Detection of PCOS using CNN and Vision Transformers
Procedia Computer Science · 2026 · 283 · 4350-4359
DOI: 10.1016/j.procs.2026.06.493 ↗Kurz gefasst
Fachartikel in der Zeitschrift Procedia Computer Science, 2026. Verfasst von S. Kavishree und 4 weiteren Autor:innen. Die Arbeit nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Polycystic Ovary Syndrome (PCOS) sometimes leads to infertility, insulin resistance, and long-term metabolic complications like diabetes, chronic heart disease in young adult women. The classification of PCOS is necessary for timely diagnosis and treatment planning. Deep Convolutional Neural Network (CNN) architecture like ResNet50, are widely adopted to detect PCOS due to their capability to record spatial elements that are hierarchical. Vision Transformer models, an alternative to CNNs, have not been extensively studied in this area. Therefore, for the detection of PCOS, this paper fine-tunes and analyzes the performance of ViT, ConvViT, and SWIN Transformer using ultrasound images. This paper also compares CNN architecture (ResNet50) architecture for the classification of PCOS-positive and negative cases using ovarian ultrasound images. It leverages transformer-based architectures to capture global contextual relationships for improved feature representation within the images, enabling more accurate and reliable diagnostic predictions. The proposed vision transformer-based architecture provides better accuracy than the existing CNN models.
Zitieren
APA
Kavishree, S., K, K., Mothieshwar, Nichitha, & Mahapatra, A. (2026). Detection of PCOS using CNN and Vision Transformers. Procedia Computer Science, 283, 4350-4359. https://doi.org/10.1016/j.procs.2026.06.493
Vancouver
Kavishree S, K K, Mothieshwar, Nichitha, Mahapatra A. Detection of PCOS using CNN and Vision Transformers. Procedia Computer Science. 2026;283:4350-4359. doi:10.1016/j.procs.2026.06.493
BibTeX
@article{Kavishree2026,
author = {Kavishree, S. and K, Kanish and Mothieshwar and Nichitha and Mahapatra, Ansuman},
title = {Detection of PCOS using CNN and Vision Transformers},
journal = {Procedia Computer Science},
year = {2026},
volume = {283},
pages = {4350--4359},
doi = {10.1016/j.procs.2026.06.493}
}
EUVIMED