Imatge de la portada

Detection of Blood Clot in Brain Using Supervised Learning Algorithms

Guardat en:
Publicat a:The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings (2025), p. 1-6
Autor principal: Pisote, Anita
Altres autors: Bhaturkar, Deepali N, Thosar, Devidas S, Thosar, Rajashree D, Deshmukh, Atharav, Borate, Vishal
Publicat:
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Matèries:
Machine learning
Algorithms
Stroke
Ischemia
Ensemble learning
Supervised learning
Accés en línia:Citation/Abstract
Etiquetes: Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
Resum:Conference Title: 2025 6th International Conference for Emerging Technology (INCET)Conference Start Date: 2025 May 23Conference End Date: 2025 May 25Conference Location: BELGAUM, IndiaIschemic Stroke and its subtypes detection in real time at early stage is a complex task and requires advanced terminologies and tools to learn the statistical pattern form raw data and gets generalize at various hospital patholabs and laboratory. In recent time, advance supervised learning based algorithms, learning algorithms are being used to developed laboratorical equipments to resolve the issue of early detection. The paper presents a analytical and diagnostic suggestion by employing variants of supervised learning - hyperparamerized machine learning algorithms, ensemble learning algorithms and a dependency based neural network architecture LSTM on patient-level data of patients with ischemic stroke are employed. The objective of research is to find and provide a best supervised learning algorithm which aims in providing effective and acceptable accuracy. For evaluating accuracy employed various aspect of classification report are being used. The experimental results ensures that advance ensemble learning algorithms have achieved accuracies above 80% for stroke and its subtypes while gated RNN architecture - LSTM has also achieved accuracies over 80% but with a less recall score as compared to ensemble and hyperparameterized machine learning algorithms highly prone to overfitting and performs well in training data only
DOI:10.1109/INCET64471.2025.11140127
Font:Science Database