In situbioprinting has been identified as a promising tissue engineering technique for treating volumetric muscle loss (VML) injuries. However, the success of this procedure significantly depends on the uniform and precise deposition of cells contributing to the regeneration of muscles. To address this critical need, in this work, we present design and quantitative evaluation of a novel autonomousin situbioprinting surgical robotic framework that can be used with a generic bioprinting material. The proposed framework consists of three main components: (i) a bioprinting tool integrated with a seven-degree-of-freedom robotic manipulator to perform a precise autonomous bioprinting procedure; (ii) a unique 3D visual measurement framework comprised of a high-accuracy structured light camera with complementary 2D/3D computer vision algorithms-to enable online and accurate measurement and reconstruction of the bioprinted constructs; and (iii) a quantitative evaluation module with novel assessment metrics-to characterize and evaluate the performance of the bioprinting process toward finding optimal bioprinting parameters. To ensure the biological functionality of a printed construct using our robotic system, we performed 90 experiments and identified optimal bioprinting parameters using the proposed novel assessment metrics.Note to Practitioners—This paper was motivated by the problem of volumetric muscle loss treatment using anin situbioprinting procedure but it also can be applied for treatment of skin and cartilage injuries. Existing approaches to performin situbioprinting is limited to either manual handheld bioprinting devices– that suffer from poor manual control and inaccurate printing constructs– or robotic systems– that have been developed without (i) considering a realistic surgical workflow and (ii) quantitatively evaluating the quality of printed constructs. To collectively address these issues, in this paper, we propose a novel autonomousin siturobotic bioprinting framework. We also introduce unique and complementary quantitative assessment metrics to characterize and evaluate the performance of the bioprinting process. Experiments suggest that the proposed framework can robustly identify optimal bioprinting parameters to ensure the biological functionality of a printed construct using our robotic system.

Accurate segmentation of brain tumors from medical images is important for diagnosis and treatment planning, and it often requires multi-modal or contrast-enhanced images. However, in practice some modalities of a patient may be absent. Synthesizing the missing modality has a potential for filling this gap and achieving high segmentation performance. Existing methods often treat the synthesis and segmentation tasks separately or consider them jointly but without effective regularization of the complex joint model, leading to limited performance. We propose a novel brain Tumor Image Synthesis and Segmentation network (TISS-Net) that obtains the synthesized target modality and segmentation of brain tumors end-to-end with high performance. First, we propose a dual-task-regularized generator that simultaneously obtains a synthesized target modality and a coarse segmentation, which leverages a tumor-aware synthesis loss with perceptibility regularization to minimize the high-level semantic domain gap between synthesized and real target modalities. Based on the synthesized image and the coarse segmentation, we further propose a dual-task segmentor that predicts a refined segmentation and error in the coarse segmentation simultaneously, where a consistency between these two predictions is introduced for regularization. Our TISS-Net was validated with two applications: synthesizing FLAIR images for whole glioma segmentation, and synthesizing contrast-enhanced T1 images for Vestibular Schwannoma segmentation. Experimental results showed that our TISS-Net largely improved the segmentation accuracy compared with direct segmentation from the available modalities, and it outperformed state-of-the-art image synthesis-based segmentation methods.

Automatic segmentation of COVID-19 pneumonia lesions is critical for quantitative measurement for diagnosis and treatment management. For this task, deep learning is the state-of-the-art method while requires a large set of accurately annotated images for training, which is difficult to obtain due to limited access to experts and the time-consuming annotation process. To address this problem, we aim to train the segmentation network from imperfect annotations, where the training set consists of a small clean set of accurately annotated images by experts and a large noisy set of inaccurate annotations by non-experts. To avoid the labels with different qualities corrupting the segmentation model, we propose a new approach to train segmentation networks to deal with noisy labels. We introduce a dual-branch network to separately learn from the accurate and noisy annotations. To fully exploit the imperfect annotations as well as suppressing the noise, we design a Divergence-Aware Selective Training (DAST) strategy, where a divergence-aware noisiness score is used to identify severely noisy annotations and slightly noisy annotations. For severely noisy samples we use an regularization through dual-branch consistency between predictions from the two branches. We also refine slightly noisy samples and use them as supplementary data for the clean branch to avoid overfitting. Experimental results show that our method achieves a higher performance than standard training process for COVID-19 pneumonia lesion segmentation when learning from imperfect labels, and our framework outperforms the state-of-the-art noise-tolerate methods significantly with various clean label percentages.