Enhancing malaria detection accuracy using deep learning techniques for improved diagnosis
Raja Ranjan, N. Usha Rani, Anita Kumari, Ravikant Nirala · 2025
Malaria is caused by plasmodium parasite infestation, a serious health problem in tropical and subtropical regions. In this work, we focus on using deep learning techniques in malaria diagnosis on cell image basis, since it is the need of hour to test the patient early and accurately so that he could be treated early and mortality can be avoided. In particular, transfer learning is applied to classify parasitized and uninfected blood smear pictures. For validation, we used a publicly available Kaggle dataset and achieved resilience across the sample noise and accuracy of about 97%. It not only provides overview of comparative analysis, evaluation metrics, and experimental setups for the sake of maximizing deep learning&s;s effectiveness in medical diagnostics, but also summarizes various researches done in medical image segmentation in the previous 5 years.