Skip to main content

Table 10 Comparison of classification performance in individual classes

From: Interpretable CNN for ischemic stroke subtype classification with active model adaptation

Subtype

Precision*

Recall*

F1-score*

Precision

Recall

F1-score

LAA

0.5960

0.8774

0.7042

0.6559 (+ 0.0599)

0.7552 (− 0.1222)

0.6994 (− 0.0048)

CE

0.6220

0.3857

0.4705

0.7208 (+ 0.0988)

0.5343 (+ 0.1486)

0.5923 (+ 0.1218)

SAO

0.3966

0.2821

0.2910

0.5690 (+ 0.1724)

0.5392 (+ 0.2571)

0.5446 (+ 0.2536)

OC

0.2917

0.0682

0.1020

0.4785 (+ 0.1868)

0.2747 (+ 0.2065)

0.3277 (+ 0.2257)

UND

0.3067

0.0280

0.0507

0.3943 (+ 0.0876)

0.2391 (+ 0.2111)

0.2825 (+ 0.2318)