Deep-silicon photon-counting x-ray projection denoising through reinforcement learning.

BACKGROUND: In recent years, deep reinforcement learning (RL) has been applied to various medical tasks and produced encouraging results. OBJECTIVE: In this paper, we demonstrate the feasibility of deep RL for denoising simulated deep-silicon photon-counting CT (PCCT) data in both full and interior scan modes. PCCT offers higher spatial and spectral resolution than conventional CT, requiring advanced denoising methods to suppress noise increase. METHODS: In this work, we apply a dueling double deep Q network (DDDQN) to denoise PCCT data for maximum contrast-to-noise ratio (CNR) and a multi-agent approach to handle data non-stationarity. RESULTS: Using our method, we obtained significant image quality improvement for single-channel scans and consistent improvement for all three channels of multichannel scans. For the single-channel interior scans, the PSNR (dB) and SSIM increased from 33.4078 and 0.9165 to 37.4167 and 0.9790 respectively. For the multichannel interior scans, the channel-wise PSNR (dB) increased from 31.2348, 30.7114, and 30.4667 to 31.6182, 30.9783, and 30.8427 respectively. Similarly, the SSIM improved from 0.9415, 0.9445, and 0.9336 to 0.9504, 0.9493, and 0.0326 respectively. CONCLUSIONS: Our results show that the RL approach improves image quality effectively, efficiently, and consistently across multiple spectral channels and has great potential in clinical applications.

Simulated deep CT characterization of liver metastases with high-resolution filtered back projection reconstruction.

Early diagnosis and accurate prognosis of colorectal cancer is critical for determining optimal treatment plans and maximizing patient outcomes, especially as the disease progresses into liver metastases. Computed tomography (CT) is a frontline tool for this task; however, the preservation of predictive radiomic features is highly dependent on the scanning protocol and reconstruction algorithm. We hypothesized that image reconstruction with a high-frequency kernel could result in a better characterization of liver metastases features via deep neural networks. This kernel produces images that appear noisier but preserve more sinogram information. A simulation pipeline was developed to study the effects of imaging parameters on the ability to characterize the features of liver metastases. This pipeline utilizes a fractal approach to generate a diverse population of shapes representing virtual metastases, and then it superimposes them on a realistic CT liver region to perform a virtual CT scan using CatSim. Datasets of 10,000 liver metastases were generated, scanned, and reconstructed using either standard or high-frequency kernels. These data were used to train and validate deep neural networks to recover crafted metastases characteristics, such as internal heterogeneity, edge sharpness, and edge fractal dimension. In the absence of noise, models scored, on average, 12.2% ( α = 0.012 ) and 7.5% ( α = 0.049 ) lower squared error for characterizing edge sharpness and fractal dimension, respectively, when using high-frequency reconstructions compared to standard. However, the differences in performance were statistically insignificant when a typical level of CT noise was simulated in the clinical scan. Our results suggest that high-frequency reconstruction kernels can better preserve information for downstream artificial intelligence-based radiomic characterization, provided that noise is limited. Future work should investigate the information-preserving kernels in datasets with clinical labels.

Disrupting adversarial transferability in deep neural networks.

Adversarial attack transferability is well recognized in deep learning. Previous work has partially explained transferability by recognizing common adversarial subspaces and correlations between decision boundaries, but little is known beyond that. We propose that transferability between seemingly different models is due to a high linear correlation between the feature sets that different networks extract. In other words, two models trained on the same task that are distant in the parameter space likely extract features in the same fashion, linked by trivial affine transformations between the latent spaces. Furthermore, we show how applying a feature correlation loss, which decorrelates the extracted features in corresponding latent spaces, can reduce the transferability of adversarial attacks between models, suggesting that the models complete tasks in semantically different ways. Finally, we propose a dual-neck autoencoder (DNA), which leverages this feature correlation loss to create two meaningfully different encodings of input information with reduced transferability.

Feasibility analysis on simultaneous electron density and attenuation coefficient reconstruction.

PURPOSE: Creating a viable reconstruction method for Compton scatter tomography remains challenging. Accounting for scatter attenuation when the underlying attenuation map is not known is particularly challenging, and current mathematical approaches to this vary widely. This work explores a novel approach to joint scatter and attenuation image reconstruction, which leverages the underlying structural similarity between the two images and incorporates a deep learning model in an alternating iterative reconstruction scheme. METHODS: A single-view computed tomography (CT) imaging procedure for recording Compton scatter is first described. A joint reconstruction model, which iterates between algebraically reconstructing scatter images and estimating the attenuation via deep learning, is then proposed. This model is tested on both a generated dataset of 2D phantom images designed to mimic human tissues as well as a realistically simulated dataset based on real CT images. RESULTS: Testing results yield convergence of the model and decent reconstruction quality to distinguish crucial features such as tumors and lesions, demonstrating the potential principled utilities of this configuration and deep learning approach. The model achieved a structural similarity index measure of at least 0.82 for scatter and 0.88 for attenuation reconstructions with the realistically simulated dataset. CONCLUSION: The iterative, deep learning approach outlined in this work shows potential for future efficient medical imaging procedures, reconstructing images with limited scatter information.

Modeling of moral decisions with deep learning.

One example of an artificial intelligence ethical dilemma is the autonomous vehicle situation presented by Massachusetts Institute of Technology researchers in the Moral Machine Experiment. To solve such dilemmas, the MIT researchers used a classic statistical method known as the hierarchical Bayesian (HB) model. This paper builds upon previous work for modeling moral decision making, applies a deep learning method to learn human ethics in this context, and compares it to the HB approach. These methods were tested to predict moral decisions of simulated populations of Moral Machine participants. Overall, test results indicate that deep neural networks can be effective in learning the group morality of a population through observation, and outperform the Bayesian model in the cases of model mismatches.

Design and characterization of low-cost fabric-based flat pneumatic actuators for soft assistive glove application.

We present the design of low-cost fabric-based Hat pneumatic actuators for soft assistive glove application. The soft assistive glove is designed to assist hand impaired patients in performing activities of daily living and rehabilitation. The actuators consist of flexible materials such as fabric and latex bladder. Using zero volume actuation concept, the 2D configuration of the actuators simplifies the manufacturing process and allows the actuators to be more compact. The actuators achieve bi-directional flexion and extension motions. Compared to previously developed inflatable soft actuators, the actuators generate sufficient force and torque to assist in both finger flexion and extension at lower air pressure. Preliminary evaluation results show that the glove is able to provide both active finger flexion and extension assistance for activities of daily living and rehabilitative training.