Cross-Global Attention Graph Kernel Network Prediction of Drug Prescription

Yao, Hao-Ren; Chang, Der–Chen; Huang, Wei; Liang, I-Chia; Hung, Chi‐Feng

doi:10.1145/3388440.3412459

Cited by 4 publications

(23 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our previous efforts [5,12,15] present a graph kernel-based system for outcome prediction of drug prescription, particularly the success or failure treatment, on short-term and chronic diseases. In [5], a Multiple Graph Kernel Fusion (MGKF) was proposed to overcome noise effect on short-term disease.…”

Section: Preliminariesmentioning

confidence: 99%

“…In [5], a Multiple Graph Kernel Fusion (MGKF) was proposed to overcome noise effect on short-term disease. A deep graph kernel learning approach, e.g., Cross-Global Attention Graph Kernel Network (Cross-Global), is proposed in [12] to handle long-term chronic disease. In short, we initially determine success and failure patients for the target disease treatment as training data within a user-defined time quantum, where a set of medical events are extracted between this time period.…”

Section: Preliminariesmentioning

confidence: 99%

“…It also motivated the deep kernel learning [10,11]. In [12], a deep metric learning based graph kernel is proposed to solve the outcome prediction problem in complex chronic disease treatment planning, where the cosine distance was shown superior to the Euclidean distance measure.…”

Section: Introductionmentioning

confidence: 99%

“…In [15], a framework was proposed to predict success and failure outcomes of a given drug prescription for antibiotic treatment-based disease. The approach is further extended to overcome biased data distribution [5] and chronic disease treatment plan [12]. Moreover, Euclidean and cosine distance measures exhibit different performance behaviors.…”

Section: Introductionmentioning

confidence: 99%

“…In [12], the cosine distance was found superior to the Euclidean measure under highly data imbalanced chronic disease, however, the Euclidean distance was used for short-term disease in [5]. To further investigate such differences, we establish a unified framework, which integrates two model structures proposed in the aforementioned works, to conduct rigorous empirical evaluation on all diseases investigated in previous studies [5,12], in addition to a theoretical discussion. The aforementioned prescription efficacy prediction approaches are now under commercial licensing.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The analysis from nonlinear distance metric to kernel-based prescription prediction system

2021

J. Nonlinear Var. Anal.

View full text Add to dashboard Cite

The distance metric and its nonlinear variant play a substantial role in machine learning, particularly yoso in building kernel functions. Often, the Euclidean distance with a radial basis function (RBF) is used to construct a RBF kernel for nonlinear classification. However, domain implications periodically constrain the distance metrics. Specifically, within the domain of drug efficacy prediction, distance measures must account for time that varies based on disease duration, short to chronic. Recently, a distance-derived graph kernel approach was commercially licensed for drug prescription efficacy prediction. The analysis of the distance functions used therein, namely the Euclidean and cosine distance measures and their respective derived graph kernels, is provided. Theoretically, we provide a formulation of our efforts and demonstrate how both the Euclidean and cosine distance induce space and discuss the difference from geometric perspectives. The aforementioned approach is likewise empirically evaluated using a million-plus patient subset of a life-spanning, real-world, electronic health record database. Diseases are characterized as either short in duration or chronic and either common, hence balanced data, or relatively rare, hence imbalanced. Empirically, the system accurately predicted the efficacy of prescriptions for both balanced and imbalanced and short-term and chronic diseases, with at least one of the measures used being statistically significantly superior to conventional prediction methods. Succinctly, for short-term, balanced diseases, the Euclidean and cosine measures were generally statistically equivalent. For short-term, imbalanced diseases however, the Euclidean measure was superior to the cosine measure, at times and not infrequently, statistically significantly so. For chronic, balanced diseases, Euclidean was slightly superior to the cosine measure, but they were statistically equivalent. In contrast, for chronic, imbalanced diseases, the cosine measure was consistently statistically significantly superior to the Euclidean measure. These findings indicate the need for both measures depending on the use case. Our empirical findings match our theoretical underpinnings.

show abstract

Section: Preliminariesmentioning

confidence: 99%

Section: Preliminariesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The analysis from nonlinear distance metric to kernel-based prescription prediction system

2021

J. Nonlinear Var. Anal.

View full text Add to dashboard Cite

show abstract

Self-Supervised Representation Learning on Electronic Health Records with Graph Kernel Infomax

Yao,

Cao,

Russell

et al. 2024

ACM Trans. Comput. Healthcare

View full text Add to dashboard Cite

Learning Electronic Health Records (EHRs) representation is a preeminent yet under-discovered research topic. It benefits various clinical decision support applications, e.g., medication outcome prediction or patient similarity search. Current approaches focus on task-specific label supervision on vectorized sequential EHR, which is not applicable to large-scale unsupervised scenarios. Recently, contrastive learning has shown great success in self-supervised representation learning problems. However, complex temporality often degrades the performance. We propose Graph Kernel Infomax, a self-supervised graph kernel learning approach on the graphical representation of EHR, to overcome the previous problems. Unlike the state-of-the-art, we do not change the graph structure to construct augmented views. Instead, we use Kernel Subspace Augmentation to embed nodes into two geometrically different manifold views. The entire framework is trained by contrasting nodes and graph representations on those two manifold views through the commonly used contrastive objectives. Empirically, using publicly available benchmark EHR datasets, our approach yields performance on clinical downstream tasks that exceeds the state-of-the-art. Theoretically, the variation in distance metrics naturally creates different views as data augmentation without changing graph structures. Practically, our method is non-ad hoc and confirms superior performance on commonly used graph benchmark datasets.

show abstract

Heat kernels on unit spheres and applications to graph kernels

2023

J. Nonlinear Var. Anal.

View full text Add to dashboard Cite

It is known that many statistical and machine learning approaches heavily rely on pairwise distance between data points. The choice of distance function on the underlying manifold has a fundamental impact on performance of these processes. This is closely related to questions of how to appropriately calculate distances, and hence, fundamental solutions (heat kernels) for heat operators can be obtained. In general, it is not so easy to obtain a closed form for heat kernels. We first survey results of heat kernels on radially symmetric Riemannian manifolds, e.g., Euclidean spaces and unit spheres in R n . For the cases n = 1, 2, 3, we may construct the heat kernel explicitly. But, the computation is much more complicated when n > 3. However, by results of Nagase, we may construct parametrices for the heat kernel by using elementary functions so that the error terms can be under controlled. In the second part of the paper, we discuss some results on subRiemannian manifolds, especially 3-dimensional sphere in C 2 as a CR-manifold. We study geodesics connecting two given points on S 3 respecting the Hopf fibration. This geodesic boundary value problem is completely solved in the case of S 3 and some partial results are obtained in the general case. The Carnot-Carathéodory distance is calculated. We also present some motivations related to quantum mechanics. Then we give a brief discussion of Greiner's methods on the heat kernel for the Cauchy-Riemann subLaplacian on S 2n+1 . We provide a brief discussion on applications of these heat kernels to graph kernels in the last part of the paper.

show abstract

Cross-Global Attention Graph Kernel Network Prediction of Drug Prescription

Cited by 4 publications

References 33 publications

The analysis from nonlinear distance metric to kernel-based prescription prediction system

The analysis from nonlinear distance metric to kernel-based prescription prediction system

Self-Supervised Representation Learning on Electronic Health Records with Graph Kernel Infomax

Heat kernels on unit spheres and applications to graph kernels

Contact Info

Product

Resources

About