What Changed at Brainomix

Brainomix announced new analyses from the phase III INBUILD trial in progressive pulmonary fibrosis. In 474 patients from the trial's CT substudy, e-Lung measures detected the effect of nintedanib, and higher baseline disease extent was linked to faster decline in lung function (FVC) over 52 weeks. e-Lung results were broadly consistent with a UCLA research algorithm run on the same scans.

Our read

Agreement with an independent algorithm and sensitivity to a known treatment effect are what drug sponsors look for before using quantitative CT as a trial endpoint. This is a post hoc analysis, so it supports e-Lung in research more than it proves benefit at the bedside.

A study was published evaluating what happened when Brainomix 360 Stroke was implemented in a high volume stroke system that already used routine perfusion imaging. The finding is that the platform improved the efficiency of the acute stroke pathway even in a setting that was not imaging constrained to begin with. The design matters here, because it is an implementation study in an operating service rather than a retrospective accuracy comparison against a reference standard.

Our read

Most stroke AI evidence measures whether the algorithm agrees with an expert reader, which tells a buyer very little about whether the service gets faster. This one measures the pathway, which is the thing hospital leaders are actually purchasing. The high volume, routine perfusion setting is the harder test case, since those centers have the least headroom for an AI tool to add value, so a positive result there travels further than one from an under resourced site.