Veterinary Medicine: Bringing low-cost low-dose 3D imaging to the veterinary surgeons treatment room

This project is to deliver a 3D imaging system for small animal use by veterinary surgeons (the 'Small Animal 3D' \['SA3D'\] system) which will be launched at the British Small Animal Veterinary Association Congress 25-28 Mar 2021\. The system will be a low-cost low-dose (~same as existing 2D imaging systems) portable system that can deliver 3D images, whilst avoiding the need to purchase a CT.

The innovation is to use 'Digital Tomosynthesis' ('DT') to create 3D images in veterinary applications. DT is routinely used in medical imaging (for instance in breast screening) but is currently only deployed in fixed high-cost installations. The innovations are:

**Innovation 1:** Deliver 3D imaging in a form-factor that can be deployed on a standard operating table in an existing veterinarian's treatment room or operating room without modification (including the use of single-phase power). SA3D would operate within the same radiation protection regulations and have similar running costs as 2D. Operation of SA3D can be by the vet or nurse without the need for specialist training or dedicated specialist technicians.

**Innovation 2:** Deliver 3D imaging that can be moved within the veterinary practice to enable 3D radiography to travel to the patient, rather than to take the patient to the 3D radiography. SA3D would be portable outside the clinic, enabling quick access to imaging diagnostics to patients who can't travel, cannot be moved, or need urgent access to imaging.

**Innovation 3:** Create a new business model (for veterinary imaging) that creates an 'archive' of data to which Machine Learning can be applied to allow new computer-aided diagnostic tools to be developed and deployed to users. It is envisaged that incidental findings could be automatically identified for the clinician to review, such as identifying tumour metastasis when investigating a primary tumour.

**Innovation 4:** Use modified acquisition protocols in conjunction with advanced image processing and mathematical techniques (compressed sensing) to allow enhanced imaging such that the system can acquire images of sufficient quality for extraoral dental imaging with a large detector with large pixel sizes. This will avoid the need for multiple intraoral images to be acquired and avoid the need for the vet to acquire a veterinary intraoral system.

There will be value in environmental terms of reducing the need for vet practices to acquire a CT or (if no CT is on-site) for vets to refer patients to other centres for imaging.