OpenAlex · Aktualisierung stündlich · Letzte Aktualisierung: 16.05.2026, 04:38

Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.

Mayo Clinic Strip AI

2025·0 Zitationen
Volltext beim Verlag öffnen

0

Zitationen

4

Autoren

2025

Jahr

Abstract

Acute ischemic stroke (AIS) pathological images reveal vascular lesions crucial for determining stroke etiology and guiding treatment decisions. Accurate identification of the origin of blood clots in stroke patients is critical for determining appropriate treatment strategies. For the management of these conditions in stroke patients, it is therefore pivotal to evaluate the etiology of blood clots. The goal is to develop an AI model that classifies blood clot origins in ischemic stroke. In this paper, we have constructed a Modified VGG-16 (MVGG-16) deep learning model and compared its performance with VGG-11 and VGG-16 models on the given dataset. The best result of the evaluation was achieved by MVGG-16 with 89.6% on the training set and 87.4% on the test set. The model MVGG-16 raised the level of each of the evaluated aspects and thus, the increase in the accuracy of the stroke diagnosis can be expected with the help of the given model.

Ähnliche Arbeiten

Autoren

Themen

Artificial Intelligence in Healthcare and Education
Volltext beim Verlag öffnen