Biomedical and Health AI Scientist

Duties of the post

The duties of the post will include:

• Developing an AI-enabled biomarker discovery engine, using multi-modal data, to address growing demand in the precision medicine market for disease risk prediction services
• Handling and analysing biomedical signals, medical text and image data
• Accessing, managing and handling new to Hurdle data types such as EHRs and wearable signals
• Fusing these new data types with multi-omics data
• Finding novel ways of bringing a temporal dimension to Hurdle’s predictive models using state-of-the-art time series analysis methodologies
• Finding ways to reduce the cost of data acquisition and speed of discovery – model performance will need to be carefully balanced with commercial drivers
• Contributing to embedding expert knowledge into Hurdle’s staff base
• Contributing to academic papers

These duties are a guide to the work that the post holder will initially be required to undertake. They may be changed from time to time to meet changing circumstances.

Key requirements

Qualifications:

• A higher degree (Master’s) in computer science, data science or AI, or bioinformatics

Experience:

• Deploying state-of-the-art AI and ML techniques in an industry setting
• Working with health-related data, particularly medical images and timeseries and bio-signal processing
• Data pre-processing, feature engineering, and statistical analysis in healthcare settings
• Creating demos and prototypes utilizing models
• Producing and presenting data visualisations
• Knowledge of cloud systems (AWS and GCP)
• Running, training, and evaluating a variety of ML models
• Working with different biomedical signals, such as medical images and timeseries
• Working with large-scale cohorts (e.g., UK Biobank, Our Future Health, All of Us) and handling population-level biomedical data
• Multimodal biomedical data, including MRI, CT, pathology imaging, and UK Biobank imaging modalities (MRI, DXA, fundus, OCT, ultrasound)

Knowledge:

• Cloud systems (AWS and GCP)
• Exposure to multi-omics datasets (genomics, proteomics, transcriptomics, metabolomics)

Skills:

• Advanced statistical and technical skills for working with large databases
• Higher level programming language skills in Python
• Skills with technologies used for data storage, access and manipulation
• High-level skills in data fusion and time series techniques
• Knowledge of fine-tuning open-source models
• Excellent communication and presentation skills
• Excellent command of written and spoken English
• Ability to work as an effective team member and independently as and when required
• Ability to manage a budget and forecast expenditure
• Excellent people management skills
• Familiarity with privacy-preserving ML techniques (e.g., federated learning, differential privacy)

Benefits

As a KTP Associate, the post will offer the following benefits:

• A personal development budget of £5340 (exclusive of salary)
• Management training and mentoring by an Innovate UK KTP Adviser
• An interesting and challenging role, with exposure to a variety of stakeholders
• Full access to university resources to complete the project
• World-leading Academic and Company project supervision, with project support by a dedicated, sector leading KTP Office

The University of Essex, in partnership with Hurdle, offers an exciting opportunity to a graduate with the relevant skills and knowledge to develop an AI-enabled biomarker discovery engine, using multi-modal data, to address growing demand in the precision medicine market for disease risk prediction services.