No abstract

No abstract

Medical image diagnosis can be achieved by deep neural networks, provided there is enough varied training data for each disease class. However, a hitherto unknown disease class not encountered during training will inevitably be misclassified, even if predicted with low probability. This problem is especially important for medical image diagnosis, when an image of a hitherto unknown disease is presented for diagnosis, especially when the images come from the same image domain, such as dermoscopic skin images.Current out-of-distribution detection algorithms act unfairly when the in-distribution classes are imbalanced, by favouring the most numerous disease in the training sets. This could lead to false diagnoses for rare cases which are often medically important.We developed a novel yet simple method to train neural networks, which enables them to classify in-distribution dermoscopic skin disease images and also detect novel diseases from dermoscopic images at test time. We show that our BinaryHeads model not only does not hurt classification balanced accuracy when the data is imbalanced, but also consistently improves the balanced accuracy. We also introduce an important method to investigate the effectiveness of out-of-distribution detection methods based on presence of varying amounts of out-of-distribution data, which may arise in real-world settings.

This paper presents the development of dynamic models for proton exchange membrane fuel cells (PEMFC). The PEMFC control system has an important effect on operation of cell. Traditional controllers could not lead to acceptable responses because of time-change, long-hysteresis, uncertainty, strong-coupling and nonlinear characteristics of PEMFCs, This paper presents a dynamic model for PEMFC system, so an intelligent or adaptive controller is needed. In this paper, a neural network predictive controller have been designed to control the voltage of at the presence of fluctuations of temperature. The results of implementation of this designed NN Predictive controller on a dynamic electrochemical model of a small size 5 KW, PEM fuel cell have been simulated by matlab/SIMULINK.

Nuclear cardiology has not witnessed development of new tracers or hardware for many years. Hence there is a need for the development of improvised techniques. Dynamic cardiac single photon emission computed tomography (SPECT) is one such technique that has a potential to overcome the limitations of conventional myocardial SPECT including the absolute quantification of blood flow. The main goal of this study is to evaluate the effect of attenuation correction (AC) on estimation of the washout parameters extracted from dynamic SPECT using a conventional protocol. The effect of the postprocessing on quantitative evaluation of dynamic SPECT is also assessed.A physical phantom was employed to physically simulate the dynamic behavior of a heart in the thorax. Using a dual detector SPECT system, 180° tomographic data in every 90 seconds were acquired. The SPECT data were reconstructed using ordered subset expectation maximization (OSEM) method while different iterations and a Butterworth filter with different cut-off frequencies were applied. Estimated washout parameter of the time activity curves (TACs) was compared with applying AC or without it.Results show that AC can improve the bias of computed washout parameter in normal regions (average bias reduction in normal ROI: 7%). Moreover, the postreconstruction filtering and reducing the number of iterations in reconstructing phase can reduce the variance of the computed washout values in normal regions (from 3.99% for cut-off frequency 0.5 cycle/cm and 32 times update in OSEM to 2.05% for cut-off frequency 0.35 cycle/cm and 16 times update in OSEM). They also reduce the actual size of the defect region (13% reduction in defect extent for above change in reconstruction parameters).According to the results, the AC and postprocessing filtration can directly affect the standard deviation of washout value acquired by cardiac dynamic SPECT. These parameters also showed a direct effect on the defect extent in final results. The study showed that the AC may partly improve the bias of calculated normal washout value. The effect of attenuation correction on the defective washout value could not be answered comprehensively in this paper.