Clinical trial analysis is one of the main business directions and services in IQVIA, and reviewing past similar studies is one of the most critical steps before starting a commercial clinical trial. The current review process is manual and time-consuming, requiring a clinical trial analyst to manually search through an extensive clinical trial database and then review all candidate studies. Therefore, it is of great interest to develop an automatic retrieval algorithm to select similar studies by giving new study information. To achieve this goal, we propose a novel group-based trial similarity learning network named GTSLNet, consisting of two kinds of similarity learning modules. The pair-wise section-level similarity learning module aims to compare the query trial and the candidate trial from the abstract semantic level via the proposed section transformer. Meanwhile, a word-level similarity learning module uses the word similarly matrix to capture the low-level similarity information. Additionally, an aggregation module combines these similarities. To address potential false negatives and noisy data, we introduce a variance-regularized group distance loss function. Experiment results show that the proposed GTSLNet significantly and consistently outperforms state-of-the-art baselines.

Predicting adverse drug reactions (ADRs) of drugs is one of the most critical steps in drug development. By pre-estimating the adverse reactions, researchers and drug development companies can greatly prevent the potential ADR risks and tragedies. However, the current ADR prediction methods suffer from several limitations. First, the prediction results are based on pure drug-related information, which makes them impossible to be directly applied for the personalized ADR prediction task. The lack of personalization of models also makes rare adverse events hard to be predicted. Therefore, it is of great interest to develop a new personalized ADR prediction method by introducing additional sources, e.g., patient health records. However, few methods have tried to use additional sources. In the meantime, the variety of different source formats and structures makes this task more challenging. To address the above challenges, we propose a novel personalized multi-sourced-based drug adverse reaction prediction model named pADR. pADR first works on every single source to transform them into proper representations. Next, a hierarchical multi-sourced Transformer is designed to automatically model the interactions between different sources and fuse them together for the final adverse event prediction. Experimental results on a new multi-sourced ADR prediction dataset show that PADR outperforms state-of-the-art drug-based baselines. Moreover, the case and ablation studies also illustrate the effectiveness of our proposed fusion strategies and the reasonableness of each module design.

Patient enrollment is critical to the success of a clinical trial. In practice, before launching a trial, one of the top priorities is to predict the enrollment rate for different countries, so that one can select clinical sites from the countries with the highest enrollment rates to accelerate patient recruitment. However, based on the limited trial information, estimating the enrollment rate is still a challenge. To deal with this problem, we adopt a very recent tensor factorization approach that aims to approximate the joint Cumulative Distribution Function (CDF) of trial enrollment data. We can always sketch a multivariate CDF in terms of multidimensional empirical cumulative probability array, i.e., a finite grid-sampled CDF tensor, and introduce a low-rank parametrization by a Canonical Polyadic Decomposition (CPD) model. The proposed model is unassuming of the structure of the data and identifiable under mild conditions by virtue of the uniqueness of CPD. At the same time, it affords both efficient sample likelihood estimation and closed-form inference. Such model can be leveraged for reliable enrollment estimation, delivering probability estimates of a specific trial meeting an expected enrollment rate or probability estimates of being within a certain interval, as well as country recommendation. Experimental results demonstrate an improved performance of the proposed method in the enrollment rate prediction task over the best baselines by up to 12.2% in mean squared error on a real country-level trial dataset, while also offering direct means of quantifying uncertainty in the predictions based on the fitted model. The improved performance and versatility highlight the societal and financial benefits of the proposed approach, which could be transformational in modern healthcare.

Low-rank tensor factorization or completion is well-studied and applied in various online settings, such as online tensor factorization (where the temporal mode grows) and online tensor completion (where incomplete slices arrive gradually). However, in many real-world settings, tensors may have more complex evolving patterns: (i) one or more modes can grow; (ii) missing entries may be filled; (iii) existing tensor elements can change. Existing methods cannot support such complex scenarios. To fill the gap, this paper proposes a Generalized Online Canonical Polyadic (CP) Tensor factorization and completion framework (named GOCPT) for this general setting, where we maintain the CP structure of such dynamic tensors during the evolution. We show that existing online tensor factorization and completion setups can be unified under the GOCPT framework. Furthermore, we propose a variant, named GOCPTE, to deal with cases where historical tensor elements are unavailable (e.g., privacy protection), which achieves similar fitness as GOCPT but with much less computational cost. Experimental results demonstrate that our GOCPT can improve fitness by up to 2.8% on the JHU Covid data and 9.2% on a proprietary patient claim dataset over baselines. Our variant GOCPTE shows up to 1.2% and 5.5% fitness improvement on two datasets with about 20% speedup compared to the best model.

Tensor decompositions are powerful tools for dimensionality reduction and feature interpretation of multidimensional data such as signals. Existing tensor decomposition objectives (e.g., Frobenius norm) are designed for fitting raw data under statistical assumptions, which may not align with downstream classification tasks. In practice, raw input tensor can contain irrelevant information while data augmentation techniques may be used to smooth out class-irrelevant noise in samples. This paper addresses the above challenges by proposing augmented tensor decomposition (ATD), which effectively incorporates data augmentations and self-supervised learning (SSL) to boost downstream classification. To address the non-convexity of the new augmented objective, we develop an iterative method that enables the optimization to follow an alternating least squares (ALS) fashion. We evaluate our proposed ATD on multiple datasets. It can achieve 0.8%~2.5% accuracy gain over tensor-based baselines. Also, our ATD model shows comparable or better performance (e.g., up to 15% in accuracy) over self-supervised and autoencoder baselines while using less than 5% of learnable parameters of these baseline models.

Existing tensor factorization methods assume that the input tensor follows some specific distribution (i.e. Poisson, Bernoulli, and Gaussian), and solve the factorization by minimizing some empirical loss functions defined based on the corresponding distribution. However, it suffers from several drawbacks: 1) In reality, the underlying distributions are complicated and unknown, making it infeasible to be approximated by a simple distribution. 2) The correlation across dimensions of the input tensor is not well utilized, leading to sub-optimal performance. Although heuristics were proposed to incorporate such correlation as side information under Gaussian distribution, they can not easily be generalized to other distributions. Thus, a more principled way of utilizing the correlation in tensor factorization models is still an open challenge. Without assuming any explicit distribution, we formulate the tensor factorization as an optimal transport problem with Wasserstein distance, which can handle non-negative inputs. We introduce SWIFT, which minimizes the Wasserstein distance that measures the distance between the input tensor and that of the reconstruction. In particular, we define the N-th order tensor Wasserstein loss for the widely used tensor CP factorization and derive the optimization algorithm that minimizes it. By leveraging sparsity structure and different equivalent formulations for optimizing computational efficiency, SWIFT is as scalable as other well-known CP algorithms. Using the factor matrices as features, SWIFT achieves up to 9.65% and 11.31% relative improvement over baselines for downstream prediction tasks. Under the noisy conditions, SWIFT achieves up to 15% and 17% relative improvements over the best competitors for the prediction tasks.

Accurate prediction of the transmission of epidemic diseases such as COVID-19 is crucial for implementing effective mitigation measures. In this work, we develop a tensor method to predict the evolution of epidemic trends for many regions simultaneously. We construct a 3-way spatio-temporal tensor (location, attribute, time) of case counts and propose a nonnegative tensor factorization with latent epidemiological model regularization named STELAR. Unlike standard tensor factorization methods which cannot predict slabs ahead, STELAR enables long-term prediction by incorporating latent temporal regularization through a system of discrete-time difference equations of a widely adopted epidemiological model. We use latent instead of location/attribute-level epidemiological dynamics to capture common epidemic profile sub-types and improve collaborative learning and prediction. We conduct experiments using both county- and state-level COVID-19 data and show that our model can identify interesting latent patterns of the epidemic. Finally, we evaluate the predictive ability of our method and show superior performance compared to the baselines, achieving up to 21% lower root mean square error and 25% lower mean absolute error for county-level prediction.

Real-world spatio-temporal data is often incomplete or inaccurate due to various data loading delays. For example, a location-disease-time tensor of case counts can have multiple delayed updates of recent temporal slices for some locations or diseases. Recovering such missing or noisy (under-reported) elements of the input tensor can be viewed as a generalized tensor completion problem. Existing tensor completion methods usually assume that i) missing elements are randomly distributed and ii) noise for each tensor element is i.i.d. zero-mean. Both assumptions can be violated for spatio-temporal tensor data. We often observe multiple versions of the input tensor with different under-reporting noise levels. The amount of noise can be time- or location-dependent as more updates are progressively introduced to the tensor. We model such dynamic data as a multi-version tensor with an extra tensor mode capturing the data updates. We propose a low-rank tensor model to predict the updates over time. We demonstrate that our method can accurately predict the ground-truth values of many real-world tensors. We obtain up to 27.2% lower root mean-squared-error compared to the best baseline method. Finally, we extend our method to track the tensor data over time, leading to significant computational savings.

Drug discovery aims at finding promising drug molecules for treating target diseases. Existing computational drug discovery methods mainly depend on molecule databases, ignoring valuable data collected from clinical trials. In this work, we propose PRIME to leverage high-quality drug molecules and drug-disease relations in historical clinical trials to narrow down the molecular search space in drug discovery. PRIME also introduces time dependency constraints to model evolving drug-disease relations using a probabilistic deep learning model that can quantify model uncertainty. We evaluated PRIME against leading models on both de novo design and drug repurposing tasks. Results show that compared with the best baselines, PRIME achieves 25.9% relative improvement (i.e., reduction) in average hit-ranking on drug repurposing and 47.6% relative improvement in success rate on de novo design.

Existing tensor completion formulation mostly relies on partial observations from a single tensor. However, tensors extracted from real-world data often are more complex due to: (i) Partial observation: Only a small subset of tensor elements are available. (ii) Coarse observation: Some tensor modes only present coarse and aggregated patterns (e.g., monthly summary instead of daily reports). In this paper, we are given a subset of the tensor and some aggregated/coarse observations (along one or more modes) and seek to recover the original fine-granular tensor with low-rank factorization. We formulate a coupled tensor completion problem and propose an efficient Multi-resolution Tensor Completion model (MTC) to solve the problem. Our MTC model explores tensor mode properties and leverages the hierarchy of resolutions to recursively initialize an optimization setup, and optimizes on the coupled system using alternating least squares. MTC ensures low computational and space complexity. We evaluate our model on two COVID-19 related spatio-temporal tensors. The experiments show that MTC could provide 65.20% and 75.79% percentage of fitness (PoF) in tensor completion with only 5% fine granular observations, which is 27.96% relative improvement over the best baseline. To evaluate the learned low-rank factors, we also design a tensor prediction task for daily and cumulative disease case predictions, where MTC achieves 50% in PoF and 30% relative improvements over the best baseline.

OBJECTIVE: The COVID-19 pandemic has created many challenges that need immediate attention. Various epidemiological and deep learning models have been developed to predict the COVID-19 outbreak, but all have limitations that affect the accuracy and robustness of the predictions. Our method aims at addressing these limitations and making earlier and more accurate pandemic outbreak predictions by (1) using patients' EHR data from different counties and states that encode local disease status and medical resource utilization condition; (2) considering demographic similarity and geographical proximity between locations; and (3) integrating pandemic transmission dynamics into deep learning models. MATERIALS AND METHODS: We proposed a spatio-temporal attention network (STAN) for pandemic prediction. It uses an attention-based graph convolutional network to capture geographical and temporal trends and predict the number of cases for a fixed number of days into the future. We also designed a physical law-based loss term for enhancing long-term prediction. STAN was tested using both massive real-world patient data and open source COVID-19 statistics provided by Johns Hopkins university across all U.S. counties. RESULTS: STAN outperforms epidemiological modeling methods such as SIR and SEIR and deep learning models on both long-term and short-term predictions, achieving up to 87% lower mean squared error compared to the best baseline prediction model. CONCLUSIONS: By using information from real-world patient data and geographical data, STAN can better capture the disease status and medical resource utilization information and thus provides more accurate pandemic modeling. With pandemic transmission law based regularization, STAN also achieves good long-term prediction performance.

Multi-user millimeter wave (mmWave) massive multi-input multi-output (MIMO) is a promising technology for the next generation mobile communication systems. However, there are still unsolved problems before such commercial MIMO networks are rolled out. One main issue is the hardware cost and power consumption which grow significantly as the number of radio frequency (RF) components increases. To tackle this issue, we consider to deploy one-bit analog-to-digital converters (ADCs) and digital-to-analog converters (DACs) at the base station (BS), and study uplink (UL)/downlink (DL) channel estimation and DL precoding techniques for the associated MIMO systems with one-bit ADCs/DACs. Specifically, we first formulate the UL channel estimation as an one-bit compressed sensing problem, and then devise an efficient gridless generalized approximate message passing-based (GL-GAMP) algorithm to handle it. Additionally, we develop an exhaustive search based proximal gradient descent method (PGM) for DL channel estimation. Note that with slight modifications, we show that PGM can also be applied to solve the DL precoding problem. Simulation results showcase that our methods have advantages over the state-of-the-art techniques and are able to offer good trade-offs between accuracy and computational complexity, which ultimately indicates their superiority in the application of mmWave MIMO systems with one-bit ADCs/DACs.

Source number detection is a critical problem in array signal processing. Conventional model-driven methods e.g., Akaikes information criterion and minimum description length, suffer from severe performance degradation when the number of samples is small or the signal-to-noise ratio is low. In this letter, we exploit the model-aided based deep neural network to estimate the source number. Specifically, we propose two eigenvalue based networks, i.e., a regression network (ERNet) and a classification network (ECNet), for source number detection, where the eigenvalues of the received signal covariance matrix and the source number are used as the input and the label of the networks, respectively. Furthermore, ERNet and ECNet can be easily generalized to handle coherent sources by adopting, e.g., the forward-backward spatial smoothing technique. Numerical results are included to showcase the remarkable improvements of ERNet and ECNet over the existing methods.

Blind estimation of parameters for M-ary frequency-shift-keying (M-FSK) signals is great of importance in intelligent receivers. Many existing algorithms have assumed white Gaussian noise. However, their performance severely degrades when grossly corrupted data, i.e., outliers, exist. This article solves this issue by developing a novel approach for parameter estimation of M-FSK signals in the presence of alpha-stable noise. Specifically, the proposed method exploits the generalized first- and second-order cyclostationarity of M-FSK signals with alpha-stable noise, which results in closed-form solutions for unknown parameters in both time and frequency domains. As a merit, it is computationally efficient and thus can be used for signal preprocessing, symbol timing estimation, signal and noise power estimation. Furthermore, substantial theoretical analysis on the performance of the proposed approach is provided. Simulations demonstrate that the proposed method is robust to alpha-stable noise and that it outperforms the state-of-the-art algorithms in many challenging scenarios.

We develop a direction of arrival (DoA) and channel estimation algorithm for the one-bit quantized millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) system. By formulating the estimation problem as a noisy one-bit compressed sensing problem, we propose a computationally efficient gridless solution based on the expectation-maximization generalized approximate message passing (EM-GAMP) approach. The proposed algorithm does not need the prior knowledge about the number of DoAs and outperforms the existing methods in distinguishing extremely close DoAs for the case of one-bit quantization. Both the DoAs and the channel coefficients are estimated for the case of one-bit quantization. The simulation results show that the proposed algorithm has effective estimation performances when the DoAs are very close to each other.

Predicting the response of cancer cells to drugs is an important problem in pharmacogenomics. Recent efforts in generation of large scale datasets profiling gene expression and drug sensitivity in cell lines have provided a unique opportunity to study this problem. However, one major challenge is the small number of samples (cell lines) compared to the number of features (genes) even in these large datasets. We propose a collaborative filtering like algorithm for modeling gene-drug relationship to identify patients most likely to benefit from a treatment. Due to the correlation of gene expressions in different cell lines, the gene expression matrix is approximately low-rank, which suggests that drug responses could be estimated from a reduced dimension latent space of the gene expression. Towards this end, we propose a joint low-rank matrix factorization and latent linear regression approach. Experiments with data from the Genomics of Drug Sensitivity in Cancer database are included to show that the proposed method can predict drug-gene associations better than the state-of-the-art methods.

This letter revisits the channel estimation problem for MIMO systems with one-bit analog-to-digital converters (ADCs) through a novel algorithm-Amplitude Retrieval (AR). Unlike the state-of-the-art methods such as those based on one-bit compressive sensing, AR takes a different approach. It accounts for the lost amplitudes of the one-bit quantized measurements, and performs channel estimation and amplitude completion jointly. This way, the direction information of the propagation paths can be estimated via accurate direction finding algorithms in array processing, e.g., maximum likelihood. The upsot is that AR is able to handle off-grid angles and provide more accurate channel estimates. Simulation results are included to showcase the advantages of AR.

In this letter, we consider the coherent direction-ofarrival estimation problem in impulsive noise. An ℓp-norm-based variant of the classical relaxation technique is proposed to tackle this problem. The proposed method successively minimizes the cost function along block coordinate directions. Specifically, at each iteration, only one block of the signal component is updated, while the remaining blocks are kept fixed. Then, instead of solving each block exactly, the proposed method optimizes the parameters in the block iteratively by solving a surrogate function that upper bounds the ℓp-norm. Numerical results show that the proposed scheme offers substantial performance improvement over the state-of-theart algorithms.

The 3GPP suggests to combine dual polarized (DP) antenna arrays with the double directional (DD) channel model for downlink channel estimation. This combination strikes a good balance between high-capacity communications and parsimonious channel modeling, and also brings limited feedback schemes for downlink channel estimation within reach. However, most existing channel estimation work under the DD model has not considered DP arrays, perhaps because of the complex array manifold and the resulting difficulty in algorithm design. In this paper, we first reveal that the DD channel with DP arrays at the transmitter and receiver can be naturally modeled as a low-rank four-way tensor, and thus the parameters can be effectively estimated via tensor decomposition algorithms. To reduce computational complexity, we show that the problem can be recast as a four-snapshot three-dimensional harmonic retrieval problem, which can be solved using computationally efficient subspace methods. On the theory side, we show that the DD channel with DP arrays is identifiable under very mild conditions, leveraging identifiability of low-rank tensors. Numerical simulations are employed to showcase the effectiveness of our methods.

In this letter, a manifold reconstruction based ESPRIT algorithm is devised for direction-of-arrival (DOA) estimation. Unlike the standard ESPRIT approach, the proposed method employs two smaller overlapping subarrays to construct the rotational invariance equation, which is then used to recover the array manifold matrix. This way, the DOA estimation can be split into K single-source DOA estimation problems, where each DOA is estimated by minimizing a nonlinear least squares fitting criterion that is solved via a computationally efficient Newton's method. Numerical results are included to validate the effectiveness of the proposed algorithm.

Phase retrieval has recently attracted renewed interest. It is revisited here through a new approach based on nonconvex quadratically constrained quadratic programming (QCQP). A least-squares (LS) formulation is adopted, and a recently developed non-convex QCQP approximation technique called feasible point pursuit (FPP) is tailored to obtain a new LS-FPP phase retrieval algorithm. The Cramér-Rao bound (CRB) is also derived for phase retrieval under additive white Gaussian noise. We demonstrate through simulations that the LS-FPP method outperforms the prior art and its mean square error approaches the CRB.

Joint estimation of the directions-of-arrival (DOAs) and frequencies of multiple signals is addressed in this paper. By constructing a set of joint diagonalization matrices, two cost functions that do not require a priori information of the source number are devised for DOA and frequency estimation in a separate manner. This enables us to estimate DOAs and frequencies via two one-dimensional search steps in their corresponding spatial and frequency domains. Thus, the tremendous two-dimensional search required in the standard approaches can be avoided. Simulation results demonstrate the effectiveness of the proposed approach.

In this work, the volume-based method for spectrum sensing is analyzed, which is able to provide the desirable properties of constant false-alarm rate, robustness against deviation from independent and identically distributed (IID) noise and being free of noise uncertainty. By computing the first and second moments for the signal-absence and signal-presence hypotheses together with using the Gamma distribution approximation, we derive accurate analytic formulae for the false-alarm and detection probabilities for IID noise situations. This enables us to develop theoretical decision threshold as well as receiver operating characteristic. Numerical results are presented to validate our theoretical findings.

A structured least squares based ESPRIT method is devised for joint direction-of-arrival and frequency estimation. By considering the errors in the estimated signal subspace and employing an iterative minimization procedure, the proposed approach is able to efficiently refine the estimated signal subspace, leading to significant enhancement in estimation performance. Simulation results demonstrate the effectiveness of the proposed approach.

A novel pseudo-noise resampling (PR) based unitary root-MUSIC algorithm for direction-of-arrival (DOA) estimation is derived in this letter. Our solution is able to eliminate the abnormal DOA estimator called outlier and obtain an approximate outlier-free performance in the unitary root-MUSIC algorithm. In particular, we utilize a hypothesis test to detect the outlier. Meanwhile, a PR process is applied to form a DOA estimator bank and a corresponding root estimator bank. We propose a distance detection strategy which exploits the information contained in the estimated root estimator to help determine the final DOA estimates when all the DOA estimators fail to pass the reliability test. Furthermore, the proposed method is realized in terms of real-valued computations, leading to an efficient implementation. Simulations show that the improved MUSIC scheme can significantly improve the DOA resolution at low signal-to-noise ratios and small samples.