Projects

Project submission form

If you would like to work with us, have a project that needs our support in any project stage, or have a product you would like us to evaluate for clinical use, please complete and submit the Project Initiation Document (PID) via email to CSCTeam@gstt.nhs.uk

Project stages

The Project Board below tracks the progress of each CSC project through the different phases of the software developmental lifecycle.

These stages are:

Scoping: Initial conversations between project stakeholders, planning of design and development ideas, discussions about verification and evaluation methods.
Documenting & Governance: Final project plans are documented and circulated to stakeholders. For projects involving SMEs, preparation of Information Security (IS) and Information Governance (IG) compliance documents.
Contracting: For projects with SMEs, preparation of Pilot Agreement and Processing Agreement documents.
Data Curating: Preparation and data access logistics, such as for training AI algorithms or retrospective evaluation studies.
Developing: Development of AI algorithms and software. This stage also includes validation and verification testing. Projects with SMEs bypass this stage.
Evaluating: Evaluation comprises two parts: 1) retrospective evaluation using legacy data to determine if the software is appropriate for clinical practice, or requires further development. 2) prospective evaluation in a live clinical setting.
Reporting: Report produced for the software owner detailing the results of the evaluation stage, potentially highlighting any additional design and development considerations. For SMEs, this report serves as the basis for submitting applications to the Trust procurement team.
Concluded: Previous projects which are now concluded.
Backlog: Projects which have not been started yet. May be due to lack of project owner engagement, missing information relating to the project, or lack of capacity within CSC team to support the project.
Graveyard: Terminated projects, possibly because of a failure at one of the stages above.

Scoping

Initial conversations between project stakeholders, planning of design and development ideas, discussions about verification and evaluation methods.

There are currently no projects at this stage of the development cycle.

Documenting & Governance

Final project plans are documented and circulated to stakeholders. For projects involving SMEs, preparation of Information Security (IS) and Information Governance (IG) compliance documents.

There are currently no projects at this stage of the development cycle.

Contracting

For projects with SMEs, preparation of Pilot Agreement and Processing Agreement documents.

Incidental Pulmonary Embolism

An automated tool for identifying and prioritising patients with Pulmonary Emboli.

Pulmonary Embolism Response Team AI

AI enabled detection and classification of high risk PE patients

Data Curating

Preparation and data access logistics, such as for training AI algorithms or retrospective evaluation studies.

Osteoporosis

Timely identification of vertebral compression fractures to promote early treatment interventions.

Developing

Development of AI algorithms and software. This stage also includes validation and verification testing. Projects with SMEs bypass this stage.

CARNAX – Computer Assisted Reporting of Neonatal Abdominal X-rays

AI-based medical imaging tool for detection of intestinal perforation.

AI fairness toolkit

Development of an AI fairness toolkit for bias checking in AI evaluation results.

Evaluating

Evaluation comprises two parts: 1) retrospective evaluation using legacy data to determine if the software is appropriate for clinical practice, or requires further development. 2) prospective evaluation in a live clinical setting.

X-Ray Occult Carpal Detection

A computer aided diagnosis tool which would automatically run when a scaphoid fracture is suspected.

Radiotherapy Auto Segmentation

Single application for all in-house auto-segmentation contours

Reporting

Report produced for the software owner detailing the results of the evaluation stage, potentially highlighting any additional design and development considerations. For SMEs, this report serves as the basis for submitting applications to the Trust procurement team.

NLP for optimisation of endoscopic resources

Natural Language Processing of Endoscopic and associated pathology text to optimise post COVID endoscopy resources

Detection and classification of lesions in dental panoramic tomograms

AI enabled detection and classification of lesions in dental panoramic tomograms

Referral Automated Vetting Engine (RAVE)

A generalisable Large Language Model (LLM) based vetting engine for analysis and management of patient referrals

Concluded

Previous projects which are now concluded.

Synthetic CT for MR-Only Prostate Radiotherapy Planning

AI to create a Synthetic CT for MR-Only Prostate Radiotherapy Planning

Smart brain MRI protocols

Evaluation of brain MRI smart protocoling software

Evaluation of Chest X-Ray AI Engine

An evaluation of an AI Engine that identifies 10 common chest radiograph pathologies

Motion-corrected 3D Fetal Brain MRI

An automated software workflow for reconstruction of motion-corrected 3D fetal MRI images

Backlog

Projects which have not been started yet. May be due to lack of project owner engagement, missing information relating to the project, or lack of capacity within CSC team to support the project.

There are currently no projects at this stage of the development cycle.

Graveyard

Terminated projects, possibly because of a failure at one of the stages above.

Lung Nodules

Automatic lung nodule detection and characterisation on CT Chest

MRI Occult Carpal Detection

AI application to interpret MRI scans to increase sensitivity and confidence of diagnosis.

Rotational Profiles

An AI tool to measure degrees of misalignment of lower limb rotation in pre-surgical planning assessments .

Early detection of rectal cancer

An application that identifies areas of concern on MRI images for the detection of rectal cancers/polyps

Characterising Adrenal Adenomas

The use of AI in characterising incidental adrenal lesions on portovenous CT abdomen and pelvis.

Radiomic features from CT to predict lung cancer

AI to extract Radiomic Features from CT images as non-invasive tool to predict outcome of patients with lung cancer.

Prosthetic Loosening

An AI tool to improve accuracy and increase confidence in reporting of prosthetic loosening.

Perioperative complexity stratification and optimisation

AI-powered algorithm for categorising patients based on their perioperative risk, enabling appropriate pathway planning and optimisation in elective surgeries.