Machine Learning-Based Optimization of Physical Layer Quantization for Diabetic Retinopathy Detection
Cheng Ding Rui Yang Xiaoke Bi Yi Wang
cheng.ding@duke.edu ry83@duke.edu yw374@duke.edu xiaoke.bi@duke.edu
| Paper PDF |
|
Early detection of Diabetic Retinopathy helps slow down the progression to visionimpairment. Such demand could be effectively satisfied by machine learning-based medical image recognition and classification. Physical layer optimizationis important for the overall CNN performance. In this final project, we reported aCNN model with a quantization-optimized physical layer that enables DR severityclassification tasks with high accuracy. Please feel free to add any additional links that are relevant here. |
|
|
| Paper: |
Code and Data:
|