Search

Although response-adaptive randomisation (RAR) has gained substantial attention in the literature, it still has limited use in clinical trials. Amongst other reasons, the implementation of RAR in the real world raises important practical questions, often neglected. Motivated by an innovative phase-II stratified RAR trial, this paper addresses two challenges: (1) How to ensure that RAR allocations are both desirable and faithful to target probabilities, even in small samples? and (2) What adaptations to trigger after interim analyses in the presence of missing data? We propose a Mapping strategy that discretises the randomisation probabilities into a vector of allocation ratios, resulting in improved frequentist errors. Under the implementation of Mapping, we analyse the impact of missing data on operating characteristics by examining selected scenarios. Finally, we discuss additional concerns including: pooling data across trial strata, analysing the level of blinding in the trial, and reporting safety results.

We present the thermal Stoner-Wohlfarth (tSW) model and apply it in the context of Molecular Dynamics simulations. The model is validated against an ensemble of immobilized, non-interacting, randomly oriented uniaxial particles (solid superparamagnet) and a classical dilute (non-interacting) ferrofluid for different combinations of anisotropy strength and magnetic field/moment coupling, at a fixed temperature. We compare analytical and simulation results to quantify the viability of the tSW model in reproducing the equilibrium and dynamic properties of magnetic soft matter systems. We show that if the anisotropy of a particle is more than four-five times higher than the thermal fluctuations, tSW is applicable and efficient. It provides a valuable insight into the interplay between internal magnetization degrees of freedom and Browninan rotation that is often neglected in the fixed point-dipole representation-based magnetic soft matter theoretical investigations.

Response-adaptive (RA) designs of clinical trials allow targeting a given objective by skewing the allocation of participants to treatments based on observed outcomes. RA designs face greater regulatory scrutiny due to potential type I error inflation, which limits their uptake in practice. Existing approaches to type I error control either only work for specific designs, have a risk of Monte Carlo/approximation error, are conservative, or computationally intractable. We develop a general and computationally tractable approach for exact analysis in two-arm RA designs with binary outcomes. We use the approach to construct exact tests applicable to designs that use either randomized or deterministic RA procedures, allowing for complexities such as delayed outcomes, early stopping or allocation of participants in blocks. Our efficient forward recursion implementation allows for testing of two-arm trials with 1,000 participants on a standard computer. Through an illustrative computational study of trials using randomized dynamic programming we show that, contrary to what is known for equal allocation, a conditional exact test has, almost uniformly, higher power than the unconditional test. Two real-world trials with the above-mentioned complexities are re-analyzed to demonstrate the value of our approach in controlling type I error and/or improving the statistical power., Comment: 35 pages, 6 figures, 11 tables

The idea of creating magnetically controllable colloids whose rheological properties can be finely tuned on the nano- or micro-scale has caused a lot of experimental and theoretical effort. The latter resulted in systems whose building blocks are ranging between single magnetic nanoparticles to complexes of such nanoparticles bound together by various mechanisms. The binding can be either chemical or physical, reversible or not. One way to create a system that is physically bound is to let the precrosslinked supracolloidal magnetic polymers (SMPs) to cluster due to both magnetic and Van-der-Waals-type forces. The topology of the SMPs in this case can be used to tune both magnetic and rheological properties of the resulting clusters as we show in this work. We employ Molecular Dynamics computer simulations coupled with explicit solvent modelled by Lattice-Boltzmann method in order to model the behaviour of the clusters formed by chains, rings, X- and Y-shaped SMPs in a shear flow with externally applied magnetic field. We find that the shear stabilises the shape of the clusters not letting them extend in the direction of the field and disintegrate. The clusters that show the highest response to an applied field and higher shape stability are those made of Y- and X-like SMPs., Comment: 13 pages, 10 figures

In complex colloidal systems, interparticle interactions strongly affect the dynamics of the constituting particles. A study of the dynamical response also provides invaluable information on the character of those interactions. Here we demonstrate how tuning the electrostatic interactions by an ionic surfactant in dispersions of magnetic nanoplatelets leads to developing new dynamic modes in magnetic response spectra. The collective modes can be induced or suppressed by either varying the concentration ratio of the magnetic nanoplatelets (MP) to the surfactant or increasing the MP concentration reflecting the nanoscale characteristics of this fluid magnet.

Designing experiments often requires balancing between learning about the true treatment effects and earning from allocating more samples to the superior treatment. While optimal algorithms for the Multi-Armed Bandit Problem (MABP) provide allocation policies that optimally balance learning and earning, they tend to be computationally expensive. The Gittins Index (GI) is a solution to the MABP that can simultaneously attain optimality and computationally efficiency goals, and it has been recently used in experiments with Bernoulli and Gaussian rewards. For the first time, we present a modification of the GI rule that can be used in experiments with exponentially-distributed rewards. We report its performance in simulated 2- armed and 3-armed experiments. Compared to traditional non-adaptive designs, our novel GI modified design shows operating characteristics comparable in learning (e.g. statistical power) but substantially better in earning (e.g. direct benefits). This illustrates the potential that designs using a GI approach to allocate participants have to improve participant benefits, increase efficiencies, and reduce experimental costs in adaptive multi-armed experiments with exponential rewards., Comment: Accepted by Computing Conference, London 2023

Nowadays, magnetoresponsive soft materials, based not simply on magnetic nanoparticles, but rather on multiple components with distinct sizes and magnetic properties, both in liquid and polymeric carriers, are becoming more and more wide-spread due to their unique and versatile macroscopic response to an applied magnetic field. The variability of the latter is related to a complex interplay of the magnetic interactions in a highly non-uniform internal fields caused by spatial inhomogeneity in multicomponent systems. In this work, we present a combine analytical and simulation study of binary superparamagnetic systems, containing nanoclusters and dispersed single-domain nanoparticles, both in liquid and solid carrier matrices. We investigate the equilibrium magnetic response of these systems in wide ranges of concentrations and interaction energies. It turns out that, while the magnetisation of a binary solid can be both above and below that of an ideal superparamagnetic gas, depending on the concentration of the dispersed phase and the interparticle interactions, the system in a liquid carrier is highly magnetically responsive. In liquid, a spatial redistribution of the initially homogeneously dispersed phase in the vicinity of the nanocluster is observed -- the effect that is reminiscent of the so-called ``haloing'' effect previously observed experimentally on micro- and milli-scales., Comment: 12 pages, 7 figures

We characterize the self-assembly of colloidal particles with surface mobile DNA linkers under kinetically limited valence conditions. For this, we put forward a computer simulation model that captures quantitatively the interplay between the main dynamic processes governing these systems and allows the simulation of the long time scales reached in experiments. The model is validated by direct comparison with available experimental results, showing an overall good agreement that includes measurements of the average effective valence and its probability distribution as a function of the density of DNA linkers on the particles surface. Finally, simulation results are used to evidence the opposite impact of particle density and characteristic DNA hybridization time on the effective valence., Comment: 10 pages, 6 figures

While interest in the application of machine learning to improve healthcare has grown tremendously in recent years, a number of barriers prevent deployment in medical practice. A notable concern is the potential to exacerbate entrenched biases and existing health disparities in society. The area of fairness in machine learning seeks to address these issues of equity; however, appropriate approaches are context-dependent, necessitating domain-specific consideration. We focus on clinical trials, i.e., research studies conducted on humans to evaluate medical treatments. Clinical trials are a relatively under-explored application in machine learning for healthcare, in part due to complex ethical, legal, and regulatory requirements and high costs. Our aim is to provide a multi-disciplinary assessment of how fairness for machine learning fits into the context of clinical trials research and practice. We start by reviewing the current ethical considerations and guidelines for clinical trials and examine their relationship with common definitions of fairness in machine learning. We examine potential sources of unfairness in clinical trials, providing concrete examples, and discuss the role machine learning might play in either mitigating potential biases or exacerbating them when applied without care. Particular focus is given to adaptive clinical trials, which may employ machine learning. Finally, we highlight concepts that require further investigation and development, and emphasize new approaches to fairness that may be relevant to the design of clinical trials., Comment: Appeared at ACM FAccT 2022

When comparing the performance of multi-armed bandit algorithms, the potential impact of missing data is often overlooked. In practice, it also affects their implementation where the simplest approach to overcome this is to continue to sample according to the original bandit algorithm, ignoring missing outcomes. We investigate the impact on performance of this approach to deal with missing data for several bandit algorithms through an extensive simulation study assuming the rewards are missing at random. We focus on two-armed bandit algorithms with binary outcomes in the context of patient allocation for clinical trials with relatively small sample sizes. However, our results apply to other applications of bandit algorithms where missing data is expected to occur. We assess the resulting operating characteristics, including the expected reward. Different probabilities of missingness in both arms are considered. The key finding of our work is that when using the simplest strategy of ignoring missing data, the impact on the expected performance of multi-armed bandit strategies varies according to the way these strategies balance the exploration-exploitation trade-off. Algorithms that are geared towards exploration continue to assign samples to the arm with more missing responses (which being perceived as the arm with less observed information is deemed more appealing by the algorithm than it would otherwise be). In contrast, algorithms that are geared towards exploitation would rapidly assign a high value to samples from the arms with a current high mean irrespective of the level observations per arm. Furthermore, for algorithms focusing more on exploration, we illustrate that the problem of missing responses can be alleviated using a simple mean imputation approach., Comment: 30 pages, 6 figures

We present a simple simulation model for analysing magnetic and frictional losses of magnetic nano-particles in viscous fluids subject to alternating magnetic fields. Assuming a particle size below the single-domain limit, we use a macrospin approach and solve the Landau-Lifshitz-Gilbert equation coupled to the mechanical torque equation. Despite its simplicity the presented model exhibits surprisingly rich physics and enables a detailed analysis of the different loss processes depending on field parameters and initial arrangement of the particle and the field. Depending on those parameters regions of different steady states emerge: a region with dominating N\'eel relaxation and high magnetic losses and another region region with high frictional losses at low fields or low frequencies. The energy increases continuously even across regime boundaries up to frequencies above the Brownian relaxation limit. At those higher frequencies the steady state can also depend on the initial orientation of the particle in the external field. The general behavior and special cases and their specific absorption rates are compared and discussed., Comment: 12 pages, 9 figures

Multi-armed bandit algorithms like Thompson Sampling (TS) can be used to conduct adaptive experiments, in which maximizing reward means that data is used to progressively assign participants to more effective arms. Such assignment strategies increase the risk of statistical hypothesis tests identifying a difference between arms when there is not one, and failing to conclude there is a difference in arms when there truly is one. We tackle this by introducing a novel heuristic algorithm, called TS-PostDiff (Posterior Probability of Difference). TS-PostDiff takes a Bayesian approach to mixing TS and Uniform Random (UR): the probability a participant is assigned using UR allocation is the posterior probability that the difference between two arms is 'small' (below a certain threshold), allowing for more UR exploration when there is little or no reward to be gained. We evaluate TS-PostDiff against state-of-the-art strategies. The empirical and simulation results help characterize the trade-offs of these approaches between reward, False Positive Rate (FPR), and statistical power, as well as under which circumstances each is effective. We quantify the advantage of TS-PostDiff in performing well across multiple differences in arm means (effect sizes), showing the benefits of adaptively changing randomization/exploration in TS in a "Statistically Considerate" manner: reducing FPR and increasing statistical power when differences are small or zero and there is less reward to be gained, while exploiting more when differences may be large. This highlights important considerations for future algorithm development and analysis to better balance reward and statistical analysis.

Magnetic nanogels (MNG) -- soft colloids made of polymer matrix with embedded in it magnetic nanoparticles (MNPs) -- are promising magneto-controllable drug carriers. In order to develop this potential, one needs to clearly understand the relationship between nanogel magnetic properties and its behaviour in a hydrodynamic flow. Considering the size of the MNG and typical time and velocity scales involved in their nanofluidics, experimental characterisation of the system is challenging. In this work, we perform molecular dynamics (MD) simulations combined with the Lattice-Boltzmann (LB) scheme aiming at describing the impact of the shear rate on the shape, magnetic structure and motion of an MNG. We find that in a shear flow the centre of mass of an MNG tends to be in the centre of a channel and to move preserving the distance to both walls. The MNG monomers along with translation are involved in two more types of motion, they rotate around the centre of mass and oscillate with respect to the latter. It results in synchronised tumbling and wobbling of the whole MNG accompanied by its volume oscillates. The fact the MNG is a highly compressible and permeable for the carrier liquid object makes its behaviour different from that predicted by classical Taylor-type models. We show that the frequency of volume oscillations and rotations are identical and are growing function of the shear rate. We find that the stronger magnetic interactions in the MNG are, the higher is the frequency of this complex oscillatory motion, and the lower is its amplitude. Finally, we show that the oscillations of the volume lead to the periodic changes in MNG magnetic energy., Comment: 14 pages, 14 figures

Using bandit algorithms to conduct adaptive randomised experiments can minimise regret, but it poses major challenges for statistical inference (e.g., biased estimators, inflated type-I error and reduced power). Recent attempts to address these challenges typically impose restrictions on the exploitative nature of the bandit algorithm$-$trading off regret$-$and require large sample sizes to ensure asymptotic guarantees. However, large experiments generally follow a successful pilot study, which is tightly constrained in its size or duration. Increasing power in such small pilot experiments, without limiting the adaptive nature of the algorithm, can allow promising interventions to reach a larger experimental phase. In this work we introduce a novel hypothesis test, uniquely based on the allocation probabilities of the bandit algorithm, and without constraining its exploitative nature or requiring a minimum experimental size. We characterise our $Allocation\ Probability\ Test$ when applied to $Thompson\ Sampling$, presenting its asymptotic theoretical properties, and illustrating its finite-sample performances compared to state-of-the-art approaches. We demonstrate the regret and inferential advantages of our approach, particularly in small samples, in both extensive simulations and in a real-world experiment on mental health aspects., Comment: 32 pages including supplementary material

Multi-armed bandit algorithms have been argued for decades as useful for adaptively randomized experiments. In such experiments, an algorithm varies which arms (e.g. alternative interventions to help students learn) are assigned to participants, with the goal of assigning higher-reward arms to as many participants as possible. We applied the bandit algorithm Thompson Sampling (TS) to run adaptive experiments in three university classes. Instructors saw great value in trying to rapidly use data to give their students in the experiments better arms (e.g. better explanations of a concept). Our deployment, however, illustrated a major barrier for scientists and practitioners to use such adaptive experiments: a lack of quantifiable insight into how much statistical analysis of specific real-world experiments is impacted (Pallmann et al, 2018; FDA, 2019), compared to traditional uniform random assignment. We therefore use our case study of the ubiquitous two-arm binary reward setting to empirically investigate the impact of using Thompson Sampling instead of uniform random assignment. In this setting, using common statistical hypothesis tests, we show that collecting data with TS can as much as double the False Positive Rate (FPR; incorrectly reporting differences when none exist) and the False Negative Rate (FNR; failing to report differences when they exist)...

Clinical trials of a vaccine during an epidemic face particular challenges, such as the pressure to identify an effective vaccine quickly to control the epidemic, and the effect that time-space-varying infection incidence has on the power of a trial. We illustrate how the operating characteristics of different trial design elements may be evaluated using a network epidemic and trial simulation model, based on COVID-19 and individually randomised two-arm trials with a binary outcome. We show that "ring" recruitment strategies, prioritising participants at high risk of infection, can result in substantial improvement in terms of power, if sufficiently many contacts of observed cases are at high risk. In addition, we introduce a novel method to make more efficient use of the data from the earliest cases of infection observed in the trial, whose infection may have been too early to be vaccine-preventable. Finally, we compare several methods of response-adaptive randomisation, discussing their advantages and disadvantages in this two-arm context and identifying particular adaptation strategies that preserve power and estimation properties, while slightly reducing the number of infections, given an effective vaccine.