EUVIMEDEUVIMED entdecken
Crossref

AI Disease Recognition System using Machine Learning

Madhuramya R, Rohith M, Vasanthkumar R, Siva Velan A

International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)

DOI: 10.15662/ijeetr.2026.0802025 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von Madhuramya R und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Healthcare diagnosis remains a major challenge in modern medicine, especially in areas with limited medical expertise. Traditional diagnostic methods depend heavily on trained professionals and specialized equipment. This creates barriers to timely access to healthcare. This paper proposes a dual-diagnosis intelligent healthcare system that combines machine learning-based symptom analysis with deep learning for medical image classification to support disease prediction. The proposed system analyzes structured symptom data and unstructured medical images to identify diseases across 41 categories and detect pneumonia from chest X-ray images. The system has two complementary modules: a symptom-based disease prediction model that uses ensemble machine learning algorithms, including Random Forest, Support Vector Machine, Decision Tree, and Naive Bayes, trained on 4,920 symptom-disease records, and a deep learning-based pneumonia detection model using transfer learning architectures like VGG16, ResNet50, and a baseline CNN trained on 5,863 chest X-ray images. The framework also provides confidence scores and visual explanations using Gradient-weighted Class Activation Mapping (Grad-CAM) to improve interpretability. Experimental results show that the Random Forest classifier achieved an accuracy of 85.3% for symptom-based prediction with an F1-score of 83.9%. The ResNet50 transfer learning model achieved 93.7% accuracy with an AUC of 0.96 for pneumonia detection. The integrated hybrid system achieved a combined accuracy of 94.1% when both symptom and imaging data were used, outperforming single-modality approaches. The system is available through a web-based interface with an average response time of 2.1 seconds, enabling real-time diagnosis suitable for telemedicine and resource-limited healthcare settings

Zitieren

APA

R, M., M, R., R, V., & A, S. V. (2026). AI Disease Recognition System using Machine Learning. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802025

Vancouver

R M, M R, R V, A SV. AI Disease Recognition System using Machine Learning. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802025

BibTeX
@article{R2026,
  author = {R, Madhuramya and M, Rohith and R, Vasanthkumar and A, Siva Velan},
  title = {AI Disease Recognition System using Machine Learning},
  journal = {International Journal of Engineering & Extended Technologies Research},
  year = {2026},
  volume = {08},
  number = {02},
  doi = {10.15662/ijeetr.2026.0802025}
}
Kostenlos starten →