Kheiron Medical is focussed on helping radiologists detect breast cancer earlier using deep learning.
Breast screening has saved many lives, but overdiagnosis and overtreatment are common. In many countries such problems have led to scans being checked routinely by a second radiologist, which improves accuracy but adds to workloads.
PROBLEM(S) TO SOLVE
The technology aims to solve two key problems:
- ease the workload on overstretched screening units by serving as a second reader to support breast screening programmes
- increase the accuracy of screening by reducing the number of false positives and false negatives, and hence number of patients getting unnecessary biopsies
IMPACT FOR USER
Patients (and their families, friends and relatives) benefit psychologically from requiring fewer unnecessary biopsies and avoiding the often agonising waits to know the final diagnosis.
IMPACT FOR HEALTHCARE SYSTEM
Reduction in workload and improved accuracy are key benefits and this is being assessed in a health economic research study underway. The resource requirements and costs are minimal to run the software; it requires an IT person to run the programme on the computer and the radiologist would also need to audit as part of a normal auditing process.
The product has received CE marking (Class IIa).
EVIDENCE TO DATE
The software is undergoing clinical trials in an NHS Test Bed and in Europe. Evidence to date has been collected through retrospective outcomes analysis.
Overall value assessment
While further research is required to validate these potential benefits, the technology could:
achieve saving in the current spend of over £3 billion on diagnostics in the NHS
deliver peace of mind and reassurance to patients with fewer false positives and false negatives
provide benefits to staff in terms of reduced workload and improved accuracy of screening