Neura had 98% accuracy when predicting negative results for inclusion in pooled testing bins.
Increase in Testing Capacity
Neura enabled efficient pool testing by accurately predicting which bins to allocate tests. This resulted in a less than 2% positive infection rate per bin (as opposed to 12% without Neura’s predictive model). Neura took samples from a population that had 12% positive infection and sorted these tests into select bins to for pooling so that each bin was less than 2% positive infection. Only by batching the negatives together, can you achieve a 6x increase in efficiency. The only way to batch the negatives together, is by allocating a predictive score, indicating which tests are negative.