Search

This study explores the utilization of Dermatoscopic synthetic data generated through stable diffusion models as a strategy for enhancing the robustness of machine learning model training. Synthetic data generation plays a pivotal role in mitigating challenges associated with limited labeled datasets, thereby facilitating more effective model training. In this context, we aim to incorporate enhanced data transformation techniques by extending the recent success of few-shot learning and a small amount of data representation in text-to-image latent diffusion models. The optimally tuned model is further used for rendering high-quality skin lesion synthetic data with diverse and realistic characteristics, providing a valuable supplement and diversity to the existing training data. We investigate the impact of incorporating newly generated synthetic data into the training pipeline of state-of-art machine learning models, assessing its effectiveness in enhancing model performance and generalization to unseen real-world data. Our experimental results demonstrate the efficacy of the synthetic data generated through stable diffusion models helps in improving the robustness and adaptability of end-to-end CNN and vision transformer models on two different real-world skin lesion datasets., Comment: Paper is submitted in EMBC 2024 Conference

For many cancer sites low-dose risks are not known and must be extrapolated from those observed in groups exposed at much higher levels of dose. Measurement error can substantially alter the dose-response shape and hence the extrapolated risk. Recently, there has been considerable attention paid to methods of dealing with shared errors, which are particularly important in occupational and environmental settings. In this paper we test Bayesian model averaging (BMA) and frequentist model averaging (FMA) methods, the first of these similar to the so-called Bayesian two-dimensional Monte Carlo (2DMC) method, and both fairly recently proposed, against a very newly proposed modification of the regression calibration method, the extended regression calibration (ERC) method. The quasi-2DMC+BMA method performs well when a linear model is assumed, but poorly when a linear-quadratic model is assumed. FMA performs as well as quasi-2DMC+BMA when a linear model is assumed, and generally much better with a linear-quadratic model, although the coverage probability for the quadratic coefficient is uniformly too high. ERC yields coverage probabilities that are too low when shared and unshared Berkson errors are both large (50%), although otherwise it performs well, and coverage is generally better than the quasi-2DMC+BMA or FMA methods, particularly for the linear-quadratic model. The bias of predicted relative risk at a variety of doses is generally smallest for ERC, and largest for quasi-2DMC+BMA and FMA, with standard regression calibration and Monte Carlo maximum likelihood exhibiting bias in predicted relative risk generally somewhat intermediate between ERC and the other two methods. In general ERC performs best in the scenarios presented, and should be the method of choice in situations where there may be substantial shared error, or suspected curvature in the dose response., Comment: 5 Tables. Fortran code supplied externally

Currently, there are 14 ongoing clinical studies on low dose radiotherapy (LDRT) for COVID-19 pneumonia. An underlying assumption is that irradiation of about 1 Gy is effective at ameliorating viral pneumonia. Its rationale, however, relies on early human case series or animal studies mostly obtained in the pre-antibiotic era, where rigorous statistical analyses were not performed. It therefore remains unclear whether those early data support such assumptions. With standard statistical survival models, and based on a systematic literature review, we re-analyzed 14 radiobiological animal datasets in which animals received mostly fractionated doses of radiation before or after bacterial/viral inoculation, and assessing various health endpoints (mortality, pneumonia morbidity). In most datasets absorbed doses did not exceed 7 Gy. Various different model systems and types of challenging infection are considered. For 7 studies that evaluated post-inoculation radiation exposure (more relevant to LDRT for COVID-19 pneumonia) the results are heterogeneous, with 2 studies showing a significant increase (p<0.001) and another showing a significant decrease (p<0.001) in mortality associated with radiation exposure. For pre-inoculation exposure the results are also heterogeneous, with 6 datasets showing a significant increase (p<0.01) in mortality risk associated with radiation exposure and the other 2 showing a significant decrease (p<0.05) in mortality risk. Collectively, these data do not provide clear support for reductions in morbidity or mortality associated with post-infection radiation exposure. For pre-infection radiation exposure the inconsistency of direction of effect makes this body of data difficult to interpret. Nevertheless, one must be cautious about adducing evidence from the published reports of these old animal datasets., Comment: 2 tables, 7 figures, 15 Appendix Tables, 36 references, 49 double spaced pages

The paper of Little et al. (PloS Comput Biol 2009 5(10) e1000539) outlined a system of reaction-diffusion equations that were used to describe induction of atherosclerotic disease. These were solved by considering an equilibrium solution and small perturbations around this equilibrium. Here we consider slight variant sets of assumptions that could be used to derive equilibrium solutions. In general they do not imply any change in the numerical results relating to monocyte chemo-attractant protein-1 (MCP-1) presented in that paper., Comment: 4 pages, 0 figures. This expands on the comments made on the PLoS Comput Biol webpage for the article of Little et al (PLoS Comput Biol 2009 5(10) e1000539). Acknowledgements have been added

In this paper we outline general considerations on parameter identifiability, and introduce the notion of weak local identifiability and gradient weak local identifiability. These are based on local properties of the likelihood, in particular the rank of the Hessian matrix. We relate these to the notions of parameter identifiability and redundancy previously introduced by Rothenberg (Econometrica 39 (1971) 577-591) and Catchpole and Morgan (Biometrika 84 (1997) 187-196). Within the exponential family parameter irredundancy, local identifiability, gradient weak local identifiability and weak local identifiability are shown to be equivalent. We consider applications to a recently developed class of cancer models of Little and Wright (Math Biosciences 183 (2003) 111-134) and Little et al. (J Theoret Biol 254 (2008) 229-238) that generalize a large number of other recently used quasi-biological cancer models, in particular those of Armitage and Doll (Br J Cancer 8 (1954) 1-12) and the two-mutation model (Moolgavkar and Venzon Math Biosciences 47 (1979) 55-77)., Comment: This is now exactly as per the version of the article published in Little et al. (PLoS ONE 2010 5 (1)e8915)

One of the biggest problems in current distributed systems is that presented by one machine attempting to determine the liveness of another in a timely manner. Unfortunately, the symptoms exhibited by a failed machine can also be the result of other causes, e.g., an overloaded machine or network which drops messages, making it impossible to detect a machine failure with cetainty until that machine recovers. This is a well understood problem and one which has led to a large body of research into failure suspectors: since it is not possible to detect a failure, the best one can do is suspect a failure and program accordingly. However, one machine's suspicions may not be the same as another's; therefore, these algorithms spend a considerable effort in ensuring a consistent view among all available machines of who is suspects of being failed. This paper describes a thought experiment on how quantum mechanics may be used to provide a failure detector that is guaranteed to give both accurate and instantaneous information about the liveness of machines, no matter the distances involved.