اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدLSTMs and Deep Residual Networks for Carbohydrate and Bolus Recommendations in Type 1 Diabetes Management
89
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
To avoid serious diabetic complications, people with type 1 diabetes must keep their blood glucose levels (BGLs) as close to normal as possible. Insulin dosages and carbohydrate consumption are important considerations in managing BGLs. Since the 1960s, models have been developed to forecast blood glucose levels based on the history of BGLs, insulin dosages, carbohydrate intake, and other physiological and lifestyle factors. Such predictions can be used to alert people of impending unsafe BGLs or to control insulin flow in an artificial pancreas. In past work, we have introduced an LSTM-based approach to blood glucose level prediction aimed at what if scenarios, in which people could enter foods they might eat or insulin amounts they might take and then see the effect on future BGLs. In this work, we invert the what-if scenario and introduce a similar architecture based on chaining two LSTMs that can be trained to make either insulin or carbohydrate recommendations aimed at reaching a desired BG level in the future. Leveraging a recent state-of-the-art model for time series forecasting, we then derive a novel architecture for the same recommendation task, in which the two LSTM chain is used as a repeating block inside a deep residual architecture. Experimental evaluations using real patient data from the OhioT1DM dataset show that the new integrated architecture compares favorably with the previous LSTM-based approach, substantially outperforming the baselines. The promising results suggest that this novel approach could potentially be of practical use to people with type 1 diabetes for self-management of BGLs.
قيم البحث
اقرأ أيضاً
Prediction of diabetes and its various complications has been studied in a number of settings, but a comprehensive overview of problem setting for diabetes prediction and care management has not been addressed in the literature. In this document we s
eek to remedy this omission in literature with an encompassing overview of diabetes complication prediction as well as situating this problem in the context of real world healthcare management. We illustrate various problems encountered in real world clinical scenarios via our own experience with building and deploying such models. In this manuscript we illustrate a Machine Learning (ML) framework for addressing the problem of predicting Type 2 Diabetes Mellitus (T2DM) together with a solution for risk stratification, intervention and management. These ML models align with how physicians think about disease management and mitigation, which comprises these four steps: Identify, Stratify, Engage, Measure.
Visual and audiovisual speech recognition are witnessing a renaissance which is largely due to the advent of deep learning methods. In this paper, we present a deep learning architecture for lipreading and audiovisual word recognition, which combines
Residual Networks equipped with spatiotemporal input layers and Bidirectional LSTMs. The lipreading architecture attains 11.92% misclassification rate on the challenging Lipreading-In-The-Wild database, which is composed of excerpts from BBC-TV, each containing one of the 500 target words. Audiovisual experiments are performed using both intermediate and late integration, as well as several types and levels of environmental noise, and notable improvements over the audio-only network are reported, even in the case of clean speech. A further analysis on the utility of target word boundaries is provided, as well as on the capacity of the network in modeling the linguistic context of the target word. Finally, we examine difficult word pairs and discuss how visual information helps towards attaining higher recognition accuracy.
Diabetes is a major public health challenge worldwide. Abnormal physiology in diabetes, particularly hypoglycemia, can cause driver impairments that affect safe driving. While diabetes driver safety has been previously researched, few studies link re
al-time physiologic changes in drivers with diabetes to objective real-world driver safety, particularly at high-risk areas like intersections. To address this, we investigated the role of acute physiologic changes in drivers with type 1 diabetes mellitus (T1DM) on safe stopping at stop intersections. 18 T1DM drivers (21-52 years, mean = 31.2 years) and 14 controls (21-55 years, mean = 33.4 years) participated in a 4-week naturalistic driving study. At induction, each participants vehicle was fitted with a camera and sensor system to collect driving data. Video was processed with computer vision algorithms detecting traffic elements. Stop intersections were geolocated with clustering methods, state intersection databases, and manual review. Videos showing driver stop intersection approaches were extracted and manually reviewed to classify stopping behavior (full, rolling, and no stop) and intersection traffic characteristics. Mixed-effects logistic regression models determined how diabetes driver stopping safety (safe vs. unsafe stop) was affected by 1) disease and 2) at-risk, acute physiology (hypo- and hyperglycemia). Diabetes drivers who were acutely hyperglycemic had 2.37 increased odds of unsafe stopping (95% CI: 1.26-4.47, p = 0.008) compared to those with normal physiology. Acute hypoglycemia did not associate with unsafe stopping (p = 0.537), however the lower frequency of hypoglycemia (vs. hyperglycemia) warrants a larger sample of drivers to investigate this effect. Critically, presence of diabetes alone did not associate with unsafe stopping, underscoring the need to evaluate driver physiology in licensing guidelines.
Deep residual network architectures have been shown to achieve superior accuracy over classical feed-forward networks, yet their success is still not fully understood. Focusing on massively over-parameterized, fully connected residual networks with R
eLU activation through their respective neural tangent kernels (ResNTK), we provide here a spectral analysis of these kernels. Specifically, we show that, much like NTK for fully connected networks (FC-NTK), for input distributed uniformly on the hypersphere $mathbb{S}^{d-1}$, the eigenfunctions of ResNTK are the spherical harmonics and the eigenvalues decay polynomially with frequency $k$ as $k^{-d}$. These in turn imply that the set of functions in their Reproducing Kernel Hilbert Space are identical to those of FC-NTK, and consequently also to those of the Laplace kernel. We further show, by drawing on the analogy to the Laplace kernel, that depending on the choice of a hyper-parameter that balances between the skip and residual connections ResNTK can either become spiky with depth, as with FC-NTK, or maintain a stable shape.
Machine unlearning has great significance in guaranteeing model security and protecting user privacy. Additionally, many legal provisions clearly stipulate that users have the right to demand model providers to delete their own data from training set
, that is, the right to be forgotten. The naive way of unlearning data is to retrain the model without it from scratch, which becomes extremely time and resource consuming at the modern scale of deep neural networks. Other unlearning approaches by refactoring model or training data struggle to gain a balance between overhead and model usability. In this paper, we propose an approach, dubbed as DeepObliviate, to implement machine unlearning efficiently, without modifying the normal training mode. Our approach improves the original training process by storing intermediate models on the hard disk. Given a data point to unlearn, we first quantify its temporal residual memory left in stored models. The influenced models will be retrained and we decide when to terminate the retraining based on the trend of residual memory on-the-fly. Last, we stitch an unlearned model by combining the retrained models and uninfluenced models. We extensively evaluate our approach on five datasets and deep learning models. Compared to the method of retraining from scratch, our approach can achieve 99.0%, 95.0%, 91.9%, 96.7%, 74.1% accuracy rates and 66.7$times$, 75.0$times$, 33.3$times$, 29.4$times$, 13.7$times$ speedups on the MNIST, SVHN, CIFAR-10, Purchase, and ImageNet datasets, respectively. Compared to the state-of-the-art unlearning approach, we improve 5.8% accuracy, 32.5$times$ prediction speedup, and reach a comparable retrain speedup under identical settings on average on these datasets. Additionally, DeepObliviate can also pass the backdoor-based unlearning verification.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
