Surgical assessment: measuring unobserved health

Background

The federal and provincial governments in Canada have invested an enormous amount of resources to measure, report and reduce surgical wait times. Yet these measures under-report the wait period that patients’ actually experience, because they do not capture the length of time a patient spends waiting to see the surgeon for a surgical assessment. This unmeasured time is referred to as the “wait one” (W1). Little is known about W1 and the effects that this has on patients’ health. Similarly, it is not understood whether patients waiting for surgical assessment actually want or need surgery. Existing administrative and clinical dataset do not capture information on health and decision-making while the patient is waiting for care form a specialist. The objective of this proposed study is to understand the impact that W1 for elective surgeries has on the health of patients and to determine whether this time can be reduced.

Methods/Design

A prospective survey design will be used to measure the health of patients waiting for surgical assessment. Working with the support of the surgical specialities in Vancouver Coastal Health, we will survey patients immediately after being referred for surgical assessment, and every four months thereafter, until they are seen by the surgeon.Validated survey instruments will be used, including: generic and condition-specific health status questionnaires, pain and depression assessments. Other factors that will be measured include: patients’ knowledge about their condition, and their desired autonomy in the decision making process. We have piloted data collection in one surgical specialty in order to demonstrate feasibility.

Discussion

The results from this study will be used to quantify changes in patients’ health while they wait for surgical assessment. Based on this, policy- and decision-makers could design care interventions during W1, aimed at mitigating any negative health consequences associated with waiting. The results from this study will also be used to better understand whether there are factors that predict patients’ desire to proceed to surgery. These could be used to guide future research into experimenting with interventions to minimize inappropriate referrals and where they are best targeted.     

Usefulness of transthoracic echocardiography in demonstrating coronary blood flow after coronary artery bypass grafting

In this pilot study, we used transthoracic echocardiography to measure coronary sinus blood flow in 15 patients before and after coronary artery bypass grafting. Coronary sinus blood flow before and after revascularization was 274 +/- 95 and 451 +/- 102 ml/min, respectively (p <0.001). This finding is consistent with those previously demonstrated by invasive techniques.     

Alveolar hypoventilation syndrome in brainstem glioma with improvement after surgical resection

A 3-year-old boy presented with brainstem astrocytoma and central alveolar hypoventilation syndrome. Contrast MRI of the brain showed that the tumor involved the cerebellum, with compression of brainstem, and resolved after surgical resection. Polysomnography performed before and after total tumor resection showed significant improvement in nocturnal respiratory rate, respiratory disturbance index, and oxygen desaturation. It is apparent that central alveolar hypoventilation syndrome secondary to brainstem tumor may improve after surgical resection for those with favorable anatomic location and histology. Serial polysomnography and MRI scans are useful for diagnosis and in the management plan, and to monitor progress.     

Density estimation using deep generative neural networks

Density estimation is one of the fundamental problems in both statistics and machine learning. In this study, we propose Roundtrip, a computational framework for general-purpose density estimation based on deep generative neural networks. Roundtrip retains the generative power of deep generative models, such as generative adversarial networks (GANs) while it also provides estimates of density values, thus supporting both data generation and density estimation. Unlike previous neural density estimators that put stringent conditions on the transformation from the latent space to the data space, Roundtrip enables the use of much more general mappings where target density is modeled by learning a manifold induced from a base density (e.g., Gaussian distribution). Roundtrip provides a statistical framework for GAN models where an explicit evaluation of density values is feasible. In numerical experiments, Roundtrip exceeds state-of-the-art performance in a diverse range of density estimation tasks.

Let p(·) be a density on a n-dimensional Euclidean space χ. The task of density estimation is to estimate p(·) based on a set of independently and identically distributed data points {xi}i=1Ndrawn from this density.Traditional density estimators such as histograms (1, 2) and kernel density estimators (KDEs) (3, 4) typically perform well only in low dimension. Recently, neural network-based approaches were proposed for density estimation and yielded promising results in problems with high-dimensional data points such as images. There are mainly two families of such neural density estimators: autoregressive models (5–7) and normalizing flows (8–11). Autoregression-based neural density estimators decompose the density into the product of conditional densities based on probability chain rule p(x)=∏ip(xi|x1:i−1). Each conditional probability p(xi|x1:i−1) is modeled by a parametric density (e.g., Gaussian or mixture of Gaussian), of which the parameters are learned by neural networks. Density estimators based on normalizing flows represent x as an invertible transformation of a latent variable z with known density, where the invertible transformation is a composition of a series of simple functions whose Jacobian is easy to compute. The parameters of these component functions are then learned by neural networks.As suggested in ref. 12, both of these are special cases of the following general framework. Given a differentiable and invertible mapping G:Rn→Rn and a base density pzz, the density of x=G(z) can be represented using the change of variable rule as follows:pxx=pzz|detJz|−1,[1]where Jz=∂Gz/∂zT is the Jacobian matrix of function G(·) at point z. Density estimation at x can be solved if the base density pzz is known and the determinant of Jacobian matrix is feasible to calculate. To achieve this, previous neural density estimators have to impose heavy constraints on the model architecture. For example, refs. 7, 10, and 12 require the Jacobian to be triangular, ref. 13 constructed low rank perturbations of a diagonal matrix as the Jacobian, and ref. 14 proposed a circular convolution where the Jacobian is a circulant matrix. These strong constraints diminish the expressiveness of neural networks, which may lead to poor performance. For example, autoregressive neural density estimators based on learning p(xi|x1:i−1) are naturally sensitive to the order of the features. Moreover, the change of variable rule is not applicable when the domain dimension in base density differs from target density. However, experiences from deep generative models [e.g., GAN (15) and VAE (16)] suggested that it is often desirable to use a latent space of smaller dimension than the data space.To overcome the limitations above, we propose a neural density estimator called Roundtrip. Our approach is motivated by recent advances in deep generative neural networks (15, 17, 18). Roundtrip differs from previous neural density estimators in two ways. 1) It allows the direct use of a deep generative network to model the transformation from the latent variable space to the data space, while previous neural density estimators use neural networks only to learn the parameters in the component functions that are used for building up an invertible transformation. 2) It can efficiently model data densities that are concentrated near learned manifolds, which is difficult to achieve by previous approaches as they require the latent space to have the same dimension as the data space. Importantly, we also provide methods, based on either importance sampling and Laplace approximation, for the pointwise evaluation of the density estimate. We summarize our major contributions in this study as follows: 1) We propose a general-purpose neural density estimator based on deep generative models, which requires less restrictive model assumptions compared to previous neural density estimators. 2) We show that the principle in previous neural density estimators can be regarded as a special case in our Roundtrip framework. 3) We demonstrate state-of-the-art performance of Roundtrip model through a series of experiments, including density estimation tasks in simulations as well as in real data applications ranging from image generation to outlier detection.     

A propeller‐like small molecule as a novel G‐quadruplex DNA binder: The study of fluorescent sensing property and preferential interactions with human telo21 structure

A new propeller‐like small molecule was synthesized with three terminal amino side groups. The molecule was found to be a selective nucleic acid binder towards telo21 G‐quadruplex DNA compared with other representative nucleic acids including single‐stranded DNA (dA21), duplex DNA (ds26) and RNA. The fluorescent signal of the molecule upon interaction with telo21 G‐quadruplex structure shows remarkable enhancement (F_max/F₀ = 17.9) while interaction with other nucleic acids shows the signal enhancement which is less than 2.1. In addition, a good linear relationship of binding signal correlated with the concentration of telo21 DNA was obtained. Molecular docking study was also performed to acquire the binding behaviour and its interaction modes of the molecule with the structure of human telomeric DNA G‐quadruplex. The modelling results show that the three conjugated terminal units (dimethylaminobenzyl groups) associated through the ethylene bridges with the central methylated pyridine ring formed a co‐planar conformation upon stacking onto the G‐quartets via pi‐pi stacking interactions. This could be the key reason that the molecule shows excellent fluorescent signal of binding towards telo21 G‐quadruplex DNA rather than other types of nucleic acids.     

A multicentre phase II study of vorinostat in patients with relapsed or refractory indolent B‐cell non‐Hodgkin lymphoma and mantle cell lymphoma

Although initial rituximab‐containing chemotherapies achieve high response rates, indolent B‐cell non‐Hodgkin lymphoma (B‐NHL), such as follicular lymphoma (FL), is still incurable. Therefore, new effective agents with novel mechanisms are anticipated. In this multicentre phase II study, patients with relapsed/refractory indolent B‐NHL and mantle cell lymphoma (MCL) received vorinostat 200 mg twice daily for 14 consecutive days in a 21‐d cycle until disease progression or unacceptable toxicity occurred. The primary endpoint was overall response rate (ORR) in FL patients and safety and tolerability in all patients. Secondary endpoints included progression‐free survival (PFS). Fifty‐six eligible patients were enrolled; 50 patients (39 with FL, seven with other B‐NHL, and four with MCL) were evaluable for ORR, and 40 patients had received rituximab‐containing prior chemotherapeutic regimens. For the 39 patients with FL, the ORR was 49% [95% confidence interval (CI): 32·4, 65·2] and the median PFS was 20 months (95% CI: 11·2, 29·7). Major toxicities were manageable grade 3/4 thrombocytopenia and neutropenia. Vorinostat offers sustained antitumour activity in patients with relapsed or refractory FL with an acceptable safety profile. Further investigation of vorinostat for clinical efficacy is warranted.