Open Access Open Access  Restricted Access Subscription Access

Improved Detection of Dermatological Conditions with AI-Based Models

SUBRAJA T, Dr. M Marikkannan

Abstract


This paper presents a machine learning-based framework for enhancing the accuracy of early skin disease detection and classification. The proposed approach involves a hybrid model combining Principal Component Analysis (PCA) for dimensionality reduction and Naive Bayes for classification. Various supervised algorithms, including Support Vector Machines (SVM), Decision Trees, K-Nearest Neighbors (KNN), and ensemble models, are also evaluated. Results show that the hybrid model outperforms other methods in accuracy and efficiency, offering promising support for dermatological diagnostics.


Full Text:

PDF

References


Biswas, N., et al. (2023). Early detection of heart disease using machine learning and feature selection. BioMed Research International.

Reddy, K. V., et al. (2021). Heart disease risk assessment with PCA-based machine learning. IEEE International Conference on Intelligent and Advanced Systems (ICIAS).

Ganesan, M., & Sivakumar, N. (2019). Heart disease diagnosis using IoT and ML models. IEEE International Conference on Smart Computing and Communication (ICSCAN).

Kumar, P. M., & Gandhi, U. D. (2018). An IoT-driven framework for early heart disease detection. Computers and Electrical Engineering, 65, 222–235. https://doi.org/10.1016/j.compeleceng.2017.09.005

Verma, P., & Sood, S. K. (2018). A cloud-based IoT framework for medical diagnosis. Journal of Parallel and Distributed Computing.

Berikol, B., & Yildiz, O. (2016). SVM-based acute coronary syndrome diagnosis. Journal of Medical Systems, 40(4). https://doi.org/10.1007/s10916-016-0452-5

Shetty, A., & Naik, C. (2016). Comparative analysis of data mining techniques for heart disease prediction. International Journal of Innovative Research.


Refbacks

  • There are currently no refbacks.