Perfusion drugs for non­muscle invasive bladder cancer (Review).

The high recurrence rate and poor prognosis of non-muscle invasive bladder cancer (BC) are challenges that need to be urgently addressed. Transurethral cystectomy for bladder tumors is often combined with bladder perfusion therapy, which can effectively reduce the recurrence and progression rates of BC. The present review integrated and analyzed currently available bladder perfusion drugs, mainly including chemotherapeutic agents, immunotherapeutic agents and other adjuvant perfusion drugs. Bacillus Calmette-Guerin (BCG) perfusion was the pioneering immunotherapy for early BC and still ranks high in the selection of perfusion drugs. However, BCG infusion has a high toxicity profile and has been shown to be ineffective in some patients. Due to the limitations of BCG, new bladder perfusion drugs are constantly being developed. Immunotherapeutic agents have opened a whole new chapter in the selection of therapeutic agents for bladder perfusion. The present review explored the mechanism of action, clinical dosage and adverse effects of a variety of bladder perfusion drugs currently in common use, described combined perfusion and compared the effects of certain drugs on BC.

Prediction model of contact forces and IOP during digital palpation of porcine eyes.

Frequent intraocular pressure (IOP) measurements are desirable in the diagnosis and management of glaucoma. Most current tonometers utilize some form of corneal deformation to estimate the IOP, since trans-scleral tonometry suffers from loss of sensitivity. Tran-scleral and trans-palpebral tonometry, however, offer a pathway towards a non-invasive home tonometry. This article presents a mathematical model capturing the relationship between the IOP and the displacements imposed onto the sclera by externally applied forces. Similar to manual digital palpation tonometry, trans-scleral mechanical palpation makes use of two force probes that are advanced in a specific order and distance. Data from the applied forces and displacements, along with concurrent measurements of IOP is used to produce a phenomenological mathematical model. The experiments were carried out on enucleated porcine eyes. Two models are presented. Model 1 predicts IOP vs forces and displacements, while Model 2 predicts the baseline IOP (prior to applying the forces) as a function of the measured forces and displacements. The proposed models result in IOP errors of 1.65 mmHG and 0.82 mmHg, respectively. Model parameters were extracted using least-squares-based system identification methods. The results show that the proposed models can be used to estimate the baseline IOP with accuracy of ±1 mmHg over a pressure range of 10-35 mmHg, solely from measurement of tactile forces and displacements.

Communication Efficient Federated Generalized Tensor Factorization for Collaborative Health Data Analytics.

Modern healthcare systems knitted by a web of entities (e.g., hospitals, clinics, pharmacy companies) are collecting a huge volume of healthcare data from a large number of individuals with various medical procedures, medications, diagnosis, and lab tests. To extract meaningful medical concepts (i.e., phenotypes) from such higher-arity relational healthcare data, tensor factorization has been proven to be an effective approach and received increasing research attention, due to their intrinsic capability to represent the high-dimensional data. Recently, federated learning offers a privacy-preserving paradigm for collaborative learning among different entities, which seemingly provides an ideal potential to further enhance the tensor factorization-based collaborative phenotyping to handle sensitive personal health data. However, existing attempts to federated tensor factorization come with various limitations, including restrictions to the classic tensor factorization, high communication cost and reduced accuracy. We propose a communication efficient federated generalized tensor factorization, which is flexible enough to choose from a variate of losses to best suit different types of data in practice. We design a three-level communication reduction strategy tailored to the generalized tensor factorization, which is able to reduce the uplink communication cost up to 99.90%. In addition, we theoretically prove that our algorithm does not compromise convergence speed despite the aggressive communication compression. Extensive experiments on two real-world electronics health record datasets demonstrate the efficiency improvements in terms of computation and communication cost.

Temporal Network Embedding via Tensor Factorization.

Representation learning on static graph-structured data has shown a significant impact on many real-world applications. However, less attention has been paid to the evolving nature of temporal networks, in which the edges are often changing over time. The embeddings of such temporal networks should encode both graph-structured information and the temporally evolving pattern. Existing approaches in learning temporally evolving network representations fail to capture the temporal interdependence. In this paper, we propose Toffee, a novel approach for temporal network representation learning based on tensor decomposition. Our method exploits the tensor-tensor product operator to encode the cross-time information, so that the periodic changes in the evolving networks can be captured. Experimental results demonstrate that Toffee outperforms existing methods on multiple real-world temporal networks in generating effective embeddings for the link prediction tasks.

Communication Efficient Tensor Factorization for Decentralized Healthcare Networks.

Tensor factorization has been proved as an efficient unsupervised learning approach for health data analysis, especially for computational phenotyping, where the high-dimensional Electronic Health Records (EHRs) with patients history of medical procedures, medications, diagnosis, lab tests, etc., are converted to meaningful and interpretable medical concepts. Federated tensor factorization distributes the tensor computation to multiple workers under the coordination of a central server, which enables jointly learning the phenotypes across multiple hospitals while preserving the privacy of the patient information. However, existing federated tensor factorization algorithms encounter the single-point-failure issue with the involvement of the central server, which is not only easily exposed to external attacks, but also limits the number of clients sharing information with the server under restricted uplink bandwidth. In this paper, we propose CiderTF, a communication-efficient decentralized generalized tensor factorization, which reduces the uplink communication cost by leveraging a four-level communication reduction strategy designed for a generalized tensor factorization, which has the flexibility of modeling different tensor distribution with multiple kinds of loss functions. Experiments on two real-world EHR datasets demonstrate that CiderTF achieves comparable convergence with the communication reduction up to 99.99%.

Generating sequential electronic health records using dual adversarial autoencoder.

OBJECTIVE: Recent studies on electronic health records (EHRs) started to learn deep generative models and synthesize a huge amount of realistic records, in order to address significant privacy issues surrounding the EHR. However, most of them only focus on structured records about patients' independent visits, rather than on chronological clinical records. In this article, we aim to learn and synthesize realistic sequences of EHRs based on the generative autoencoder. MATERIALS AND METHODS: We propose a dual adversarial autoencoder (DAAE), which learns set-valued sequences of medical entities, by combining a recurrent autoencoder with 2 generative adversarial networks (GANs). DAAE improves the mode coverage and quality of generated sequences by adversarially learning both the continuous latent distribution and the discrete data distribution. Using the MIMIC-III (Medical Information Mart for Intensive Care-III) and UT Physicians clinical databases, we evaluated the performances of DAAE in terms of predictive modeling, plausibility, and privacy preservation. RESULTS: Our generated sequences of EHRs showed the comparable performances to real data for a predictive modeling task, and achieved the best score in plausibility evaluation conducted by medical experts among all baseline models. In addition, differentially private optimization of our model enables to generate synthetic sequences without increasing the privacy leakage of patients' data. CONCLUSIONS: DAAE can effectively synthesize sequential EHRs by addressing its main challenges: the synthetic records should be realistic enough not to be distinguished from the real records, and they should cover all the training patients to reproduce the performance of specific downstream tasks.

Predict or draw blood: An integrated method to reduce lab tests.

Serial laboratory testing is common, especially in Intensive Care Units (ICU). Such repeated testing is expensive and may even harm patients. However, identifying specific tests that can be omitted is challenging. The search space of different lab tests is large and the optimal reduction is hard to determine without modeling the time trajectory of decisions, which is a nontrivial optimization problem. In this paper, we propose a novel deep-learning method with a very concise architecture to jointly predict future lab test events to be omitted and the values of the omitted events based on observed testing values. Using our method, we were able to omit 15% of lab tests with <5% prediction accuracy loss. Although the application is specific to repeated lab tests, our proposed framework is highly generalizable and can be used to tackle a family of similar business decision making problems.

Privacy-Preserving Tensor Factorization for Collaborative Health Data Analysis.

Tensor factorization has been demonstrated as an efficient approach for computational phenotyping, where massive electronic health records (EHRs) are converted to concise and meaningful clinical concepts. While distributing the tensor factorization tasks to local sites can avoid direct data sharing, it still requires the exchange of intermediary results which could reveal sensitive patient information. Therefore, the challenge is how to jointly decompose the tensor under rigorous and principled privacy constraints, while still support the model's interpretability. We propose DPFact, a privacy-preserving collaborative tensor factorization method for computational phenotyping using EHR. It embeds advanced privacy-preserving mechanisms with collaborative learning. Hospitals can keep their EHR database private but also collaboratively learn meaningful clinical concepts by sharing differentially private intermediary results. Moreover, DPFact solves the heterogeneous patient population using a structured sparsity term. In our framework, each hospital decomposes its local tensors and sends the updated intermediary results with output perturbation every several iterations to a semi-trusted server which generates the phenotypes. The evaluation on both real-world and synthetic datasets demonstrated that under strict privacy constraints, our method is more accurate and communication-efficient than state-of-the-art baseline methods.

[Characteristics of aerosol water-soluble inorganic ions in three types air-pollution incidents of Nanjing City].

In order to compare aerosol water-soluble inorganic species in different air-pollution periods, samples of PM10, PM2.1, PM1.1 and the main water-soluble ions (NH4+, Mg2+, Ca2+, Na+, K+, NO2(-), F(-), NO3(-), Cl(-), SO4(2-)) were measured, which were from 3 air-pollution incidents (continued pollution in October 16-30 of 2009, sandstorm pollution in April 27-30 of 2010, and crop burning pollution in June 14 of 2010. The results show that aerosol pollution of 3 periods is serious. The lowest PM2.1/PM10 is only 0.27, which is from sandstorm pollution period, while the largest is 0. 7 from crop burning pollution period. In continued pollution periods, NO3(-) and SO4(2-) are the dominant ions, and the total anions account for an average of 18.62%, 32.92% and 33.53% of PM10, PM2.1 and PM1.1. Total water-soluble ions only account for 13.36%, 23.72% and 28.54% of PM10, PM2.1 and PM1.1 due to the insoluble species is increased in sandstorm pollution period. The mass concentration of Ca2+ in sandstorm pollution period is higher than the other two pollution periods, and which is mainly in coarse particles with diameter larger than 1 microm. All the ten water-soluble ions are much higher in crop burning pollution especially K+ which is the tracer from crop burning. The peak mass concentrations of NO3(-), SO4(2-) and NH4+ are in 0.43-0.65 microm.