1
00:00:00,250 --> 00:00:01,884
A result should be inspectable.

2
00:00:02,034 --> 00:00:07,872
This engineering demo connects a visual path to recorded inputs, explanations
and source references.

3
00:00:10,250 --> 00:00:11,600
Select an evidence layer.

4
00:00:11,750 --> 00:00:17,215
Inspect the recorded inputs, read the explanation, and follow the source
identifiers.

5
00:00:17,365 --> 00:00:19,812
Each layer stays connected to the same case.

6
00:00:24,250 --> 00:00:27,126
An absent record must not become a negative finding.

7
00:00:27,276 --> 00:00:30,635
Here, a missing note remains distinct from a recorded zero.

8
00:00:30,785 --> 00:00:33,024
The explanation preserves what is unknown.

9
00:00:40,250 --> 00:00:42,746
Conflicting evidence stays visible for review.

10
00:00:42,896 --> 00:00:49,731
Inspect the explanation and both source identifiers, rather than silently
collapsing different records into one answer.

11
00:00:53,250 --> 00:00:57,338
Filter the synthetic cohort, then inspect the case behind its state.

12
00:00:57,488 --> 00:01:00,600
The table and chart describe the same prepared records.

13
00:01:05,250 --> 00:01:08,476
This supports a trustworthy, knowledge-driven architecture.

14
00:01:08,626 --> 00:01:11,488
Neural interpretation is a future integration layer.

15
00:01:11,638 --> 00:01:14,032
Start with one evidence inspection module.

16
00:01:14,182 --> 00:01:17,571
These are prepared states, not a clinical inference run.
