Samra Irshad

PhD Student (2022.9~)

Department of Computer Science and Engineering, Kyung Hee University

Contact
samra@khu.ac.kr
#351, Electronic Information College Building, 1732 Deogyeong-daero, Giheung-gu, Yongin, Gyeonggi-do, 17104, South Korea

Research Interest

  • Robustness
  • Adversarial Attack

Publication

  • [J3] Interpretability of Multimodal Neural Networks for Prediction of Visual Acuity in Patients with Branch Retinal Vein Occlusion
    Soyoun Won†, Kiyoung Kim†, Youngseob Won, Samra Irshad, Sungyoung Lee, Seung-Young Yu, Seong Tae Kim
    Scientific Reports, 2026

  • [J2] Adversarial Wear and Tear: Exploiting Natural Damage for Generating PHysical-world Adversarial Examples
    Samra Irshad, Seungkyu Lee, Nassir Navab, Hong Joo Lee*, Seong Tae Kim*
    IEEE Transactions on Dependable and Secure Computing, 2026

  • [C1] No Evidence of Disease: Clinically-Risky Adversarial Chest CT Report Generation
    Samra Irshad, Junho Kim, Seong Tae Kim*
    Medical Imaging with Deep Learning conference (MIDL), 2026 (Taipei)

  • [J1] Improved Abdominal Multi-Organ Segmentation via 3D Boundary-Constrained Deep Neural Networks
    Samra Irshad, Douglas P. S. Gomes, Seong Tae Kim*
    IEEE Access, 2023