Simulation and synthesis in medical imaging : ... International Workshop, SASHIMI ..., held in conjunction with MICCAI ..., proceedings. SASHIMI (Workshop)最新文献

Cochlear implants (CIs) are considered the standard-of-care treatment for profound sensory-based hearing loss. Several groups have proposed computational models of the cochlea in order to study the neural activation patterns in response to CI stimulation. However, most of the current implementations either rely on high-resolution histological images that cannot be customized for CI users or CT images that lack the spatial resolution to show cochlear structures. In this work, we propose to use a deep learning-based method to obtain μCT level tissue labels using patient CT images. Experiments showed that the proposed super-resolution segmentation architecture achieved very good performance on the inner-ear tissue segmentation. Our best-performing model (0.871) outperformed the UNet (0.746), VNet (0.853), nnUNet (0.861), TransUNet (0.848), and SRGAN (0.780) in terms of mean dice score.

The rapid tracer kinetics of rubidium-82 (⁸²Rb) and high variation of cross-frame distribution in dynamic cardiac positron emission tomography (PET) raise significant challenges for inter-frame motion correction, particularly for the early frames where conventional intensity-based image registration techniques are not applicable. Alternatively, a promising approach utilizes generative methods to handle the tracer distribution changes to assist existing registration methods. To improve frame-wise registration and parametric quantification, we propose a Temporally and Anatomically Informed Generative Adversarial Network (TAI-GAN) to transform the early frames into the late reference frame using an all-to-one mapping. Specifically, a feature-wise linear modulation layer encodes channel-wise parameters generated from temporal tracer kinetics information, and rough cardiac segmentations with local shifts serve as the anatomical information. We validated our proposed method on a clinical ⁸²Rb PET dataset and found that our TAI-GAN can produce converted early frames with high image quality, comparable to the real reference frames. After TAI-GAN conversion, motion estimation accuracy and clinical myocardial blood flow (MBF) quantification were improved compared to using the original frames. Our code is published at https://github.com/gxq1998/TAI-GAN.

Laser interstitial thermal therapy (LITT) is a novel minimally invasive treatment that is used to ablate intracranial structures to treat mesial temporal lobe epilepsy (MTLE). Region of interest (ROI) segmentation before and after LITT would enable automated lesion quantification to objectively assess treatment efficacy. Deep learning techniques, such as convolutional neural networks (CNNs) are state-of-the-art solutions for ROI segmentation, but require large amounts of annotated data during the training. However, collecting large datasets from emerging treatments such as LITT is impractical. In this paper, we propose a progressive brain lesion synthesis framework (PAVAE) to expand both the quantity and diversity of the training dataset. Concretely, our framework consists of two sequential networks: a mask synthesis network and a mask-guided lesion synthesis network. To better employ extrinsic information to provide additional supervision during network training, we design a condition embedding block (CEB) and a mask embedding block (MEB) to encode inherent conditions of masks to the feature space. Finally, a segmentation network is trained using raw and synthetic lesion images to evaluate the effectiveness of the proposed framework. Experimental results show that our method can achieve realistic synthetic results and boost the performance of down-stream segmentation tasks above traditional data augmentation techniques.

Magnetic resonance imaging (MRI) with gadolinium contrast is widely used for tissue enhancement and better identification of active lesions and tumors. Recent studies have shown that gadolinium deposition can accumulate in tissues including the brain, which raises safety concerns. Prior works have tried to synthesize post-contrast T1-weighted MRIs from pre-contrast MRIs to avoid the use of gadolinium. However, contrast and image representations are often entangled during the synthesis process, resulting in synthetic post-contrast MRIs with undesirable contrast enhancements. Moreover, the synthesis of pre-contrast MRIs from post-contrast MRIs which can be useful for volumetric analysis is rarely investigated in the literature. To tackle pre- and post- contrast MRI synthesis, we propose a BI-directional Contrast Enhancement Prediction and Synthesis (BICEPS) network that enables disentanglement of contrast and image representations via a bi-directional image-to-image translation(I2I)model. Our proposed model can perform both pre-to-post and post-to-pre contrast synthesis, and provides an interpretable synthesis process by predicting contrast enhancement maps from the learned contrast embedding. Extensive experiments on a multiple sclerosis dataset demonstrate the feasibility of applying our bidirectional synthesis and show that BICEPS outperforms current methods.

Nonlinear inter-modality registration is often challenging due to the lack of objective functions that are good proxies for alignment. Here we propose a synthesis-by-registration method to convert this problem into an easier intra-modality task. We introduce a registration loss for weakly supervised image translation between domains that does not require perfectly aligned training data. This loss capitalises on a registration U-Net with frozen weights, to drive a synthesis CNN towards the desired translation. We complement this loss with a structure preserving constraint based on contrastive learning, which prevents blurring and content shifts due to overfitting. We apply this method to the registration of histological sections to MRI slices, a key step in 3D histology reconstruction. Results on two public datasets show improvements over registration based on mutual information (13% reduction in landmark error) and synthesis-based algorithms such as CycleGAN (11% reduction), and are comparable to registration with label supervision. Code and data are publicly available at https://github.com/acasamitjana/SynthByReg.

Cochlear implants (CIs) restore hearing using an array of electrodes implanted in the cochlea to directly stimulate auditory nerve fibers (ANFs). Hearing outcomes with CIs are dependent on the health of the ANFs. In this research, we developed an approach to estimate the health of ANFs using patient-customized, image-based computational models of CI stimulation. Our stimulation models build on a previous model-based solution to estimate the intra-cochlear electric field (EF) created by the CI. Herein, we propose to use the estimated EF to drive ANF models representing 75 nerve bundles along the length of the cochlea. We propose a method to detect the neural health of the ANF models by optimizing neural health parameters to minimize the sum of squared differences between simulated and the physiological measurements available via patients' CIs. The resulting health parameters provide an estimate of the health of ANF bundles. Experiments with 8 subjects show promising model prediction accuracy, with excellent agreement between neural stimulation responses that are clinically measured and those that are predicted by our parameter optimized models. These results suggest our modeling approach may provide an accurate estimation of ANF health for CI users.