MY Mithila Yeasmin Mitu

Home / Research & Publications

Research & Publications

Applied deep learning research in medical image analysis and AI-assisted diagnostic systems.

Journal ArticleAmerican Journal of Smart Technology and Solutions (AJSTS)5(2), 2026

Deep Learning-Based Detection of Renal Abnormalities from CT Images: A Study on a Bangladeshi Cohort

Authors: Mithila Yeasmin Mitu; M. A. Hossain

DOI: 10.54536/ajsts.v5i2.8226

This study presents an AI-driven CT diagnostic system for automated detection and classification of renal abnormalities in the Bangladeshi population. Xception, VGG16, and ResNet152V2 were compared across four classes: normal, cyst, stone, and tumor. Xception achieved the highest reported accuracy at 99.84%.

Artificial IntelligenceCTDeep LearningComputer-Aided DiagnosisRenal Abnormalities
View DOIJournal Page
Journal ArticleAmerican Journal of Innovation in Science and Engineering (AJISE)5(2), 2026

Automated Skin Disease Classification Using Deep Neural Networks: A Study on a Bangladeshi Dataset

Authors: Mithila Yeasmin Mitu; M. A. Siddiki

DOI: 10.54536/ajise.v5i2.8227

This study develops a deep learning framework for automated skin disease classification using a dermatologist-annotated Bangladeshi dataset. Six architectures were evaluated; NASNetLarge achieved the highest reported accuracy at 91.0%, with 90.5% precision, 90.8% recall, and 90.6% F1-score.

Convolutional Neural NetworksDeep LearningImage ClassificationMedical Image ProcessingSkin Disease Detection
View DOIJournal Page