Karim Mohammadi
Papers
2
Total Citations
7
H-Index
2
About
Karim Mohammadi’s research bridges two transformative frontiers: medical AI and neuromorphic engineering. In medical imaging, he developed a fast, accurate approach for detecting and segmenting melanoma skin cancer by fine-tuning YOLOv3 and SegNet with deep transfer learning. This work, which has garnered 5 citations, directly addresses the critical need for early diagnosis—improving segmentation accuracy to aid clinicians and surgical robots in precisely removing malignant lesions, thereby increasing cure rates. On the neuromorphic side, Mohammadi has advanced biologically inspired computing by emulating a central pattern generator (CPG) using CMOS neurons and memristor-based synapses. This hardware implementation, cited 2 times, holds promise for neural prosthetics that restore lost motor functions due to neurological injury, as well as for industrial robotics requiring rhythmic control. By tackling both cancer detection and neural circuit emulation, Mohammadi demonstrates a rare versatility—applying deep learning to save lives and hardware neuroscience to restore movement. His work exemplifies how computational methods can address pressing medical challenges while pushing the boundaries of intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2