Background: The epigenetic regulation of immune response has been demonstrated in recent studies. Nonetheless, potential roles of RNA N6-methyladenosine (m 6 A) modification in tumor microenvironment (TME) cell infiltration remain unknown. Methods: We comprehensively evaluated the m 6 A modification patterns of 1938 gastric cancer samples based on 21 m 6 A regulators, and systematically correlated these modification patterns with TME cell-infiltrating characteristics. The m6Ascore was constructed to quantify m 6 A modification patterns of individual tumors using principal component analysis algorithms. Results: Three distinct m 6 A modification patterns were determined. The TME cell-infiltrating characteristics under these three patterns were highly consistent with the three immune phenotypes of tumors including immuneexcluded, immune-inflamed and immune-desert phenotypes. We demonstrated the evaluation of m 6 A modification patterns within individual tumors could predict stages of tumor inflammation, subtypes, TME stromal activity, genetic variation, and patient prognosis. Low m6Ascore, characterized by increased mutation burden and activation of immunity, indicated an inflamed TME phenotype, with 69.4% 5-year survival. Activation of stroma and lack of effective immune infiltration were observed in the high m6Ascore subtype, indicating a non-inflamed and immuneexclusion TME phenotype, with poorer survival. Low m6Ascore was also linked to increased neoantigen load and enhanced response to anti-PD-1/L1 immunotherapy. Two immunotherapy cohorts confirmed patients with lower m6Ascore demonstrated significant therapeutic advantages and clinical benefits. Conclusions: This work revealed the m 6 A modification played a nonnegligible role in formation of TME diversity and complexity. Evaluating the m 6 A modification pattern of individual tumor will contribute to enhancing our cognition of TME infiltration characterization and guiding more effective immunotherapy strategies.

Cerebral microbleeds (CMBs) are small haemorrhages nearby blood vessels. They have been recognized as important diagnostic biomarkers for many cerebrovascular diseases and cognitive dysfunctions. In current clinical routine, CMBs are manually labelled by radiologists but this procedure is laborious, time-consuming, and error prone. In this paper, we propose a novel automatic method to detect CMBs from magnetic resonance (MR) images by exploiting the 3D convolutional neural network (CNN). Compared with previous methods that employed either low-level hand-crafted descriptors or 2D CNNs, our method can take full advantage of spatial contextual information in MR volumes to extract more representative high-level features for CMBs, and hence achieve a much better detection accuracy. To further improve the detection performance while reducing the computational cost, we propose a cascaded framework under 3D CNNs for the task of CMB detection. We first exploit a 3D fully convolutional network (FCN) strategy to retrieve the candidates with high probabilities of being CMBs, and then apply a well-trained 3D CNN discrimination model to distinguish CMBs from hard mimics. Compared with traditional sliding window strategy, the proposed 3D FCN strategy can remove massive redundant computations and dramatically speed up the detection process. We constructed a large dataset with 320 volumetric MR scans and performed extensive experiments to validate the proposed method, which achieved a high sensitivity of 93.16% with an average number of 2.74 false positives per subject, outperforming previous methods using low-level descriptors or 2D CNNs by a significant margin. The proposed method, in principle, can be adapted to other biomarker detection tasks from volumetric medical data.

BackgroundDistal radius fracture is common in the general population. Fracture management includes a plaster cast, splint and synthetic material cast to immobilise the injured arm. Casting complications are common in those conventional casting technologies. 3D printing technology is a rapidly increasing application in rehabilitation. However, there is no clinical study investigating the application of a 3D–printed orthopaedic cast for the treatment of bone fractures. We have developed a patient-specific casting technology fabricated by 3D printing. This pioneering study aims to use 3D–printed casts we developed for the treatment of distal radius fractures, to provide the foundation for conducting additional clinical trials, and to perform clinical assessments.MethodTen patients with ages between 5 and 78 years are involved in the clinical trial. Patients are applied 3D–printed casts we developed. Orthopaedic surgeons carried out a six-week follow-up to examine clinical outcomes. Two questionnaires were developed for the assessment of clinical efficacy and patients’ satisfaction. These questionnaires are completed by physicians and participating patients.ResultsA 3D–printed cast creates a custom-fitted design to maintain the fractured bone alignment. No loss of reduction is found in all patients. Compartment syndrome and pressure sores are not present. Patient comfort gets positive scores on the questionnaire. All (100%) of the patients opt for the 3D–printed cast instead of the conventional plaster cast.DiscussionA patient-specific, 3D–printed cast offers a proper fit to immobilise an injured arm and holds the fracture reduction appropriately. A custom-fitted structure reduces the risk of pressure-related complications due to the high and concentrated local stress. The ventilated and lightweight design minimises interference with a patient’s daily activities and reduces the risk of cutaneous complications. Patients express a strong preference for using a 3D–printed cast instead of a plaster cast. Limitations of the novel cast include a slight odour after heavy sweating and the relatively high cost due to the limitations of current 3D printing technologies.ConclusionsThis pioneering study is the first clinical trial on the application of a 3D–printed cast for the treatment of forearm fractures. The novel casting technology heals the fracture effectively without casting complications. The 3D–printed cast is patient-specific and ventilated as well as lightweight, and it features both increased patient comfort and satisfaction.

Acute ischemic stroke is recognized as a common cerebral vascular disease in aging people. Accurate diagnosis and timely treatment can effectively improve the blood supply of the ischemic area and reduce the risk of disability or even death. Understanding the location and size of infarcts plays a critical role in the diagnosis decision. However, manual localization and quantification of stroke lesions are laborious and time-consuming. In this paper, we propose a novel automatic method to segment acute ischemic stroke from diffusion weighted images (DWIs) using deep 3-D convolutional neural networks (CNNs). Our method can efficiently utilize 3-D contextual information and automatically learn very discriminative features in an end-to-end and data-driven way. To relieve the difficulty of training very deep 3-D CNN, we equip our network with dense connectivity to enable the unimpeded propagation of information and gradients throughout the network. We train our model with Dice objective function to combat the severe class imbalance problem in data. A DWI data set containing 242 subjects (90 for training, 62 for validation, and 90 for testing) with various types of acute ischemic stroke was constructed to evaluate our method. Our model achieved high performance on various metrics (Dice similarity coefficient: 79.13%, lesionwise precision: 92.67%, and lesionwise F1 score: 89.25%), outperforming the other state-of-the-art CNN methods by a large margin. We also evaluated the model on ISLES2015-SSIS data set and achieved very competitive performance, which further demonstrated its generalization capacity. The proposed method is fast and accurate, demonstrating a good potential in clinical routines.

The generalization error bounds found by current error models using the number of effective parameters of a classifier and the number of training samples are usually very loose. These bounds are intended for the entire input space. However, support vector machine (SVM), radial basis function neural network (RBFNN), and multilayer perceptron neural network (MLPNN) are local learning machines for solving problems and treat unseen samples near the training samples to be more important. In this paper, we propose a localized generalization error model which bounds from above the generalization error within a neighborhood of the training samples using stochastic sensitivity measure. It is then used to develop an architecture selection technique for a classifier with maximal coverage of unseen samples by specifying a generalization error threshold. Experiments using 17 University of California at Irvine (UCI) data sets show that, in comparison with cross validation (CV), sequential learning, and two other ad hoc methods, our technique consistently yields the best testing classification accuracy with fewer hidden neurons and less training time.

No abstract

Background and Purpose-We explored the association between pulsatility index (PI) as derived from transcranial Doppler ultrasound with various measures of small vessel disease in the community.

Background/AimsPatients with irritable bowel syndrome (IBS) are characterized by abnormal central processing with altered brain activation in response to visceral nociceptive signals. The effect of electroacupuncture (EA) on IBS patients is unclear. The study is set to study the effect of EA on brain activation during noxious rectal distension in IBS patients using a randomized sham-controlled model.MethodsThirty IBS-diarrhea patients were randomized to true electroacupuncture or sham acupuncture. Functional MRI was performed to evaluate cerebral activation at the following time points: (1) baseline when there was rectal distension only, (2) rectal distension during application of EA, (3) rectal distension after cessation of EA and (4) EA alone with no rectal distension. Group comparison was made under each condition using SPM5 program.ResultsRectal distension induced significant activation of the anterior cingulated cortex, prefrontal cortex, thalamus, temporal regions and cerebellum at baseline. During and immediately after EA, increased cerebral activation from baseline was observed in the anterior cingulated cortex, bilateral prefrontal cortex, thalamus, temporal regions and right insula in both groups. However, true electroacupuncture led to significantly higher activation at right insula, as well as pulvinar and medial nucleus of the thalamus when compared to sham acupuncture.ConclusionsWe postulate that acupuncture might have the potential effect of pain modulation in IBS by 2 actions: (1) modulation of serotonin pathway at insula and (2) modulation of mood and affection in higher cortical center via ascending pathway at the pulvinar and medial nucleus of the thalamus.