RESUMO
The ethanol/water separation challenge highlights the adsorption capacity/selectivity trade-off problem. We show that the target guest can serve as a gating component of the host to block the undesired guest, giving molecular sieving effect for the adsorbent possessing large pores. Two hydrophilic/water-stable metal azolate frameworks were designed to compare the effects of gating and pore-opening flexibility. Large amounts (up to 28.7â mmol g-1 ) of ethanol with fuel-grade (99.5 %+) and even higher purities (99.9999 %+) can be produced in a single adsorption process from not only 95 : 5 but also 10 : 90 ethanol/water mixtures. More interestingly, the pore-opening adsorbent possessing large pore apertures showed not only high water adsorption capacity but also exceptionally high water/ethanol selectivity characteristic of molecular sieving. Computational simulations demonstrated the critical role of guest-anchoring aperture for the guest-dominated gating process.
RESUMO
Wavelength-selective thermal emitters (WS-EMs) are of interest due to the lack of cost-effective, narrow-band sources in the mid- to long-wave infrared. WS-EMs can be realized via Tamm plasmon polaritons (TPPs) supported by distributed Bragg reflectors on metals. However, the design of multiple resonances is challenging as numerous structural parameters must be optimized simultaneously. Here we use stochastic gradient descent to optimize TPP emitters (TPP-EMs) composed of an aperiodic distributed Bragg reflector deposited on doped cadmium oxide (CdO) film, where layer thicknesses and carrier density are inversely designed. The combination of the aperiodic distributed Bragg reflector with the designable plasma frequency of CdO enables multiple TPP-EM modes to be simultaneously designed with arbitrary spectral control not accessible with metal-based TPPs. Using this approach, we experimentally demonstrated and numerically proposed TPP-EMs exhibiting single or multiple emission bands with designable frequencies, line-widths and amplitudes. This thereby enables lithography-free, wafer-scale WS-EMs that are complementary metal-oxide-semiconductor compatible for applications such as free-space communications and gas sensing.
RESUMO
With the rapid development of image acquisition and storage, multiple images per class are commonly available for computer vision tasks (e.g., face recognition, object detection, medical imaging, etc.). Recently, the recurrent neural network (RNN) has been widely integrated with convolutional neural networks (CNN) to perform image classification on ordered (sequential) data. In this paper, by permutating multiple images as multiple dummy orders, we generalize the ordered "RNN+CNN" design (longitudinal) to a novel unordered fashion, called Multi-path x-D Recurrent Neural Network (MxDRNN) for image classification. To the best of our knowledge, few (if any) existing studies have deployed the RNN framework to unordered intra-class images to leverage classification performance. Specifically, multiple learning paths are introduced in the MxDRNN to extract discriminative features by permutating input dummy orders. Eight datasets from five different fields (MNIST, 3D-MNIST, CIFAR, VGGFace2, and lung screening computed tomography) are included to evaluate the performance of our method. The proposed MxDRNN improves the baseline performance by a large margin across the different application fields (e.g., accuracy from 46.40% to 76.54% in VGGFace2 test pose set, AUC from 0.7418 to 0.8162 in NLST lung dataset). Additionally, empirical experiments show the MxDRNN is more robust to category-irrelevant attributes (e.g., expression, pose in face images), which may introduce difficulties for image classification and algorithm generalizability. The code is publicly available.
RESUMO
Objective: Anodal transcranial direct current stimulation (a-tDCS) has been used to improve athletic performance in various populations; however, its role in improving performance in elite athletes is unclear. This study aimed to investigate the effects of a-tDCS on athletic performance in elite athletes. Methods: We used a single-blind, randomized controlled experimental design and recruited 24 national-level freestyle swimmers from China. All athletes were randomly divided into two groups; the experimental group underwent a-tDCS (current 2 mA for 20 min) combined with physical training, and the control group underwent a-tDCS sham stimulation combined with physical training. The physical training program was identical in the experimental and control groups. The intervention period was 6 weeks, with five weekly sessions of 110 min each, consisting of 20 min of a-tDCS and 90 min of physical training. Base strength, explosive strength, aerobic endurance, and anaerobic endurance were measured in the athletes before and after the intervention. Results: The results were as follows. 1) Basic strength: There was a significant improvement in 5RM pull-ups in the experimental and control groups before and after the intervention (p < 0.05). 2) Explosive strength: There was a significant improvement in vertical jump and swimming start distance into the water in the experimental and control groups before and after the intervention (p < 0.05). 3) Aerobic endurance: There was no significant improvement in the experimental and control groups before and after the intervention. 4) Anaerobic endurance: There was a significant improvement in 400 m running performance in the experimental and control groups before and after the intervention (p < 0.05). Conclusion: Compared to physical training alone, a-tDCS combined with physical training can better improve the athletic performance of high-level swimmers, especially in basic strength, explosive strength, and aerobic endurance.
RESUMO
Objectives: This study examined and compared the effects of functional and running high-intensity interval training (HIIT) on body composition, cardiorespiratory fitness, and muscular fitness of young adults with overweight or obesity. Methods: Forty-five participants (22.1 ± 2.1 years, BMI = 25.2 ± 1.0 kg/m2) were assigned to functional HIIT (HIIT-F; n = 15), running HIIT (HIIT-R; n = 15), or non-training control group (CON; n = 15). Participants in HIIT-F and HIIT-R performed functional exercise based-HIIT (four sets of all-out whole-body exercises including jumping jacks, squats, twist jumps and mountain climbers, et al.) and running HIIT (four sets of running on a treadmill) for 12 weeks, respectively. Body composition, muscular fitness, and cardiorespiratory fitness were assessed pre and post intervention. Results: Both HIIT-F and HIIT-R significantly improved the body composition and cardiorespiratory fitness, with HIIT-F induced greater improvements in lean mass (+1.623 vs. -1.034 kg, p < 0.001), back strength (+6.007 vs. +3.333 kg, p < 0.01), and push-ups (+5.692 vs. 1.923 reps, p < 0.001) than that in HIIT-R. HIIT-R reduced more visceral fat area (VFA) (-11.416 vs. -4.338 cm2, p = 0.052) and induced similar improvement in cardiorespiratory fitness (VO2max, +2.192 vs. +2.885 mL/kg/min, p = 0.792) with HIIT-F. Conclusion: Twelve weeks of HIIT-R or HIIT-F improved physical fitness among young adults with overweight or obesity. Despite the similar impact on cardiorespiratory fitness, HIIT-F generates a better positive effect on muscular fitness relative to HIIT-R, which could be partly explained by the greater increase in lean mass after HIIT-F intervention.
RESUMO
The advancement of artificial intelligence (AI) for organ segmentation and tumor detection is propelled by the growing availability of computed tomography (CT) datasets with detailed, per-voxel annotations. However, these AI models often struggle with flexibility for partially annotated datasets and extensibility for new classes due to limitations in the one-hot encoding, architectural design, and learning scheme. To overcome these limitations, we propose a universal, extensible framework enabling a single model, termed Universal Model, to deal with multiple public datasets and adapt to new classes (e.g., organs/tumors). Firstly, we introduce a novel language-driven parameter generator that leverages language embeddings from large language models, enriching semantic encoding compared with one-hot encoding. Secondly, the conventional output layers are replaced with lightweight, class-specific heads, allowing Universal Model to simultaneously segment 25 organs and six types of tumors and ease the addition of new classes. We train our Universal Model on 3410 CT volumes assembled from 14 publicly available datasets and then test it on 6173 CT volumes from four external datasets. Universal Model achieves first place on six CT tasks in the Medical Segmentation Decathlon (MSD) public leaderboard and leading performance on the Beyond The Cranial Vault (BTCV) dataset. In summary, Universal Model exhibits remarkable computational efficiency (6× faster than other dataset-specific models), demonstrates strong generalization across different hospitals, transfers well to numerous downstream tasks, and more importantly, facilitates the extensibility to new classes while alleviating the catastrophic forgetting of previously learned classes. Codes, models, and datasets are available at https://github.com/ljwztc/CLIP-Driven-Universal-Model.
RESUMO
Parkinson's disease (PD) is the second most common neurodegenerative disorder characterized by the accumulation of α-synuclein (α-syn) aggregates called Lewy bodies leading to the gradual loss of dopaminergic (DA) neurons in the substantia nigra. Although α-syn expression can be attenuated by antisense oligonucleotides (ASOs) and heteroduplex oligonucleotide (HDO) by intracerebroventricular (ICV) injection, the challenge to peripheral targeted delivery of oligonucleotide safely and effectively into DA neurons remains unresolved. Here, we designed a new DNA/DNA double-stranded (complementary DNA, coDNA) molecule with cholesterol conjugation (Chol-HDO (coDNA)) based on an α-syn-ASO sequence and evaluated its silence efficiency. Further, Chol-HDO@LMNPs, Chol-HDO-loaded, cerebrovascular endothelial cell membrane with DSPE-PEG2000-levodopa modification (L-DOPA-CECm)-coated nanoparticles (NPs), were developed for the targeted treatment of PD by tail intravenous injection. CECm facilitated the blood-brain barrier (BBB) penetration of NPs, together with cholesterol escaped from reticuloendothelial system uptake, as well as L-DOPA was decarboxylated into dopamine which promoted the NPs toward the PD site for DA neuron regeneration. The behavioral tests demonstrated that the nanodecoys improved the efficacy of HDO on PD mice. These findings provide insights into the development of biomimetic nanodecoys loading HDO for precise therapy of PD. STATEMENT OF SIGNIFICANCE: The accumulation of α-synuclein (α-syn) aggregates is a hallmark of PD. Our previous study designed a specific antisense oligonucleotide (ASO) targeting human SNCA, but the traumatic intracerebroventricular (ICV) is not conducive to clinical application. Here, we further optimize the ASO by creating a DNA/DNA double-stranded molecule with cholesterol-conjugated, named Chol-HDO (coDNA), and develop a DA-targeted biomimetic nanodecoy Chol-HDO@LMNPs by engineering cerebrovascular endothelial cells membranes (CECm) with DSPE-PEG2000 and L-DOPA. The in vivo results demonstrated that tail vein injection of Chol-HDO@LMNPs could target DA neurons in the brain and ameliorate motor deficits in a PD mouse model. This investigation provides a promising peripheral delivery platform of L-DOPA-CECm nanodecoy loaded with a new Chol-HDO (coDNA) targeting DA neurons in PD therapy.
Assuntos
Doença de Parkinson , Camundongos , Humanos , Animais , Doença de Parkinson/genética , alfa-Sinucleína/genética , alfa-Sinucleína/metabolismo , Neurônios Dopaminérgicos/metabolismo , Levodopa , Oligonucleotídeos/farmacologia , Oligonucleotídeos/genética , Oligonucleotídeos/metabolismo , Biomimética , Células Endoteliais/metabolismo , DNA/metabolismoRESUMO
The lack of annotated datasets is a major bottleneck for training new task-specific supervised machine learning models, considering that manual annotation is extremely expensive and time-consuming. To address this problem, we present MONAI Label, a free and open-source framework that facilitates the development of applications based on artificial intelligence (AI) models that aim at reducing the time required to annotate radiology datasets. Through MONAI Label, researchers can develop AI annotation applications focusing on their domain of expertise. It allows researchers to readily deploy their apps as services, which can be made available to clinicians via their preferred user interface. Currently, MONAI Label readily supports locally installed (3D Slicer) and web-based (OHIF) frontends and offers two active learning strategies to facilitate and speed up the training of segmentation algorithms. MONAI Label allows researchers to make incremental improvements to their AI-based annotation application by making them available to other researchers and clinicians alike. Additionally, MONAI Label provides sample AI-based interactive and non-interactive labeling applications, that can be used directly off the shelf, as plug-and-play to any given dataset. Significant reduced annotation times using the interactive model can be observed on two public datasets.
Assuntos
Inteligência Artificial , Imageamento Tridimensional , Humanos , Imageamento Tridimensional/métodos , Algoritmos , SoftwareRESUMO
Acute inflammation has the potential for the recruitment of immune cells, inhibiting tumor angiogenesis, metastasis, and drug resistance thereby overcoming the tumor immunosuppressive microenvironment caused by chronic inflammation. Here, an acute inflammation inducer using bacteria outer membrane vesicles (OMVs) loaded in thermal-sensitive hydrogel (named OMVs-gel) for localized and controlled release of OMVs in tumor sites is proposed. OMVs trigger neutrophil recruitment and amplify acute inflammation inside tumor tissues. The hydrogel ensures drastic inflammation is confined within the tumor, addressing biosafety concerns that the direct administration of free OMVs may cause fatal effects. This strategy eradicated solid tumors safely and rapidly. The study further elucidates one of the possible immune mechanisms of OMVs-gel therapy, which involves the assembly of antitumor neutrophils and elastase release for selective tumor killing. Additionally, tumor vascular destruction induced by OMVs-gel results in tumor darkening, allowing for combinational photothermal therapy. The findings suggest that the use of OMVs-gel can safely induce acute inflammation and enhance antitumor immunity, representing a promising strategy to promote acute inflammation application in tumor immunotherapy.
RESUMO
Purpose: Two-dimensional single-slice abdominal computed tomography (CT) provides a detailed tissue map with high resolution allowing quantitative characterization of relationships between health conditions and aging. However, longitudinal analysis of body composition changes using these scans is difficult due to positional variation between slices acquired in different years, which leads to different organs/tissues being captured. Approach: To address this issue, we propose C-SliceGen, which takes an arbitrary axial slice in the abdominal region as a condition and generates a pre-defined vertebral level slice by estimating structural changes in the latent space. Results: Our experiments on 2608 volumetric CT data from two in-house datasets and 50 subjects from the 2015 Multi-Atlas Abdomen Labeling Challenge Beyond the Cranial Vault (BTCV) dataset demonstrate that our model can generate high-quality images that are realistic and similar. We further evaluate our method's capability to harmonize longitudinal positional variation on 1033 subjects from the Baltimore longitudinal study of aging dataset, which contains longitudinal single abdominal slices, and confirmed that our method can harmonize the slice positional variance in terms of visceral fat area. Conclusion: This approach provides a promising direction for mapping slices from different vertebral levels to a target slice and reducing positional variance for single-slice longitudinal analysis. The source code is available at: https://github.com/MASILab/C-SliceGen.
RESUMO
Transformer, one of the latest technological advances of deep learning, has gained prevalence in natural language processing or computer vision. Since medical imaging bear some resemblance to computer vision, it is natural to inquire about the status quo of Transformers in medical imaging and ask the question: can the Transformer models transform medical imaging? In this paper, we attempt to make a response to the inquiry. After a brief introduction of the fundamentals of Transformers, especially in comparison with convolutional neural networks (CNNs), and highlighting key defining properties that characterize the Transformers, we offer a comprehensive review of the state-of-the-art Transformer-based approaches for medical imaging and exhibit current research progresses made in the areas of medical image segmentation, recognition, detection, registration, reconstruction, enhancement, etc. In particular, what distinguishes our review lies in its organization based on the Transformer's key defining properties, which are mostly derived from comparing the Transformer and CNN, and its type of architecture, which specifies the manner in which the Transformer and CNN are combined, all helping the readers to best understand the rationale behind the reviewed approaches. We conclude with discussions of future perspectives.
Assuntos
Diagnóstico por Imagem , Redes Neurais de Computação , HumanosRESUMO
Immunotherapy has been emerging as a powerful strategy for cancer management. Recently, accumulating evidence has demonstrated that bacteria-based immunotherapy including naive bacteria, bacterial components, and bacterial derivatives, can modulate immune response via various cellular and molecular pathways. The key mechanisms of bacterial antitumor immunity include inducing immune cells to kill tumor cells directly or reverse the immunosuppressive microenvironment. Currently, bacterial antigens synthesized as vaccine candidates by bioengineering technology are novel antitumor immunotherapy. Especially the combination therapy of bacterial vaccine with conventional therapies may further achieve enhanced therapeutic benefits against cancers. However, the clinical translation of bacteria-based immunotherapy is limited for biosafety concerns and non-uniform production standards. In this review, we aim to summarize immunotherapy strategies based on advanced bacterial therapeutics and discuss their potential for cancer management, we will also propose approaches for optimizing bacteria-based immunotherapy for facilitating clinical translation.
Assuntos
Imunoterapia , Neoplasias , Humanos , Bactérias , Neoplasias/terapia , Antígenos de Bactérias , Vacinas Bacterianas , Microambiente TumoralRESUMO
Metabolic health is increasingly implicated as a risk factor across conditions from cardiology to neurology, and efficiency assessment of body composition is critical to quantitatively characterizing these relationships. 2D low dose single slice computed tomography (CT) provides a high resolution, quantitative tissue map, albeit with a limited field of view. Although numerous potential analyses have been proposed in quantifying image context, there has been no comprehensive study for low-dose single slice CT longitudinal variability with automated segmentation. We studied a total of 1816 slices from 1469 subjects of Baltimore Longitudinal Study on Aging (BLSA) abdominal dataset using supervised deep learning-based segmentation and unsupervised clustering method. 300 out of 1469 subjects that have two year gap in their first two scans were pick out to evaluate longitudinal variability with measurements including intraclass correlation coefficient (ICC) and coefficient of variation (CV) in terms of tissues/organs size and mean intensity. We showed that our segmentation methods are stable in longitudinal settings with Dice ranged from 0.821 to 0.962 for thirteen target abdominal tissues structures. We observed high variability in most organ with ICC<0.5, low variability in the area of muscle, abdominal wall, fat and body mask with average ICC≥0.8. We found that the variability in organ is highly related to the cross-sectional position of the 2D slice. Our efforts pave quantitative exploration and quality control to reduce uncertainties in longitudinal analysis.
RESUMO
Batch size is a key hyperparameter in training deep learning models. Conventional wisdom suggests larger batches produce improved model performance. Here we present evidence to the contrary, particularly when using autoencoders to derive meaningful latent spaces from data with spatially global similarities and local differences, such as electronic health records (EHR) and medical imaging. We investigate batch size effects in both EHR data from the Baltimore Longitudinal Study of Aging and medical imaging data from the multimodal brain tumor segmentation (BraTS) challenge. We train fully connected and convolutional autoencoders to compress the EHR and imaging input spaces, respectively, into 32-dimensional latent spaces via reconstruction losses for various batch sizes between 1 and 100. Under the same hyperparameter configurations, smaller batches improve loss performance for both datasets. Additionally, latent spaces derived by autoencoders with smaller batches capture more biologically meaningful information. Qualitatively, we visualize 2-dimensional projections of the latent spaces and find that with smaller batches the EHR network better separates the sex of the individuals, and the imaging network better captures the right-left laterality of tumors. Quantitatively, the analogous sex classification and laterality regressions using the latent spaces demonstrate statistically significant improvements in performance at smaller batch sizes. Finally, we find improved individual variation locally in visualizations of representative data reconstructions at lower batch sizes. Taken together, these results suggest that smaller batch sizes should be considered when designing autoencoders to extract meaningful latent spaces among EHR and medical imaging data driven by global similarities and local variation.
RESUMO
With the confounding effects of demographics across large-scale imaging surveys, substantial variation is demonstrated with the volumetric structure of orbit and eye anthropometry. Such variability increases the level of difficulty to localize the anatomical features of the eye organs for populational analysis. To adapt the variability of eye organs with stable registration transfer, we propose an unbiased eye atlas template followed by a hierarchical coarse-to-fine approach to provide generalized eye organ context across populations. Furthermore, we retrieved volumetric scans from 1842 healthy patients for generating an eye atlas template with minimal biases. Briefly, we select 20 subject scans and use an iterative approach to generate an initial unbiased template. We then perform metric-based registration to the remaining samples with the unbiased template and generate coarse registered outputs. The coarse registered outputs are further leveraged to train a deep probabilistic network, which aims to refine the organ deformation in unsupervised setting. Computed tomography (CT) scans of 100 de-identified subjects are used to generate and evaluate the unbiased atlas template with the hierarchical pipeline. The refined registration shows the stable transfer of the eye organs, which were well-localized in the high-resolution (0.5 mm3) atlas space and demonstrated a significant improvement of 2.37% Dice for inverse label transfer performance. The subject-wise qualitative representations with surface rendering successfully demonstrate the transfer details of the organ context and showed the applicability of generalizing the morphological variation across patients.
RESUMO
Features learned from single radiologic images are unable to provide information about whether and how much a lesion may be changing over time. Time-dependent features computed from repeated images can capture those changes and help identify malignant lesions by their temporal behavior. However, longitudinal medical imaging presents the unique challenge of sparse, irregular time intervals in data acquisition. While self-attention has been shown to be a versatile and efficient learning mechanism for time series and natural images, its potential for interpreting temporal distance between sparse, irregularly sampled spatial features has not been explored. In this work, we propose two interpretations of a time-distance vision transformer (ViT) by using (1) vector embeddings of continuous time and (2) a temporal emphasis model to scale self-attention weights. The two algorithms are evaluated based on benign versus malignant lung cancer discrimination of synthetic pulmonary nodules and lung screening computed tomography studies from the National Lung Screening Trial (NLST). Experiments evaluating the time-distance ViTs on synthetic nodules show a fundamental improvement in classifying irregularly sampled longitudinal images when compared to standard ViTs. In cross-validation on screening chest CTs from the NLST, our methods (0.785 and 0.786 AUC respectively) significantly outperform a cross-sectional approach (0.734 AUC) and match the discriminative performance of the leading longitudinal medical imaging algorithm (0.779 AUC) on benign versus malignant classification. This work represents the first self-attention-based framework for classifying longitudinal medical images. Our code is available at https://github.com/tom1193/time-distance-transformer.
RESUMO
Children with optic pathway gliomas (OPGs), a low-grade brain tumor associated with neurofibromatosis type 1 (NF1-OPG), are at risk for permanent vision loss. While OPG size has been associated with vision loss, it is unclear how changes in size, shape, and imaging features of OPGs are associated with the likelihood of vision loss. This paper presents a fully automatic framework for accurate prediction of visual acuity loss using multi-sequence magnetic resonance images (MRIs). Our proposed framework includes a transformer-based segmentation network using transfer learning, statistical analysis of radiomic features, and a machine learning method for predicting vision loss. Our segmentation network was evaluated on multi-sequence MRIs acquired from 75 pediatric subjects with NF1-OPG and obtained an average Dice similarity coefficient of 0.791. The ability to predict vision loss was evaluated on a subset of 25 subjects with ground truth using cross-validation and achieved an average accuracy of 0.8. Analyzing multiple MRI features appear to be good indicators of vision loss, potentially permitting early treatment decisions.Clinical relevance- Accurately determining which children with NF1-OPGs are at risk and hence require preventive treatment before vision loss remains challenging, towards this we present a fully automatic deep learning-based framework for vision outcome prediction, potentially permitting early treatment decisions.
Assuntos
Neurofibromatose 1 , Glioma do Nervo Óptico , Humanos , Criança , Glioma do Nervo Óptico/complicações , Glioma do Nervo Óptico/diagnóstico por imagem , Glioma do Nervo Óptico/patologia , Neurofibromatose 1/complicações , Neurofibromatose 1/diagnóstico por imagem , Neurofibromatose 1/patologia , Imageamento por Ressonância Magnética/métodos , Transtornos da Visão , Acuidade VisualRESUMO
Medical image segmentation, or computing voxel-wise semantic masks, is a fundamental yet challenging task in medical imaging domain. To increase the ability of encoder-decoder neural networks to perform this task across large clinical cohorts, contrastive learning provides an opportunity to stabilize model initialization and enhances downstream tasks performance without ground-truth voxel-wise labels. However, multiple target objects with different semantic meanings and contrast level may exist in a single image, which poses a problem for adapting traditional contrastive learning methods from prevalent "image-level classification" to "pixel-level segmentation". In this article, we propose a simple semantic-aware contrastive learning approach leveraging attention masks and image-wise labels to advance multi-object semantic segmentation. Briefly, we embed different semantic objects to different clusters rather than the traditional image-level embeddings. We evaluate our proposed method on a multi-organ medical image segmentation task with both in-house data and MICCAI Challenge 2015 BTCV datasets. Compared with current state-of-the-art training strategies, our proposed pipeline yields a substantial improvement of 5.53% and 6.09% on Dice score for both medical image segmentation cohorts respectively (p-value 0.01). The performance of the proposed method is further assessed on external medical image cohort via MICCAI Challenge FLARE 2021 dataset, and achieves a substantial improvement from Dice 0.922 to 0.933 (p-value 0.01).
Assuntos
Diagnóstico por Imagem , Aprendizado de Máquina , Humanos , Processamento de Imagem Assistida por Computador , Redes Neurais de Computação , Semântica , Diagnóstico por Imagem/métodos , Conjuntos de Dados como AssuntoRESUMO
BACKGROUND: Clinical evidence of immune checkpoint inhibitors combined with antiangiogenic drugs in patients with advanced non-small cell lung cancer (NSCLC) was limited. Recombinant human endostatin (rh-endostatin), an antiangiogenic drug, and camrelizumab, an anti-PD-1 antibody, have been approved for the treatment of advanced NSCLC in China. This study aimed to investigate the efficacy and safety of rh-endostatin plus camrelizumab and chemotherapy in the treatment of advanced NSCLC. METHODS: Eligible patients were enrolled and received camrelizumab (200 mg, day 1) every 3 weeks and continuous intravenous infusion of rh-endostatin (70 mg/day, days 1-3) and cisplatin combined with pemetrexed (for adenocarcinoma) or paclitaxel (for NSCLC other than adenocarcinoma) every 3 weeks. Primary endpoint was progression-free survival (PFS). Secondary endpoints were objective response rate (ORR), disease control rate (DCR), overall survival (OS), and safety profiles. RESULTS: Overall, 27 patients were included, and 25 patients were eligible for efficacy evaluation. For these 25 patients, ORR was 48.15% (13/27) and DCR was 85.19% (23/27). With a median follow-up of 10.37 months, the median PFS was 8.9 (95% CI: 4.23-13.57) months. Median OS was not reached. Overall, 96.3% of patients experienced at least one treatment-related adverse event, and grade 3 TRAEs occurred in 9 (33.3%) patients. No unexpected AEs were observed. CONCLUSION: Rh-endostatin plus camrelizumab and chemotherapy showed favorable efficacy and safety profile in patients with advanced NSCLC, representing a promising treatment regimen for these patients.
Assuntos
Adenocarcinoma , Carcinoma Pulmonar de Células não Pequenas , Neoplasias Pulmonares , Humanos , Endostatinas/efeitos adversos , Estudos Retrospectivos , Protocolos de Quimioterapia Combinada Antineoplásica/efeitos adversos , Inibidores da Angiogênese/uso terapêutico , Adenocarcinoma/tratamento farmacológicoRESUMO
Transformer-based models, capable of learning better global dependencies, have recently demonstrated exceptional representation learning capabilities in computer vision and medical image analysis. Transformer reformats the image into separate patches and realizes global communication via the self-attention mechanism. However, positional information between patches is hard to preserve in such 1D sequences, and loss of it can lead to sub-optimal performance when dealing with large amounts of heterogeneous tissues of various sizes in 3D medical image segmentation. Additionally, current methods are not robust and efficient for heavy-duty medical segmentation tasks such as predicting a large number of tissue classes or modeling globally inter-connected tissue structures. To address such challenges and inspired by the nested hierarchical structures in vision transformer, we proposed a novel 3D medical image segmentation method (UNesT), employing a simplified and faster-converging transformer encoder design that achieves local communication among spatially adjacent patch sequences by aggregating them hierarchically. We extensively validate our method on multiple challenging datasets, consisting of multiple modalities, anatomies, and a wide range of tissue classes, including 133 structures in the brain, 14 organs in the abdomen, 4 hierarchical components in the kidneys, inter-connected kidney tumors and brain tumors. We show that UNesT consistently achieves state-of-the-art performance and evaluate its generalizability and data efficiency. Particularly, the model achieves whole brain segmentation task complete ROI with 133 tissue classes in a single network, outperforming prior state-of-the-art method SLANT27 ensembled with 27 networks. Our model performance increases the mean DSC score of the publicly available Colin and CANDI dataset from 0.7264 to 0.7444 and from 0.6968 to 0.7025, respectively. Code, pre-trained models, and use case pipeline are available at: https://github.com/MASILab/UNesT.