Diagnosis Stratification Tool (DST)

Diagnosis Stratification� (DST)

is an algorithm based clinical evaluation tool that can identify undiagnosed patients with a high likelihood of having a Long Term Condition (LTC). It uses past medical history of a patient and estimates the probability of each patient to have an LTC.


To put it simply, the algorithms are based on AI and use millions of clinical patterns from patients with a LTC during training. Once training is completed, the same algorithms can be applied to patients that do not have a diagnosed LTC but present similar patterns to patients with a LTC. Depending on how closely those patterns correlate, a probability percentage is calculated. The table below shows the number of patients identified with undiagnosed conditions for three confidence intervals for various LTCs in England. Patients within the 90% confidence interval would benefit from accurate diagnosis, patients up to 80% may be invited to clinics and patients up to 70% may be invited for targeted screening programmes.