this abstract? Indeed, every 40 seconds, a heart attack occurs, and a heart attack fatality occurs about every minute. For long, medical science has established that frequent monitoring of heart rate and blood pressure is the key to mitigating significant risks for stroke, heart failure and coronary artery disease. Current gold-standard vital sign monitoring devices are invasive, cause discomfort and interfere with users' activities. We challenge this status quo and ask - can your next earable be the secret weapon to manage your cardiovascular health? We aren't talking PopSci here but the remarkable capabilities of a sensory earable ushering a new pathway to a healthy heart. However, ensuring the accuracy and robustness of in-ear vital sign measurements is the most complex challenge in this vision. This talk will reflect on the algorithms and their systematic characterisations to address this challenge in continuously measuring the five most critical vital signs in our ear - Heart Rate (and Heart Rate Variability), SPO2, Respiration Rate, Temperature and Blood Pressure.

Biography: Fahim Kawsar leads Pervasive Systems research at Nokia Bell Labs, Cambridge and holds a Mobile Systems Professorship in Computing Science at Glasgow. At his Cambridge lab, he studies forms and intelligence of multi-sensory devices to learn, infer and augment human behaviour in three application areas - Digital Health, Quantified Lifestyle and Smart Built Environment. At Glasgow, he is building up a new group to study system and algorithmic challenges for Mobile Systems. Fahim's research has led to developing multiple artefacts for Nokia (e.g., Consumer Wearables, Sensing-as-a-Service, and Camera Analytics). He has (co-)authored 160+ research papers (Six Best Paper Awards) and granted 40+ patents (Two Top Nokia Innovator Awards). Fahim is a frequent keynote speaker across academic and industrial forums, is an AEIC of IEEE Pervasive Computing, sits at the editorial board of ACM IMWUT, serves (d) as a committee member of leading mobile computing conferences, received multiple million-scale research funds, and has a career accumulation of over €5 Million.

Registration is required:  https://uofglasgow.zoom.us/meeting/register/tJEvcOiqrjMrEtRUDgwKSAWB3aTbhFha5G1x 

Hybrid Format: Room 422-423 Sir Alwyn Williams Building.

Consider subscribing to our emailing list: https://samoa.dcs.gla.ac.uk/events/series.jsp?series=179 

Speaker: Dr. Stamatia Giannarou

This is a special session organised on behalf of the Computing Technologies for Healthcare Theme and the Athena Swan. The session includes:

• Research talk and Q/A (60 minutes)
• Discussion on career progression and fellowships (45 minutes) 

Registration is required:  https://uofglasgow.zoom.us/meeting/register/tJwsduqhrDwiHtdMKCADLEHgdNagRGveehZr

Consider subscribing to our emailing list: https://samoa.dcs.gla.ac.uk/events/series.jsp?series=179 

Abstract: With recent advances in medical imaging and surgical robotics, surgical oncology is entering a new era that is set to bring major healthcare and socio-economic benefits. The main goal of surgical oncology is to achieve complete resection of cancerous tissue with minimal iatrogenic injury to surrounding tissue. In practice, this often presents a formidable challenge to surgeons. Surgery on tumours residing within the brain is particularly demanding, and the prognosis for patients afflicted with such tumours remains very poor. Intrinsic brain tumours are highly infiltrative making it difficult to distinguish tumour tissue from surrounding tissue. Moreover, it is imperative to preserve unaffected brain tissue, which is delicate, often eloquent, and has little capacity for regeneration.

The aim of my research is to integrate multimodal intraoperative imaging and navigation technologies into a cognitive robotic platform. In this talk, I will present an intraoperative vision system for surgical navigation and real-time tissue characterisation during robot-assisted neurosurgery to improve both the efficacy and safety of tumour resections. The focus will be on the recovery of 3D morphological structures in the presence of tissue deformation, the efficient robot-assisted tissue scanning with imaging probes and the tissue characterisation for on-line diagnosis support.

Biosketch: Stamatia (Matina) Giannarou received the MEng degree in Electrical and Computer Engineering from Democritus University of Thrace, Greece in 2003, the MSc degree in communications and signal processing and the Ph.D. degree in image processing from the department of Electrical and Electronic Engineering, Imperial College London, UK in 2004 and 2008, respectively. Currently she is a Royal Society University Research Fellow and a Lecturer in Surgical Cancer Technology and Imaging at the Hamlyn Centre for Robotic Surgery, Department of Surgery and Cancer, Imperial College London, UK. Her research focuses on enhanced surgical vision for intraoperative navigation in minimally invasive and robot-assisted operations. In 2017, she won “The President’s Award for Outstanding Early Career Researcher” at Imperial College London. She has been selected as a member of the IdeasLab of Imperial College London on the “Frontiers of Imaging” at the World Economic Forum Annual Meeting of the New Champions 2016 in Tianjin, China. She received best paper awards at international conferences and workshops including the IPCAI 2016, AE-CAI-MICCAI 2020, IPCAI 2020, AE-CAI-MICCAI 2021. She has also been invited to present her work at a number of international workshops and symposia. She is a regular reviewer for high impact journals and conferences in the fields of medical robotics, medical imaging and biomedical engineering and the chair of the annual Hamlyn Winter School on Surgical Imaging and Vision.

Speaker: Dr. Karim Lekadir

Abstract: Amid hope and hype, artificial intelligence (AI) is widely regarded as one of the most promising and disruptive technologies for future healthcare. The application of medical AI has the potential to increase the productivity and efficiency of clinicians, improve medical diagnosis and treatment, optimise the allocation of human and technical resources, and lead to better health outcomes for patients and citizens. However, there are currently several technical, clinical, ethical and legal risks associated with medical AI that have limited its deployment in the real world. This talk will discuss these risks, including the potential lack of clinical safety, the limited generalisability of the AI solutions across settings, and important ethical issues such as algorithmic bias against under-represented groups. I will also discuss the need for guidelines and best practices to support the design, development and deployment of future AI tools in healthcare that are accurate and robust, but also trustworthy and ethical, to maximise their acceptance and adoption by medical professionals and patients alike. 

Biography:  Dr. Karim Lekadir is a Ramon y Cajal Researcher and Director of the Artificial Intelligence in Medicine Lab at the Universitat de Barcelona (BCN-AIM). He holds a PhD from Imperial College London (UK) and was previously a visiting scholar at Stanford University (USA). His current research focuses on the development of data science and machine learning approaches for the analysis of large-scale biomedical data, including imaging, clinical, lifestyle, and mobile data. The software he developed during his PhD for cardiac functional quantification has been CE marked and commercialised by CMRtools, and is now used in more than 250 clinical centres worldwide. He is the Coordinator of the following Horizon 2020 projects: euCanSHare (2018-2022), developing a big data platform for cardiovascular research; EarlyCause (2019-2023), which investigates multi-morbidity using experimental and data science approaches; and EuCanImage (2020-2024), which is building a federated artificial intelligence environment for cancer imaging. He is also work package leader in the longITools H2020 project  (2019-2024), developing a mobile app for cardio-metabolic risk prediction based on exposome data. In addition, Karim is General Chair for the MICCAI 2024 Conference (Medical Image Computing and Computer-Assited Intervention) which for the first time will take place in Africa – in Marrakesh, Morocco. He is an Associate Editor of IEEE Transactions on Medical Imaging.

Speaker: Prof. Hatice Gunes

This is a special session organised on behalf of the Computing Technologies for Healthcare Theme and the Athena Swan. The session includes:

• Research talk and Q/A (60 minutes)
• Discussion on career progression and fellowships (30 minutes) 

Registration is required: https://uofglasgow.zoom.us/meeting/register/tJ0od-msqTotHdMlo7hMP2WxCs8jkbNj6J00 

Short summary of research talk: Designing artificially intelligent interfaces and robots with socio-emotional skills is a challenging task. Progress in industry and developments in academia provide us a positive outlook, however, the artificial emotional intelligence of the current technology is still limited. In this talk, I will present some of our research explorations in this area with applications to well-being, specifically in virtual reality, in work-like settings, and with/for robotic mental well-being coaching.

Biography: Hatice Gunes is a Professor of Affective Intelligence and Robotics (AFAR) and the Head of the AFAR Lab at the University of Cambridge's Department of Computer Science and Technology. Her expertise is in the areas of affective computing and social signal processing cross-fertilising research in multimodal interaction, computer vision, signal processing, machine learning and social robotics. She has published over 125 papers in these areas (h-index=35, citations > 6,300),  with  most  recent  works  on lifelong learning for facial expression recognition, fairness and affective  robotics;  and  longitudinal  HRI  for  wellbeing.  Some of her research highlights  include  RSJ/KROS  Distinguished  Interdisciplinary Research Award Finalist at IEEE RO-MAN’21, Distinguished PC  Award  at  IJCAI’21,  Best  Paper  Award  Finalist  at  IEEE RO-MAN’20, Finalist for the 2018 Frontiers Spotlight Award, Outstanding  Paper  Award  at  IEEE  FG’11,  and  Best  Demo Award at IEEE ACII’09. Prof Gunes is the former President of the Association for the Advancement of Affective Computing (AAAC), and was the General Co-Chair of ACII’19, and the Program Co-Chair of ACM/IEEE HRI’20 and IEEE FG’17. She was a member of the Human-Robot Interaction Steering Committee (2018-2021) and was the Chair of the Steering Board of IEEE Transactions on Affective Computing (2017-2019). In 2019 she was awarded the prestigious EPSRC Fellowship as a personal grant (2019-2024) to investigate adaptive robotic emotional intelligence for well-being, and was named a Faculty Fellow of the Alan Turing Institute– UK’s national centre for data science and artificial intelligence (2019-2021). Prof Gunes is a Senior Member of the IEEE and a member of the AAAC.

Speaker: Andrew Simpson

Computer Science career in the NHS: The Scientist Training Programme (STP)

Andrew Simpson will join us on Zoom on the 25th of January at 6:30pm to talk about the Scientist Training Programme in the NHS. The STP is a three year fully NHS funded training programme, leading to an Clinical Scientist. During training, the student is paid at NHS Band 6 (at a training Annex). The entry requirements are: An undergraduate degree in either Computer Science, Mathematics or Scientific Engineering.

Computer Scientists who are Clinical Scientists can and do: 

• Develop medical software and technology
• Use that software/technology clinically on patients (in theatres, on wards, etc.)  (Under Annex A of the Medical Device Regulations) 
• Understand and be involved in the research of the science behind the diagnostic test
• Advise doctors on the results and on the technology available for patients.

All as a state registered healthcare professional in a truly unique, in demand and rewarding role.

Higher Specialist Scientific Programme (HSST)

As your career progresses, Clinical Scientists are eligible to further train as Consultant Clinical Scientists in Clinical and Scientific Computing, to demonstrate significant expertise at the intersection of Computer Science and Healthcare.

STP Equivalence
This is portfolio route by the Academy of Healthcare Sciences, where the pre-registrant demonstrates equivalency to the Scientist Training Programme outcomes. A master's degree is not required although the pre-registrant must demonstrate working to master's level.

Route II
This route allows a work-based competency learning route for those who have a masters and relevant experience. The pre-registrant works towards and compiles a portfolio of evidence, demonstrating competency in Clinical Computing to the level of Clinical Scientist.

About Andrew:

He studied BSc & MSc Computer Science, worked in IT industry for a while and he is now training as a clinical scientist in physiological measurement & clinical computing, under Route II, in the James Cook University Hospital's Medical Physics Department.

(For more information email Dr. Tim Storer: timothy.storer@glasgow.ac.uk)

Speaker: Prof. Frederick Matsen

Phylogenetic (evolutionary tree) inference is a key tool for understanding evolutionary systems. This includes viral adaptation and genomic epidemiology, as well as the antibody response to infection and vaccination. Bayesian phylogenetic analysis allows us to assess and integrate out tree uncertainty to obtain more reliable estimates of other model variables of interest (e.g. transmission rates). However, Bayesian posterior distributions on phylogenetic trees remain difficult to sample despite decades of effort. The complex discrete and continuous model structure of trees means that recent inferential methods developed for Euclidean space are not easily applicable to the phylogenetic case. Thus, we are left with random-walk Markov Chain Monte Carlo (MCMC) with uninformed tree modification proposals; these traverse tree space slowly because phylogenetic posteriors are concentrated on a small fraction of the very many possible trees.

In this talk, I will give a relatively non-technical overview of the work we have done to enable Bayesian phylogenetic inference via optimization. This work has led to a new discrete inferential target, which we call the "subsplit directed acyclic graph," and a new algorithm that will allow us to infer this structure using methods analogous to much faster maximum-likelihood (point-estimate) methods for phylogenetics. I will also describe how, once this structure is in hand, we can perform variational inference for continuous parameters via stochastic gradient descent. 

Biography:

Dr. Frederick “Erick” Matsen is an expert in computational biology, which is the science of using biological data to develop computer algorithms, or programs, to understand biological systems and relationships. His research team has developed new methods to analyze data generated by sequencing the DNA of viruses, immune cells and complex environmental samples containing many microorganisms. The team also pursues more abstract questions about the methods used to construct evolutionary trees. Another focus of Dr. Matsen’s work is on improving software used in computational biology, both by developing open source tools and by contributing to work on larger, collaborative projects.

Registration: https://uofglasgow.zoom.us/meeting/register/tJYsf-2tpjkqHte7oAoHA8svaCuDku4j0eJ8  

Speaker: Prof. Nazli Goharian

Abstract:

With the ever-increasing usage of social media for either explicitly seeking help or for simply sharing thoughts and feelings, we, in the computational disciplines, have the opportunity to utilize such data for building datasets, models, and doing analysis.  In the first part of my talk, I will share our collaborative work done at the Information Retrieval Lab at Georgetown University on detecting mental health concerns on social media posts. The first application is on a dedicated mental health forum where the users who register to share and communicate their thoughts and feelings are suffering from some sort of mental distress (sadness, depression, potential of self-harm….). The task is to triage the severity of users’ posts to detect early the potential of self-harm as well as to evaluate the impact of forum activities and conversations on the users during a period of time. In the second type of platform, i.e., non-dedicated, I focus on the question of whether we can detect if a user is suffering from any one or more of the nine mental health conditions, only using the *general language* of the user; that is, the posts are not in mental health [sub]forums nor have any mental health related words. For addressing this question, we had to construct large scale datasets; I will explain how we have identified the diagnosed users, and how carefully selected the controls.  Further I will show the results of several baselines to detect the conditions. To conclude this segment of the talk, I will present our preliminary efforts in summarizing mental health posts.

In the second part of the talk, I switch to physical health, and more specifically, to our efforts to address the discrepancies in the radiology reports. Each radiology report is written by residents and reviewed by the attending physician. The question we worked on was how we can detect significant discrepancies in these reports. This system is in use and is made available to the residents in training.

I will conclude with glimpses of additional health related projects of our lab.

Biography

Nazli Goharian is Clinical Professor of Computer Science and Associate Director of the Information Retrieval Lab at Georgetown University, which she co-founded in 2010. She joined the Illinois Institute of Technology (IIT) from industry in 2000. Her research and doctoral student mentorship span the domains of information retrieval, text mining, and natural language processing. Specifically, her interest lies in humane-computing applications such as medical/health domain. Joint with her doctoral students, she received an EMNLP 2017 Best Long Paper Award and COLING 2018 Honorable Mention both for papers on mental health and social media.  For contributions to undergraduate and graduate curriculum development and teaching excellence, she was recognized with the IIT Julia Beveridge Award for faculty (university-wide female faculty of the year) in 2009, the College of Science and Letters Dean’s Excellence Award in Teaching in 2005, and in 2002, 2003, and 2007, the Computer Science Department Teacher of the Year Award.  She served as Senior/Area Chairs at ACL 2018, ACL 2019, ACL 2020 and ACL 2021.  She is co-chair of SIGIR Women in Information retrieval (WIR) since 2019, focusing on gender pay inequity and women leadership.

Registration is required to attend:  https://uofglasgow.zoom.us/meeting/register/tJMtdOGtrT4iE9aX2cDxzQ6TbisiGWEn7Y-M

Speaker: Dr. Alison ONeil

Radiologists are under pressure to handle an ever-increasing volume of medical imaging. In the meantime, deep learning solutions have demonstrated impressive performance for imaging tasks of classification, segmentation and translation, opening up opportunities for AI to assist and automate imaging workflows. Training deep learning solutions for real-world healthcare applications involves ethical, legal and practical performance considerations. As we look to scale deep learning to a wide range of clinical applications, we also need to find ways to reduce reliance on pixel-level expert annotations whilst retaining clinically acceptable accuracy. This talk will cover some of Canon’s research in this area.

Biography:

r Alison Q O’Neil is a Principal Scientist in the AI Research Team at Canon Medical Research Europe and Honorary Research Fellow at the University of Edinburgh. She leads an AI Research team at Canon who focus on imaging and natural language processing for healthcare problems. Her research interests span techniques for learning with less, multimodal learning, knowledge informed machine learning, and causality.

 Registration is required to attend the event: https://uofglasgow.zoom.us/meeting/register/tJUuf-2prD0sG91BDJPMlW9HKqBz6itv8CgP 

Speaker: Dr. Jennifer Stinson

 Dr. Stinson leads the iOUCH lab (improving outcomes in child health using digital health technologies) in The Research Institute at the Hospital for Sick Children in Toronto.  Her program of research is focused on improving the way healthcare is delivered to children with life-threatening and chronic illnesses and their families in Canada using the latest in information and communications technologies (ICTs). More specifically, her research focuses on the use of ICTs to (a) assess and manage disease-related pain and other symptoms and (b) deliver innovative disease self-management and transitional care programs to these at-risk populations. During this talk she will describe the user-centred design approach she employs to build virtual care self-management pain solutions, discuss research on the impact of CAVID-19 pandemic on youth with chronic pain in Canada and the development of the Power Over Pain Portal to improve equity and access to pain care in Canada. Finally, she will discuss key lessons learned regarding virtual care for clinical and research.

Lab website - https://lab.research.sickkids.ca/iouch/

Biography:

Dr. Jennifer Stinson is the inaugural Mary Jo Haddad Nursing Chair in Child Health and a Senior Scientist in the Child Health Evaluative Sciences (CHES) research program in the Research Institute at The Hospital for Sick Children (SickKids) in Toronto. She is also an Advanced Practice Nurse in the Chronic Pain Program in the Department of Anesthesia and Pain Medicine at SickKids and a Professor in the Lawrence S. Bloomberg Faculty of Nursing, the Institute of Health Policy, Management and Evaluation and Institute of Medical Science at the University of Toronto.

Registration is required: 

Speaker: Dr. Rajeev.Krishnadas

Biography

Dr Rajeev Krishnadas is a visiting consultant at the Priory Hospital Glasgow. He is also a consultant psychiatrist at NHS Greater Glasgow and Clyde and an Honorary Senior Clinical Lecturer at the University of Glasgow. He also sees adults with a wide range of mental health problems, like depression, psychosis, anxiety and bipolar disorder. His particular area of interest is diagnosis and treatment of adults with ADHD. 

Dr Krishnadas completed his basic medical training and psychiatric training from Bombay University, India. He then worked as a consultant psychiatrist in India before moving to the UK, bringing extensive experience in managing patients who speak Malayalam, Tamil and Hindi. He then completed his advanced training in general adult psychiatry along with a PhD in psychological medicine, from the University of Glasgow. 

To attend you need to register here:
https://uofglasgow.zoom.us/meeting/register/tJMsdO6tpzksHN0ai5MzpOoM6UtEn3ISk3Qr 

Speaker: https://socialcdt.org/aimh/

The World Health Organisation states that mental health problems costs the global economy one trillion US dollars each year, whereas in UK, mental health services report that 19.7% of people over 16 years old show symptoms of anxiety and depression. The latest interactive technologies of social robots and virtual reality, powered by Artificial Intelligence (AI) promise new approaches for clinical treatment of psychiatric, developmental and cognitive problems. This workshop brings together experts in AI, computing and mental health, from both academia and industry, that work towards the development of new, computing based methodologies for detection, treatment and analysis of mental health issues. Besides providing an extensive overview of the state-of-the-art in the domain, the workshop aims at helping young researchers (including PhD students and postdocs) and raise awareness of multidisciplinary opportunities of computing and medical sciences in mental health.

Speakers:

• Dr. Nicholas Cummins
  Title: Speech analysis for mental health: opportunities and challenges
• Prof. Helen Minnis
  Title: Social Signal Processing in Attachment
• Prof. Mohamed Chetouani
  Title: Putting the social interaction at the center of mental health analysis and modeling with AI-based approaches
  
• Dr. Esther Papies
  Title: The role of habits in health behaviour: Challenges and opportunities for digital interventions
  
• Dr Pietro Cipresso
  Title: Computational Psychometrics integrating artificial intelligence, virtual reality and mental health

Organised by Computing Technologies for HealthCare and Social AI CDT. For more information and registration details visit the official website.

Speaker: Prof Robert Insall, Dr Breda Cullen, Dr Richard Reeve, Dr Mayetri Gupta

Organised by staff and PhD students, the First University of Glasgow Computational Biology Conferences aims to bring together researchers across the University of Glasgow who are interested or work in the field of computational biology. Topics of interest including but not limited to: computational approaches in omics, systems biology, population genetics, medical imaging, public health and clinical data. The conference will take place in the afternoons of 27th and 28th of May. We are pleased to confirm our keynote speakers: Prof Robert Insall, Dr Breda Cullen, Dr Richard Reeve, and Dr Mayetri Gupta. 

We welcome abstract submissions from early career researchers and PhD students. Abstract deadline is April 30th, 2021. Abstract submission form available on our website: https://cbc.dcs.gla.ac.uk/

Register to attend here. 

If you have any questions, please do not hesitate to drop us a message: 

Speaker: Dr Kitty Meeks

Abstract:

Which characteristics should be used to determine whether a patient is offered routine cancer screening? Are there two or three qualitatively different types of heart failure? Which journeys should be forbidden to restrict the spread of an infectious disease?
Data-driven approaches to answering any of these questions - as well as many others in the field of digital health - typically involve searching for a single mathematical object which is optimal with respect to some criterion. For example, we might aim to partition patients into a fixed number of groups or "clusters" in a way that minimises the maximum "difference" in the characteristics of patients assigned to the same cluster.
However, there will often be many solutions that are equally good with respect to our chosen criterion, in which case it is misleading to consider just a single example: if there are many optimal ways to split our patient group into clusters, and there is little agreement between these optimal solutions about which patients belong to the same cluster, then we should not draw conclusions based on just a single optimal solution. It is therefore important to find out more about the whole set of good solutions.
Unfortunately, in most settings, even finding a single optimal solution is a very computationally challenging problem, and finding all good solutions (or even estimating how many of these there are) is even more difficult.

This talk will contain many more questions than answers.  In October I will start a 5-year EPSRC fellowship in which I will be developing new theoretical machinery for solving counting, enumeration and sampling problems - with provably efficient and correct algorithms - and exploring the applications of these algorithms to digital health.  In this talk I will give a high-level introduction to my proposed methodology, as well as discussing several specific healthcare challenges where I believe this approach can offer new insights.  The talk will also act as an open call for collaboration: I would be delighted to hear from others about further potential applications, or opportunities to compare my new approaches to existing techniques.

Biography

As a child, she dreamed of becoming a professional cake decorator. However, it soon became clear that her future was more likely to lie in mathematics. She went on to obtain an MMath in Mathematics and Computer Science at the University of Oxford (Merton College) in 2009; She stayed in Oxford, working under the supervision of Computational Counting. She came to the University of Glasgow in 2014, where she held a lectureship in the School of Mathematics and Statistics for two years. In October 2016 she moved to the School of Computing Science, where she currently hold a five-year Royal Society of Edinburgh Personal Research Fellowship (funded by the Scottish Government).

Speaker: Prof. Kazem Rahimi / Dr. Mohammad Hossein Mamouei

Abstract: This talk will first provide an overview of the claimed expectations from machine learning and AI in medicine. It criticises the misconceptions around personalised medicine in the context of multi-cause conditions such as cardiovascular diseases. It then provides a general overview of the machine learning research at Deep Medicine with examples of how this is guiding research and could lead to better decision making in clinical practice. Examples will primarily focus on the application of a range of machine learning models to complex Electronic Health Records for risk prediction, co-morbidity and patient clustering, risk factor identification, causal inference and model interpretability. 

Kazem Rahimi is a Professor of Cardiovascular Medicine and Population Health, at the University of Oxford and a consultant cardiologist at the Oxford University Hospitals NHS Trust. His research interests include hypertension, heart failure, multimorbidity and cardiovascular risk management, using a variety of methodologies such as individual-patient meta-analysis, large-scale decentralised clinical trials, and digital health technologies.

Kazem leads the Deep Medicine programme at the Nuffield Department of Women’s and Reproductive Health with a major interest in the application of machine learning approaches to electronic health records. He also leadsPEAK-Urban programme.
----

Dr. Mohammadhossein Mamouei is a Machine Learning Scientist at the University of Oxford. As a member of the interdisciplinary research group, Deep Medicine led by Prof Kazem Rahimi, Mo specialises in the applications of machine learning on large-scale electronic health records to extract new insights and develop models for prognostication and diagnostication. He is involved in the PEAK Urban and Informal Cities programmes where he focuses on the effects of environmental and sociodemographic factors on health outcomes. 

Before joining the University of Oxford, he was a postdoctoral researcher at City, University of London focusing on the analysis of high-dimensional optical spectra and bio-signals, and a research assistant at University of Southampton focusing on urban mobility. 

Mo's educational background is in Applied Mathematics-Systems and Modelling (PhD, City, University of London), and Electrical Engineering (MSc, City, University of London & BSc, Iran University of Science and Technology).

He also leads a team of data scientists at a charity, Apart of Me, that has embarked upon the ambitious project of delivering the first mobile game to help children cope with bereavement.

Speaker: Diana Robinson

Diana Robinson is an incoming PhD candidate in computer science at Cambridge and Student Fellow at the Leverhulme Centre for the Future of Intelligence. Diana specialises in human-computer interaction, philosophy and business. She was a Visiting Scholar at the MIT Media Lab in the Opera of the Future group. Prior to that, she worked as a Commodity Risk Analyst in BP's Integrated Supply and Trading business. She was a Princeton Project 55 Fellow in 2012/13. Diana holds an MBA from the Cambridge Judge Business School and a BA in philosophy from Princeton University.

Focused discussions are events that aim to provide feedback to researchers for work in progress and encourage discussion. Diana's research focus is on representing and modeling uncertainty in decisions around postoperative bleeding following cardiac surgery. Her aim is to see whether a tool that is built on probabilistic programming could be useful in incorporating clinical assumptions, modeling the uncertainty explicitly and reasoning through to likelihood estimates of the next best course of action. One of her interests is in how quantifying uncertainty might impact the social process of decision making, particularly the coordination and handovers between multidisciplinary teams. By developing this from an end user perspective, and co-designing with clinicians, the aim is for clinicians themselves to eventually be able to build and adapt their own models to inform their decisions. 

Speaker: Dr. Emma Robinson

Dr Robinson's research focuses on the development of computational methods for brain imaging analysis,  and covers a wide range of image processing and machine learning topics. Most notably, her software for cortical surface registration (Multimodal Surface Matching, MSM) has been central to the development of  of the Human Connectome Project’s “Multi-modal parcellation of the Human Cortex “ (Glasser et al, Nature 2016), and has featured as a central tenet in the HCP’s paradigm for neuroimage analysis (Glasser et al, Nature NeuroScience 2016). This work has been widely reported in the media including Wired, Scientific American, and Wall Street Journal). Current research interests are focused on the application of advanced machine learning, and particularly Deep Learning to diverse data sets combining multi-modality imaging data with genetic samples. We are particualrly interested in building sensitive models of cognitive development and developmental outcome for prematurely born babies from data collected for the Developing Human Connectome Project (dHCP).

Speaker: Matt Muckley, PhD

Speaker: Prof. John Cleland

Professor John Cleland, Cardiologist & Director of the Robertson Centre for Biostatistics & Clinical Trials

The basis of modern, evidence-based medicine is the randomised controlled trial. This reflects a growing mistrust in the use of observational data to make causal inferences about the effects of lifestyle, diet or medical interventions.

There are many forms of bias in observational data including ascertainment bias, selection bias, information bias, prescribing/intervention bias and confirmation bias. Confounding may or may not be due to one or more of the above.

Amongst these biases, the one that receives the least attention (ie:- usually ignored) in medicine is ascertainment bias, both in terms of ascertaining cases and outcomes. Many analyses depend on medical diagnosis and/or adjudication committees. Although medical opinions may be of value, they are often not based on verifiable facts; a policy of ‘mistrust and verify’ should be applied. Circumstantial evidence is often more compelling than diagnostic opinion.

The application of modern analytical approaches to large datasets is an exciting possibility fraught with hazard. Focussing on associations that create hypotheses and being cautious about any perceived causal relationship is required to avoid damaging academic and clinical reputations and harm to people, patients and populations.  

“To err is human, to really foul things up requires a computer.” (attributed to William E. Vaughan, adapted from Agatha Christie, who wrote: ‘To err is human’ but a human error is nothing to what a computer can do if it tries.”)

Biography:

Professor John Cleland is a leading international researcher in heart failure, focusing on improving diagnosis, management and monitoring, and has led large, international clinical trials to study treatments. Professor John Cleland receives substantial funding from the British Heart Foundation to support his research on heart failure. He is a Clarivate Analytics Global Highly Cited Researcher for 2018, one of ten Glasgow researchers recognised as being in the top 1% for citations in their academic field. He also conducted one of the earliest large trials of home telemonitoring for heart failure and continues to promote innovations in this field. His main research interest is heart failure, including its epidemiology and prevention, and development and implementation of guidelines. He has participated in a number of randomised trials to study interventions for heart failure, including the CARE-HF, PEP-CHF and HeartCycle studies. He has been involved in research on the role of myocardial hibernation contributing to heart failure and its treatment (including beta-blockers and revascularisation), diastolic heart failure in the elderly, ventricular resynchronisation, implantable haemodynamic monitoring devices, atrial fibrillation in heart failure and advanced electrophysiology.

Speaker: Dr. Daniel Leff

Neuroergonomics to Assess and Improve Surgical Performance

In this talk, Dr. Daniel Leff will discuss his work within the Hamlyn Centre’s ‘Neuroergonomics and Perception Laboratory’ whose research remit is the evaluation of brain function in surgeons with a view to gaining insights into technical skills training, motor learning, decision-making and fatigue. Recently his focus has been in the detection of changes in brain function that can be used to differentiate high and low demand surgical tasks, facilitate workflow analysis, and which may indicate task induced burden. These factors if reliably detected may form the basis of intelligent feedback loops as means to realise brain computer interfacing to enhance safety. 

Biography:

Dr. Daniel Leff is currently a Reader in Surgery working in the Departments of BioSurgery and Surgical Technology, the Hamlyn Centre for Robotic Surgery and the Cancer Research UK Centre at Imperial College London. He is an Honorary Consultant in Oncoplastic Breast Surgery working within the Breast Unit at Imperial College Healthcare NHS Trust.  He received his PhD in Surgery from Imperial College London in 2009, working under the supervision of Professor Ara Darzi and Professor Guang-Zhong Yang. Daniel has published over 100 scientific papers in peer-reviewed journals on subjects relating to the application of surgical technologies, assessment of cortical brain function and surgical oncology. His research has been shortlisted for numerous prizes including nominations for the Hounsfield Prize (Imperial College London), Patey Prize (SARS), young scientist award at MICCAI. He has attracted funding from Industry (Hitachi Medical Corp), the Academy of Medical Sciences and Wellcome Trust, and Cancer Research UK.