TernScope is a model-hardware co-design approach for endoscopic segmentation using ternary weights and 8-bit activations. The resulting 0.2483M parameter model enables multiplier-free computation and packed storage, making it suitable for low-power edge medical devices.
HOW THIS AFFECTS YOU
●
builderYou can deploy highly efficient segmentation models on resource-constrained hardware using ternary quantization.
●
healthThis enables real-time, high-accuracy AI analysis directly on medical edge hardware.