Konstantin Georgiev (MSc, BEng (Hons), PhD)

Postdoctoral Fellow in Health Data Science

  • Institute of Neuroscience and Cardiovascular Research
  • College of Medicine & Veterinary Medicine

Contact details

Address

Street

Chancellor's Building
49 Little France Crescent
Little France Campus
The Royal Infirmary of Edinburgh

City
Edinburgh
Post code
EH164SB

Biography

Background

I am a Health Data Scientist and postdoctoral researcher at the University of Edinburgh. My current research focus is on the development and validation of Multimodal AI algorithms for the diagnosis of acute cardiac conditions. Throughout my data science career, I have worked with multi-centre routine healthcare data across NHS Scotland for risk and resource forecasting and development of decision-support tools. My PhD topic was on the use of Electronic Health Records (EHRs) and data-driven algorithms to support the optimum allocation of care in older people with geriatric syndromes. During this time, I led a number of studies examining the effectiveness and predictive qualities of routine data across the NHS Lothian population. My industry experience at Red Star was also tied to the development of decision-support and auditing tools driven by Machine Learning for automated risk assessment in chronic diseases such as dementia, bone fracture and heart failure.

Qualifications

PhD in Precision Medicine (University of Edinburgh)

MSc graduate in Artificial Intelligence (University of Aberdeen), MSc DataLab alumni

BEng graduate in Software Engineering (Technical University of Varna)

Responsibilities & affiliations

Former Precision Medicine DTP student

Health Data Scientist at RedStar AI (https://redstar.ai/)

HighSTEACS, Ageing and Health, AIAI (School of Informatics) research groups

Open to PhD supervision enquiries?

Yes

Research summary

I am currently interested in Multimodal AI for precise risk diagnostics in cardiovascular care using fusion of EHR, ECG and X-ray imaging data, as well as Trustworthy/Explainable AI approaches for interpreting multimodal interactions and mitigating induced biases from routine healthcare data.

Current project grants

British Heart Foundation project grant (PG/24/12136) titled "Artificial intelligence to improve the diagnosis of acute cardiac conditions in clinical practice"

Past project grants

Sir Jules Thorn PhD scholarship (21/01PhD)
Health Improvement Scotland research grant for the project "Developing an artificial intelligence tool for dementia risk from routine healthcare data" in collaboration with Red Star
The MSc DataLab Scholarship award

Invited speaker

European Geriatric Medicine Society (EuGMS 2022, 2024 and 2025)

Process-oriented Data Science for Healthcare (International Conference for Process Mining 2023)

NHS Lothian R&D Conference 2025

Papers delivered

UNDERSTANDING HOSPITAL ACTIVITY AND OUTCOMES FOR PEOPLE WITH MULTIMORBIDITY USING ELECTRONIC HEALTH RECORDS

In: Scientific Reports (2025) https://doi.org/10.1038/s41598-025-92940-7 Statistical analysis using logistic regression to describe relationships between multimorbidity patterns and complexity of healthcare delivery in urgently hospitalised older patients. It uses routine healthcare data from NHS Lothian (via DataLoch) to study associations between long-term conditions, healthcare contacts and early readmission.

PREDICTING INCIDENT DEMENTIA IN COMMUNITY-DWELLING OLDER ADULTS USING PRIMARY AND SECONDARY CARE DATA FROM ELECTRONIC HEALTH RECORDS In: Brain Communications (2024)

https://doi.org/10.1093/braincomms/fcae469 Describes the development and validation of a Machine Learning model for predicting pre-symptomatic dementia risk across the NHS Lothian community up to 13 years prior to clinical diagnosis. It uses linked routine primary and secondary care data from DataLoch, alongside a clinically-supervised approach to develop features associated with future dementia incidence. The study aimed to provide novel insights on risk factor modification to help target novel immunotherapies in patients at high-risk of developing dementia.

COMPARING CARE PATHWAYS BETWEEN COVID-19 PANDEMIC WAVES USING ELECTRONIC HEALTH RECORDS : A PROCESS MINING CASE STUDY In: Journal of Healthcare Informatics Research (2024)

https://doi.org/10.1007/s41666-024-00181-6 Describes the development of a Process Mining algorithm (Inductive Miner) to compare COVID-19 care pathways in hospitalised patients, constructing Petri Net data structures for examining transitions in hospital. I used routine secondary care data from DataLoch alongside event data describing in-hospital contacts with healthcare providers to measure complexity of delivered care during the first two waves of the COVID-19 pandemic. I used standardised measures of Conformance Checking and Graph Edit Distance to compare patient subgroups and describe differences in delivered care between the two pandemic waves.

UNDERSTANDING HOSPITAL REHABILITATION USING ELECTRONIC HEALTH RECORDS IN PATIENTS WITH AND WITHOUT COVID-19 In: BMC Health Services Research (2024)

https://doi.org/10.1186/s12913-024-11665-x A statistical analysis describing variation in delivered hospital rehabilitation during the COVID-19 pandemic, within NHS Lothian hospitals. We used the DataLoch service to examine rehabilitation pathways in older hospitalised patients studying associations between COVID-19 and intensity of care, including total rehabilitation time, sessions per day and time to first contact. I used a mixed-effects regression approach adjusted for baseline confounders including age, sex and socioeconomic status to study associations with total healthcare contacts.

This article was published on