A 65 nm Trustworthy Hypoglycemia Forecasting Engine Achieving 11.3 nJ per Inference
2026-06-05 • Hardware Architecture
Hardware Architecture
AI summaryⓘ
The authors developed a special computer chip to help predict low blood sugar levels early in people with diabetes. Their chip uses a clever mix of methods to make decisions that are both reliable and energy efficient, while also explaining how it made those decisions. It works well even when sensor data is noisy or incomplete, making it trustworthy for medical use. This chip can run on-the-spot glucose data and forecast hypoglycemia with good accuracy, using less power than other methods.
diabetescontinuous glucose monitoringhypoglycemia forecastingprobabilistic decision treesenergy efficient AIexplainable AIRISC V coreCMOS chipsensor noise robustness
Authors
Boyang Cheng, Jianbo Liu, Pengyu Ren, Xueji Zhao, Steven Davis, Likai Pei, Zephan M. Enciso, Kai Ni, Ningyuan Cao
Abstract
Diabetes affects millions of people and requires reliable continuous glucose monitoring for early hypoglycemia warning. However, medical AI systems must be not only accurate and energy efficient, but also explainable, noise robust, and uncertainty aware. This work presents a 65 nm hypoglycemia forecasting engine based on probabilistic decision trees for trustworthy medical inference. The proposed hybrid architecture combines exact arithmetic evaluation for shallow tree layers with sampling based inference for deeper layers, reducing soft decision tree complexity from exponential to sample efficient traversal. A reconfigurable 4 by 24 by 24 probabilistic node array supports arbitrary tree structures with a maximum depth of 12, coordinated by an on chip low power RISC V core. Fabricated in 65 nm CMOS, the chip achieves 11.3 nJ per inference and a state of the art 30 min forecasting F1 score of 0.825 on continuous glucose monitoring data. Compared with conventional decision tree and random forest models, the proposed engine improves robustness to sensor noise and data point drop off by 4.1x to 16.1x. These results demonstrate an energy efficient, explainable, and uncertainty aware edge AI engine for trustworthy hypoglycemia forecasting.