Robust Vascular Segmentation for Raw Complex Images of Laser Speckle Contrast Based on Weakly Supervised Learning.

Laser speckle contrast imaging (LSCI) is widely used for in vivo real-time detection and analysis of local blood flow microcirculation due to its non-invasive ability and excellent spatial and temporal resolution. However, vascular segmentation of LSCI images still faces a lot of difficulties due to numerous specific noises caused by the complexity of blood microcirculation's structure and irregular vascular aberrations in diseased regions. In addition, the difficulties of LSCI image data annotation have hindered the application of deep learning methods based on supervised learning in the field of LSCI image vascular segmentation. To tackle these difficulties, we propose a robust weakly supervised learning method, which selects the threshold combinations and processing flows instead of labor-intensive annotation work to construct the ground truth of the dataset, and design a deep neural network, FURNet, based on UNet++ and ResNeXt. The model obtained from training achieves high-quality vascular segmentation and captures multi-scene vascular features on both constructed and unknown datasets with good generalization. Furthermore, we intravital verified the availability of this method on a tumor before and after embolization treatment. This work provides a new approach for realizing LSCI vascular segmentation and also makes a new application-level advance in the field of artificial intelligence-assisted disease diagnosis.

Osteoporosis prediction in lumbar spine X-ray images using the multi-scale weighted fusion contextual transformer network.

Osteoporosis is a bone-related disease characterized by decreased bone density and mass, leading to brittle fractures. Osteoporosis assessment from radiographs using a deep learning algorithm has proven a low-cost alternative to the golden standard DXA. Due to the considerable noise and low contrast, automated diagnosis of osteoporosis in X-ray images still poses a significant challenge for traditional diagnostic methods. In this paper, an end-to-end transformer-style network was proposed, termed FCoTNet, to overcome the shortcoming of insufficient fusion of texture information and local features in the traditional CoTNet. To extract complementary geometric representations at each scale of the transformer module, we integrated parallel multi-scale feature extraction architectures in each unit layer of FCoTNet to utilize convolution to aggregate features from different receptive fields. Moreover, in order to extract small-scale texture features which were more critical to the diagnosis of osteoporosis in radiographs, larger fusion weights were assigned to the feature maps with small-size receptive fields. Afterward, the multi-scale global modeling was conducted by self-attention mechanism. The proposed model was first investigated on a private lumbar spine X-ray dataset with the 5-fold cross-validation strategy, obtaining an average accuracy of 78.29 ± 0.93 %, an average sensitivity of 69.72 ± 2.35 %, and an average specificity of 88.92 ± 0.67 % for the multi-classification of normal, osteopenia, and osteoporosis categories. We then conducted a controlled trial with five orthopedic clinicians to evaluate the clinical value of the model. The average clinician's accuracy improved from 61.50 ± 10.79 % unaided to 80.00 ± 5.92 % aided (18.50 % improvement), sensitivity improved from 64.38 ± 8.07 % unaided to 83.31 ± 5.43 % aided (18.93 % improvement), and specificity improved from 80.11 ± 4.72 % unaided to 89.94 ± 3.82 % aided (9.83 % improvement). Meanwhile, the prediction consistency among clinicians significantly improved with the assistance of FCoTNet. Furthermore, the proposed model showed good robustness on an external test dataset. These investigations indicate that the proposed deep learning model achieves state-of-the-art performance for osteoporosis prediction, which substantially improves osteoporosis screening and reduced osteoporosis fractures.

Robust vessel segmentation in laser speckle contrast images based on semi-weakly supervised learning.

Objective.The goal of this study is to develop a robust semi-weakly supervised learning strategy for vessel segmentation in laser speckle contrast imaging (LSCI), addressing the challenges associated with the low signal-to-noise ratio, small vessel size, and irregular vascular aberration in diseased regions, while improving the performance and robustness of the segmentation method.Approach.For the training dataset, the healthy vascular images denoted as normal-vessel samples were manually labeled, while the diseased LSCI images involving tumor or embolism were denoted as abnormal-vessel samples and annotated as pseudo labels by the traditional semantic segmentation methods. In the training phase, the pseudo labels were constantly updated to improve the segmentation accuracy based on DeepLabv3+. Objective evaluation was conducted on the normal-vessel test set, while subjective evaluation was performed on the abnormal-vessel test set.Main results.The proposed method achieved an IOU of 0.8671, a Dice of 0.9288, and a mean relative percentage difference (mRPD) with supervised learning of 0.5% in the objective evaluation. In the subjective evaluation, our method significantly outperformed other methods in main vessel segmentation, tiny vessel segmentation, and blood vessel connection. Additionally, our method exhibited robustness when abnormal-vessel style noise was added to normal-vessel samples using a style translation network.Significance.The proposed semi-weakly supervised learning strategy demonstrates high efficiency and excellent robustness for vascular segmentation in LSCI, providing a potential tool for assessing the morphological and structural features of vessels in clinical applications.

Neurog2 directly converts astrocytes into functional neurons in midbrain and spinal cord.

Conversion of astrocytes into neurons in vivo offers an alternative therapeutic approach for neuronal loss after injury or disease. However, not only the efficiency of the conversion of astrocytes into functional neurons by single Neurog2, but also the conundrum that whether Neurog2-induced neuronal cells (Neurog2-iNs) are further functionally integrated into existing matured neural circuits remains unknown. Here, we adopted the AAV(2/8) delivery system to overexpress single factor Neurog2 into astrocytes and found that the majority of astrocytes were successfully converted into neuronal cells in multiple brain regions, including the midbrain and spinal cord. In the midbrain, Neurog2-induced neuronal cells (Neurog2-iNs) exhibit neuronal morphology, mature electrophysiological properties, glutamatergic identity (about 60%), and synapse-like configuration local circuits. In the spinal cord, astrocytes from both the intact and lesioned sources could be converted into functional neurons with ectopic expression of Neurog2 alone. Notably, further evidence from our study also proves that Neurog2-iNs in the intact spinal cord are capable of responding to diverse afferent inputs from dorsal root ganglion (DRG). Together, this study does not merely demonstrate the feasibility of Neurog2 for efficient in vivo reprogramming, it gives an indication for the Neurog2-iNs as a functional and potential factor in cell-replacement therapy.

Molecular Mechanisms Underlying Ascl1-Mediated Astrocyte-to-Neuron Conversion.

Direct neuronal reprogramming potentially provides valuable sources for cell-based therapies. Proneural gene Ascl1 converts astrocytes into induced neuronal (iN) cells efficiently both in vitro and in vivo. However, the underlying mechanisms are largely unknown. By combining RNA sequencing and chromatin immunoprecipitation followed by high-throughput sequencing, we found that the expression of 1,501 genes was markedly changed during the early stages of Ascl1-induced astrocyte-to-neuron conversion and that the regulatory regions of 107 differentially expressed genes were directly bound by ASCL1. Among Ascl1's direct targets, Klf10 regulates the neuritogenesis of iN cells at the early stage, Myt1 and Myt1l are critical for the electrophysiological maturation of iN cells, and Neurod4 and Chd7 are required for the efficient conversion of astrocytes into neurons. Together, this study provides more insights into understanding the molecular mechanisms underlying Ascl1-mediated astrocyte-to-neuron conversion and will be of value for the application of direct neuronal reprogramming.

Combined-Indications Significance Level of Multiple Related Indications Developed Simultaneously.

BACKGROUND: Typically, regulatory requirements include 2 confirmatory studies, each at a 1-sided .025 significance level, for a medicine to be approved for a specific indication. When the same medicine has been approved in related indications, 1 confirmatory study at a 1-sided .025 significance level could constitute adequate evidence of efficacy for a new indication. METHODS: This article does not contain any studies with human or animal subjects performed by any of the authors. For multiple related indications developed simultaneously to constitute sufficient evidence of clinical efficacy, the combined-studies significance level can be set at the same level as if those indications are developed sequentially. RESULTS: This article establishes possible strategies to develop a few related indications at the same time for marketing registration approval, maintaining a desired combined-studies significance level; for example, 1-sided .0000156 for 2 indications, with 1 option having each indication assessed with 1 confirmatory study at .00395 significance level. CONCLUSION: It is possible to develop a few indications at the same time for marketing registration approval, where the combinedstudies significance level is less stringent than that of the usual paradigm with 2 confirmatory studies each at 1-sided .025 significance level for every indication.

Reducing Patient Burden in Clinical Trials Through the Use of Historical Controls: Appropriate Selection of Historical Data to Minimize Risk of Bias.

Historical data have been used to augment or replace control arms in some rare disease and pediatric clinical trials. With greater availability of historical data and new methodology such as dynamic borrowing, the inclusion of historical data in clinical trials is an increasingly appealing approach for larger disease areas as well, as this can result in increased power and precision and can minimize the burden on patients in clinical trials. However, sponsors must assess whether the potential biases incurred with this approach outweigh the benefits and discuss this trade-off with the regulatory agencies. This paper discusses important points for the appropriate selection of historical controls for inclusion in the analysis of primary and/or key secondary endpoint(s) in clinical trials. The general steps are as follows: (1) Assess whether a trial is a suitable candidate for this approach. (2) If it is, then carefully identify appropriate historical trials to minimize selection bias. (3) Refine the historical control set if appropriate, for example, by selecting subsets of studies or patients. Identification of trial settings that are amenable to historical borrowing and selection of appropriate historical data using the principles discussed in this paper has the potential to lead to more efficient estimation and decision making. Ultimately, this efficiency gain results in lower patient burden and gets effective drugs to patients more quickly.

Cenicriviroc for the treatment of liver fibrosis in adults with nonalcoholic steatohepatitis: AURORA Phase 3 study design.

INTRODUCTION: Nonalcoholic steatohepatitis (NASH) is a sub-classification of nonalcoholic fatty liver disease (NAFLD) characterized by increased risk of progressive liver fibrosis. Cenicriviroc (CVC) is a novel, orally administered, potent chemokine 2 and 5 receptor antagonist currently in development for the treatment of liver fibrosis in adults with NASH. METHODS AND ANALYSIS: Efficacy and safety of CVC will be comprehensively evaluated in a global, Phase 3, multicenter, randomized, double-blind, placebo-controlled study (AURORA, NCT03028740) of subjects with NASH and Stage F2 or F3 fibrosis. Approximately 2000 adults (Part 1, 1200 subjects; Part 2, 800 additional subjects) aged 18-75 years with histological evidence of NASH with Stage F2 or F3 fibrosis (NASH Clinical Research Network classification system) will be randomized 2:1 to CVC 150 mg or placebo orally once daily. Primary efficacy endpoints will include the proportion of subjects with ≥1-stage improvement in liver fibrosis and no worsening of steatohepatitis at Month 12 relative to screening (Part 1), and time to first occurrence of any adjudicated event: death; histopathologic progression to cirrhosis; liver transplant; Model of End-Stage Liver Disease score ≥ 15; ascites; hospitalization due to liver decompensation (Part 2). Patient-reported outcomes will assess changes in health outcomes from baseline (Chronic Liver Disease Questionnaire - NAFLD; Work Productivity and Activity Impairment in NASH; 36-Item Short Form Health Survey version 2). Adverse events will be assessed throughout the study. As there are currently no approved treatments indicated for NASH, the AURORA CVC Phase 3 study addresses an unmet medical need.

Conversion of Astrocytes and Fibroblasts into Functional Noradrenergic Neurons.

Dysfunction of noradrenergic (NA) neurons is associated with a number of neuronal disorders. Diverse neuronal subtypes can be generated by direct reprogramming. However, it is still unknown how to convert non-neuronal cells into NA neurons. Here, we show that seven transcription factors (TFs) (Ascl1, Phox2b, AP-2α, Gata3, Hand2, Nurr1, and Phox2a) are able to convert astrocytes and fibroblasts into induced NA (iNA) neurons. These iNA neurons express the genes required for the biosynthesis, release, and re-uptake of noradrenaline. Moreover, iNA neurons fire action potentials, receive synaptic inputs, and control the beating rate of co-cultured ventricular myocytes. Furthermore, iNA neurons survive and integrate into neural circuits after transplantation. Last, human fibroblasts can be converted into functional iNA neurons as well. Together, iNA neurons are generated by direct reprogramming, and they could be potentially useful for disease modeling and cell-based therapies.