To design an efficient intervention, decision-makers need not only to estimate the total effect of the intervention, but also understand the underlying causal mechanisms driving this effect.

Signos is the first FDA-cleared, AI-enabled system that uses CGMs to nudge you towards healthier behaviors. According to the American Diabetes Association, around 7 million people in the United States are undiagnosed, with 1 in 3 Americans at risk for developing type 2 diabetes. If you do not go on medication, you can manage the condition--a chronic metabolic disease that's characterized by elevated blood sugar levels--by exercising and watching what you eat (very, very closely). In the past few years, the tools that diabetics use to help manage their condition have become more widely available. Continuous glucose monitors (CGMs) like the Abbott Lingo and the Dexcom Stelo used to be available only by prescription.

Deep sequence models for blood glucose forecasting consistently fail to leverage clinically informative drivers--insulin, meals, and activity--despite well-understood physiological mechanisms. We term this Driver-Blindness and formalize it via $Δ_{\text{drivers}}$, the performance gain of multivariate models over matched univariate baselines. Across the literature, $Δ_{\text{drivers}}$ is typically near zero. We attribute this to three interacting factors: architectural biases favoring autocorrelation (C1), data fidelity gaps that render drivers noisy and confounded (C2), and physiological heterogeneity that undermines population-level models (C3). We synthesize strategies that partially mitigate Driver-Blindness--including physiological feature encoders, causal regularization, and personalization--and recommend that future work routinely report $Δ_{\text{drivers}}$ to prevent driver-blind models from being considered state-of-the-art.

Type 2 diabetes prevention and treatment can benefit from personalized lifestyle prescriptions. However, the delivery of personalized lifestyle medicine prescriptions is limited by the shortage of trained professionals and the variability in physicians' expertise. We propose an offline contextual bandit approach that learns individualized lifestyle prescriptions from the aggregated NHANES profiles of 119,555 participants by minimizing the Magni glucose risk-reward function. The model encodes patient status and generates lifestyle medicine prescriptions, which are trained using a mixed-action Soft Actor-Critic algorithm. The task is treated as a single-step contextual bandit. The model is validated against lifestyle medicine prescriptions issued by three certified physicians from Xiangya Hospital. These results demonstrate that offline mixed-action SAC can generate risk-aware lifestyle medicine prescriptions from cross-sectional NHANES data, warranting prospective clinical validation.

To design an efficient intervention, decision-makers need not only to estimate the total effect of the intervention, but also understand the underlying causal mechanisms driving this effect.

Abstract--Continuous glucose monitoring (CGM) provides dense and dynamic glucose profiles that enable reliable estimation of Ambulatory Glucose Profile (AGP) metrics, such as Time in Range (TIR), Time Below Range (TBR), and Time Above Range (T AR). However, the high cost and limited accessibility of CGM restrict its widespread adoption, particularly in low-and middle-income regions. In contrast, self-monitoring of blood glucose (SMBG) is inexpensive and widely available, but yields sparse and irregular data that are challenging to translate into clinically meaningful glycemic metrics. In this work, we propose a Dual-Path Attention Neural Network (DPA-Net), to estimate AGP metrics directly from SMBG data. DPA-Net integrates two complementary paths: (1) a spatial-channel attention path that reconstructs a CGM-like trajectory from sparse SMBG observations, and (2) a multi-scale ResNet path that directly predicts AGP metrics. An alignment mechanism between the two paths is introduced to reduce bias and mitigate overfitting. In addition, we develop an active point selector to identify realistic and informative SMBG sampling points that reflect patient behavioral patterns. Experimental results on a large, real-world dataset demonstrate that DPA-Net achieves robust accuracy with low errors, while reducing systematic bias. T o the best of our knowledge, this is the first supervised machine learning framework for estimating AGP metrics from SMBG data, offering a practical and clinically relevant decision-support tool in settings where CGM is not accessible. With the steadily increasing prevalence of diabetes, it has become one of the most common and challenging chronic diseases worldwide, imposing a substantial burden on public health [1].

Continuous glucose monitoring (CGM) generates dense data streams critical for diabetes management, but most used forecasting models lack interpretability for clinical use. We present SSM-CGM, a Mamba-based neural state-space forecasting model that integrates CGM and wearable activity signals from the AI-READI cohort. SSM-CGM improves short-term accuracy over a Temporal Fusion Transformer baseline, adds interpretability through variable selection and temporal attribution, and enables counterfactual forecasts simulating how planned changes in physiological signals (e.g., heart rate, respiration) affect near-term glucose. Together, these features make SSM-CGM an interpretable, physiologically grounded framework for personalized diabetes management.

Non-invasive glucose monitoring remains a long-standing challenge in biomedical sensing, with the potential to transform diabetes management by eliminating painful finger pricks, reducing consumable costs, and enabling real-time glucose tracking. Despite decades of research, no solution has matched the accuracy and reliability of invasive methods in real-world deployment. Near-infrared (NIR) spectroscopy -- leveraging glucose's absorption features in the 850-1150 nm range -- remains the most promising modality, but progress has stalled due to the problem's intrinsic complexity. At its core, non-invasive glucose monitoring is an ill-posed inverse problem. The NIR signal measured at the skin surface reflects a convoluted mixture of tissue scattering, overlapping spectral absorption from water, hemoglobin, and fat, and variations in skin thickness, perfusion, and melanin content. These physiological variables are compounded by hardware noise (e.g., ADC quantization, LED instability, photodiode dark current) and environmental drift (e.g., temperature, humidity, ambient light). In the field, these effects suppress glucose-NIR correlation to ρ 0.21, rendering many lab-trained models ineffective. 1

Despite recent advances in insulin preparations and technology, adjusting insulin remains an ongoing challenge for the majority of people with type 1 diabetes (T1D) and longstanding type 2 diabetes (T2D). In this study, we propose the Adaptive Basal-Bolus Advisor (ABBA), a personalised insulin treatment recommendation approach based on reinforcement learning for individuals with T1D and T2D, performing self-monitoring blood glucose measurements and multiple daily insulin injection therapy. We developed and evaluated the ability of ABBA to achieve better time-in-range (TIR) for individuals with T1D and T2D, compared to a standard basal-bolus advisor (BBA). The in-silico test was performed using an FDA-accepted population, including 101 simulated adults with T1D and 101 with T2D. An in-silico evaluation shows that ABBA significantly improved TIR and significantly reduced both times below- and above-range, compared to BBA. ABBA's performance continued to improve over two months, whereas BBA exhibited only modest changes. This personalised method for adjusting insulin has the potential to further optimise glycaemic control and support people with T1D and T2D in their daily self-management. Our results warrant ABBA to be trialed for the first time in humans.

Accurately estimating parameters of physiological models is essential to achieving reliable digital twins. For Type 1 Diabetes, this is particularly challenging due to the complexity of glucose-insulin interactions. Traditional methods based on Markov Chain Monte Carlo struggle with high-dimensional parameter spaces and fit parameters from scratch at inference time, making them slow and computationally expensive. In this study, we propose a Simulation-Based Inference approach based on Neural Posterior Estimation to efficiently capture the complex relationships between meal intake, insulin, and glucose level, providing faster, amortized inference. Our experiments demonstrate that SBI not only outperforms traditional methods in parameter estimation but also generalizes better to unseen conditions, offering real-time posterior inference with reliable uncertainty quantification.