Negative diagnostic_consensus when only one sublevel is not reported

[rmnldwg]

I noticed a case where I am unsure about how to handle it: If a patient's pathology report indicates a healthy sublevel Ib, but makes no statement about sublevel Ia, what should the superlevel I be?

Generally in this dataset, we assume the diagnostic_consensus to be negative (healthy), when the respective pathology report is missing (None/null). And LyProX only reports the superlevel (I in this case) to be healthy when both sublevels show no involvement. So, in the case screenshotted below, the superlevel I will always ends up being in the state "unknown", since the superlevel is None both for pathology and the diagnostic consensus.

Is that correct? Or should the superlevel involvement rather be displayed after taking into account the most likely state of all sublevels?

Currently, this affects only the LNL I (ipsi- & contralateral) of something between 58 and 67 patients in the second CLB cohort.

[rmnldwg]

The issue was resolved for now by setting the diagnostic_consensus of the second CLB dataset to healthy when the pathology information for the LNL is either missing or it is negative.

This is a sort of "hacky" solution and has the consequence that the diagnostic_consensus is potentially not correct, but since it is useless anyways (i.e., contains only "healthy" entries), it probably does not matter...