Search

Ultrasonography (USG) has become an invaluable tool in the assessment of neurocritical patients in the operating theaters and critical care units. Due to its easy availability, reliability, safety, and repeatability, neuro-intensivists and neuro-anesthesiologists utilize USG to make a diagnosis, assess prognosis, and decide upon treatment. In neurocritical care units, USG has myriad indications for use, both systemic and neurologic. The neurological indications include the assessment of stroke, vasospasm, traumatic brain injury, brain death, acute brain damage, optic nerve sheath diameter, and pupillary reflexes to name a few. The systemic indications range from assessment of cardio-pulmonary function and intravascular volume status to detection of deep venous thromboses, vocal cord assessment in intubated patients, placement of central venous catheters, and percutaneous tracheostomy. In this narrative review, we iterate the clinical applications of USG in neuroanesthesia and neurocritical care, which we penned after searching relevant databases in PubMed, Medline, Ovid, and Google Scholar by using terms such as 'applications of transcranial Doppler', 'optic nerve sheath diameter', 'USG applications in the critical care unit', and so on. Our search database includes several research papers, neurocritical care books, review articles, and scientific databases. This article reviews various applications of USG in neuroanesthesia, neurosurgery, and neurocritical care. [ABSTRACT FROM AUTHOR]

• This article summarized the development course of biodegradable cardiovascular stents. • The present in vivo experimental results of Mg-based stents were briefly reviewed. • The strategies for surface endothelialization of cardiovascular stents were summarized. • The recent advances in surface endothelialization of magnesium alloy stent materials were reviewed. • The key issues and future direction of magnesium alloy stents were proposed in the present article. Magnesium and its alloy have good mechanical properties and biodegradability, and have become the hotspot of the next-generation biodegradable vascular stent materials. However, their rapid degradation in vivo and poor biocompatibility are still the bottlenecks of clinical applications for the cardiovascular stents. In particular, how to induce the repair and regeneration of the vascular endothelial with normal physiological functions on the surface of the magnesium alloy stent materials represents the key to its clinical application in the field of cardiovascular stents. It has been believed that it is an ideal way to completely solve the postoperative complications through constructing the multifunctional anti-corrosive bioactive coating on the magnesium alloy surface to induce the formation of vascular endothelium with normal physiological functions. However, how to construct a corrosion-resistant multifunctional bioactive coating with the good endothelial regeneration abilities on the magnesium alloy surface still faces a great challenge. This paper mainly focused on highlighting and summarizing the recent advances in the surface endothelialization of the magnesium alloy materials for the vascular stent, including the bio-inert coating, in-situ immobilization of bioactive molecules on the surface, polymer coating loaded with bioactive factors, novel multifunctional polymer coating, bioactive micropatterns, bioactive layer with glycocalyx-like structure, NO-releasing coating and bioactive sol-gel coating. The advantages and disadvantages of these strategies were discussed and analyzed. Finally, in the senses of future development and clinical application, this paper analyzed and summarized the development direction and prospect of surface endothelialization of the magnesium alloy vascular stents. It is anticipated that this review can give the new cues to the surface endothelialization of the cardiovascular magnesium alloy stents and promote future advancements in this field. [ABSTRACT FROM AUTHOR]

Finite element (FE) analysis has emerged as the primary tool to investigate complex problems that are difficult to test clinically on living tissues, yielding valuable but not necessarily applicable information on orthodontic treatment modalities. Many studies are being published but the findings are difficult to interpret, lack in relevance, or cannot be critically appraised. In this paper, the FE modeling design and impact on the quality and outcome of research are discussed, with focus on how to narrow the gap between theory and clinical application. [ABSTRACT FROM AUTHOR]

Young patients with a severely traumatized maxillary central incisor have limited possibilities to restore its loss. Autotransplantation of immature premolars has been documented as a predictable long-term outcome with the potential for preservation and regeneration of alveolar bone. They can also be successfully reshaped to the central incisor's morphology using direct composite build-ups. However, tooth transplantation is a technique sensitive procedure and requires good case selection, an understanding of relevant tooth transplantation biological principles and careful execution of the optimal surgical and follow-up protocols. Orthodontists play an important role in cooperation with the surgeon before and after surgery. This includes selection of suitable candidates based on orthodontic evaluation of occlusion and indications for premolar extraction, opening adequate space before transplantation and post-surgical alignment of the transplant for an optimal reshaping and establishing a stable occlusion. The aim of the paper is to present clinical application of autotransplanation of immature premolars to replace a traumatized central incisor in growing patients and to discuss the role of an orthodontist in the interdisciplinary team. [ABSTRACT FROM AUTHOR]

• Digital Mental Health Services (DMHS) can greatly increase access to care. • Training is needed so providers can effectively integrate DMHS into their treatments. • More research on DMHS is needed, along with regulatory standards. • DMHS may help reduce health disparities but questions remain about cultural tailoring. The papers in this special series make a compelling case for the value of digital mental health services (DMHS; including technology-based interventions, assessments, and prevention programs) to help address some of the currently unmet needs in mental health care. At the same time, the papers highlight the work that needs to be accomplished for DMHS to fulfill their promise. We review the papers' contributions in terms of (a) the imperative to increase access to evidence-informed, high-quality care, especially for underserved populations, both in the United States and globally; (b) ways to use DMHS to improve the ways that clinical care is provided to make treatment provision more effective and efficient; and (c) the current state of the research on DMHS for emotional disorders. We then consider lessons learned and recommendations to move the field forward, such as increasing (and making transparent) the research base on DMHS, adopting regulatory standards for DMHS, attending carefully to training issues for DMHS and best practices for dissemination and implementation, designing specifically for digital platforms, and being intentional about efforts to reduce disparities regarding who benefits from DMHS. [ABSTRACT FROM AUTHOR]

The aim of this paper is to discuss the significance of the intersectionality framework for addressing prejudices, racism and inequalities in nursing education and clinical learning environments. Discrimination and racism against nursing students and educators based on their gender, ethnicity, race and social identities is well-documented in the nursing literature. Despite documented discrimination and incivility based on intersectional factors, it is reported that often nurse educators show limited interest in the culture, diverse experiences and values of nursing students with culturally and linguistically diverse backgrounds. Discussion paper The discussion was based on contemporary literature about intersectionality, discrimination and racism in nursing. We completed a cursory search of literature in nursing education journal and selected nursing and health science databases. This was not a formal literature review. Using a fictional example, the application of intersectionality to address inequalities in educational settings is illustrated. Intersectionality is an invaluable tool for examining interwoven power relations and power struggles arising from racial, gender, ethnic, religious and sexuality and disability-related differences. Nurse educators, students and leaders should be more cognizant of their preconceived views, sociocultural stereotypes and varied forms of sociocultural oppression affecting their interactions with each other in clinical learning environments. Incorporating intersectionality can address prejudices, racism and inequalities arising due to sociocultural, ethnic, power-related and intergenerational issues among educators, students and other personnel involved in creating clinical learning environments. [ABSTRACT FROM AUTHOR]

Accurate 3D tracking of hand and fingers movements poses significant challenges in computer vision. The potential applications span across multiple domains, including human–computer interaction, virtual reality, industry, and medicine. While gesture recognition has achieved remarkable accuracy, quantifying fine movements remains a hurdle, particularly in clinical applications where the assessment of hand dysfunctions and rehabilitation training outcomes necessitate precise measurements. Several novel and lightweight frameworks based on Deep Learning have emerged to address this issue; however, their performance in accurately and reliably measuring finger movements requires validation against well-established gold standard systems. In this paper, the aim is to validate the hand-tracking framework implemented by Google MediaPipe Hand (GMH) and an innovative enhanced version, GMH-D, that exploits the depth estimation of an RGB-Depth camera to achieve more accurate tracking of 3D movements. Three dynamic exercises commonly administered by clinicians to assess hand dysfunctions, namely hand opening–closing, single finger tapping and multiple finger tapping are considered. Results demonstrate high temporal and spectral consistency of both frameworks with the gold standard. However, the enhanced GMH-D framework exhibits superior accuracy in spatial measurements compared to the baseline GMH, for both slow and fast movements. Overall, our study contributes to the advancement of hand tracking technology, and the establishment of a validation procedure as a good-practice to prove efficacy of deep-learning-based hand-tracking. Moreover, it proves that GMH-D is a reliable framework for assessing 3D hand movements in clinical applications. • Hands are vital for carrying out daily-life tasks without external aid. • Measuring objectively hand dexterity can support new digital health treatments. • Deep Learning allows for video-based hand tracking of fine motor tasks. • We validate accuracy of Google Mediapipe Hand and GMH-D against motion capture. • Both frameworks achieve good-to-high spatial and temporal hand tracking performance. [ABSTRACT FROM AUTHOR]

This paper discusses the historical construction of the Expanded Family Health Center (NASF, in Portuguese), based on the analysis of 17 documents edited by the Ministry of Health (MH) between 2005 and 2021. This is a qualitative study of documental review that seeks to understand how the regulations and official instructive manuals have been shaping the way NASF teams operate. It proposes to divide the NASF construction process into five periods: previous movements (2003 to 2007); support guidelines (2008 to 2011); the universalization of nasf (2012 to 2015); expansion of support (2016 to 2018); and the dismantling of NASF? (2019 to 2021). The results show changes in guidelines over the years of the team's existence, especially in relation to the matrix support concept and its two dimensions: technical-pedagogical and clinical care. This study also demonstrates the effects of the Previne Brasil Program on the NASF, which resulted in the reduction of 379 teams in 2020 and 2021. Added to this scenario is the SARS-CoV-2 pandemic, which may be repositioning NASF interventions in the Brazilian Unified Health System (SUS, in Portuguese). [ABSTRACT FROM AUTHOR]

Atrous convolution and attention have improved the performance of the UNet architecture for segmentation purposes. However, a perfect combination of atrous convolution and attention to improve brain tumor segmentation performance is still an interesting challenge. In this paper, we propose UNet architecture with the addition of attention in the skip connection and the replacement of the processing block with two atrous convolution sequences connected to the attention unit combined with one residual path called the Multipath Residual Attention Block (MRAB). The architecture was trained using the Brain Tumor Segmentation(BraTS) 2018, 2019, 2020, and 2021 challenge datasets. The ensembled model was validated online and obtained dice scores of 77.71%, 79.77%, 89.59% for BraTS2018, 74.91%, 80.98%, 88.48% for BraTS2019, 72.91%, 80.19%, 88.57% for BraTS2020, and 77.73%, 82.19%, 89.33% for BraTS2021 validation datasets for Enhanced Tumor(ET), Tumor Core(TC), and Whole Tumor(WT) areas, respectively. These dice score performances outperformed state-of-the-art brain tumor segmentation architectures and promised to be developed for clinical application. [ABSTRACT FROM AUTHOR]

Recent trends and current market analysis suggest demand for herbal formulations, including herbal creams, is increasing over synthetic creams due to their lower toxicity and fewer side effects, followed by more skin nourishment capability, according to several published studies. Moreover, published literature lacks any updated, compiled explanatory review on herbal creams. Therefore, we designed the present communication to accomplish the current requirement. The major objective of the present communication is to provide the clinical data-based evidence and state of the art adopted for the formulation and development of herbal cream, including its composition and manufacturing processes, to establish therapeutic effectiveness. Both online and offline literature searches were performed and screened. The publications accounted for in vitro and clinical studies, followed by various preparation methods, classification and manufacturing process were included. There were a total of 150 papers finally selected and reviewed in the present communications addressing concerns related to herbal cream from 1995 to July 2023. It was established that herbal creams can be produced using a variety of techniques and methods. Majorly anti-inflammatory, dermatitis, anti-trauma, anti-aging, anti-acne, and vulvovaginal herbal creams were enlisted in terms of different clinical and therapeutic application. The evidence from the present systematic review suggests that more human clinical trials with larger sample sizes are needed in order to achieve higher accuracy in terms of the safety and efficacy of herbal creams. [Display omitted] • Composition, preparation and classification of herbal cream. • Manufacturing of herbal cream. • Clinical and therapeutic application of herbal cream from 1995 to 2023. • Regulations and the future perspective of herbal creams. [ABSTRACT FROM AUTHOR]

The ability to automatically segment anatomical targets on medical images is crucial for clinical diagnosis and interventional therapy. However, supervised learning methods often require a large number of pixel-wise labels that are difficult to obtain. This paper proposes a weakly supervised glottis segmentation (WSGS) method for training end-to-end neural networks using only point annotations as training labels. This method functions by iteratively generating pseudo-labels and training the segmentation network. An automatic seeded region growing (ASRG) algorithm is introduced to generate quality pseudo labels to diffuse point annotations based on network prediction and image features. Additionally, a novel loss function based on the structural similarity index measure (SSIM) is designed to enhance boundary segmentation. Using the trained network as its core, a glottis state monitor is developed to detect the motion behavior of the glottis and assist the anesthesiologist. Finally, the performance of the proposed approach was evaluated on two datasets, achieving an average mIoU and accuracy of 82.7% and 91.3%. The proposed monitor was demonstrated to be effective, which holds significance in clinical applications. • Weakly supervised glottis segmentation method only with point annotations. • Geometry-constrained diffusion strategy for quality pseudo-label generation. • Development of a glottis state monitor to detect the motion behavior of the glottis. [ABSTRACT FROM AUTHOR]

The surgical repair of full-thickness defects involving the alae nasi is complex. Pedicle flaps such as frontal and nasolabial flaps can be used, but require several interventions with different techniques. In contrast, free flaps from the foot of the ear helix allow the three layers of the nasal wings to be reconstructed in a single operation. Nevertheless, in the classical approach, the vascular pedicle is short. Although some authors have proposed raising the flap in a retrograde manner, this still yields a relatively short pedicle with narrow vessels. In the companion paper, we demonstrated that a posterior auricular artery helix root free flap (PAAHF) can be harvested from the posterior auricular vessels, thus increasing the useful pedicle length. The case of a patient with basal cell carcinoma of the left ala is presented here. A right helix root free flap was anastomosed with the facial vessels at the left mandibular notch. This new flap overcomes the main limitation of the classical helix root flap, namely the length of the pedicle. It has all of the morphological qualities of the classical flap, but with simpler vascular assembly, since autologous venous grafts and complex anastomoses are not required. [ABSTRACT FROM AUTHOR]

Personal data is highly affected by the witnessed digital transformation of healthcare processes. This process relies deeply on the connectivity and decentralization of healthcare systems and data repositories. In this context, value creation and quality enhancement are obviously leveraged, however both health providers and individuals could be exposed to many risks ranging from privacy violations to medical identity theft and personal harm. Hence, it is essential that healthcare stakeholders ensure privacy protection and systemic compliance to personal data regulations such as HIPPA (Health Insurance Portability and Accountability Act) and GDPR (General Data Protection Regulation). Taking clinical processes as a starting point is very important to highlight the personal data in use and to assess whether such usage is justifiable and subsequently allow privacy management decisions to be made. In this paper we combine BPMN (Business Process Model and Notation) and DMN (Decision Model and Notation) to model clinical care pathways as standard business processing constituting the hospital information system. Business process modelling presents a useful mean to model clinical care pathways. It allows a complete discovery of data processing scenarios. DMN (Decision Model and Notation) is implemented in BPMN models to present the rules that lead to a decision in easy-to-read tables which are executed directly by a decision engine. In addition, the integration of verifiable security labels of the manipulated data, we make sure compliance to legislation is ensured at the level of decision rules for each decision table of the DMN. [ABSTRACT FROM AUTHOR]

Aim: Demonstrating the validated tool to compress CSR writing time with capabilities that facilitate data retrieval from multiple sources to assist in compilation, evaluation and interpretation of information that supports the content development Method: Generates the pre-filled CSR automatically using AI techniques such as ML/NLP/NLG. Most of the CSR content assembled from source documents such as protocol, SAP etc. The tool automatically writes the CSR content from the documents mentioned above. Results / Discussion: Creating Regulatory documents such as eCTD modules 2.7.3 & 2.7.4 (Clinical summary efficacy & safety respectively), Investigator brochure, Clinical Study Report (CSR) etc., is highly manual and time consuming for medical writers. Good percentage of contents for the above-mentioned documents comes from various source documents such as Protocol, SAP, Safety Narratives, In-text tables, Integrated summary of safety and efficacy etc. Automating these scientific documents writing by utilizing AI techniques such as ML and Natural language processing/understanding /Generation (NLP/NLU/NLG) will reduce the efforts significantly. This paper will demonstrate the tool that we developed for automating the scientific document to CSR writing by using Artificial Intelligence (AI) techniques. The CSR template follows the ICH-E3 guideline, and the tool can also accommodate sponsor-defined templates. Conclusion: The benefits of this application are lean writing, multi-person authoring, Traceability report, Interpretation of tables in simple English, Post-text to In-text table conversion, authoring safety narratives and Integration of sponsors workflow. The author can visualize the consolidated comments and edits from multiple reviewer(s) in one place. These features can save lot of time for medical writers so that, they can focus on discussion points and interpretation of study results. [ABSTRACT FROM AUTHOR]

Aim: To determine whether the wound records on electronic recording platforms guarantee the continuity of care, providing health indicators in the treatment of wounds and in the quality of care provided. Method: A systematic review was conducted in accordance with the Joanna Briggs Institute (JBI) recommendations. The search was conducted in September 2022 on 6 electronic databases: B-On, BVS, EBSCO/CINAHL, MEDLINE/PubMed, Scielo, Cochrane, using the descriptors "Wound healing", "Electronic health records", "Outcome assessment", "Pressure ulcer", "Wounds", having been selected 6 articles. Results / Discussion: It was found that electronic records in the assessment and treatment of wounds are fundamental regarding internal regulations, continuity of care, reducing the level of documentation error, reducing running costs and increasing wound care capacity. When comparing the use of paper and electronic registers, there was a wide preference of patients and health professionals in the electronic version of the records. The use of electronic health records, although at an early stage, is extremely useful in nursing care, although it is necessary to ensure that they are intuitive and user-friendly, so as to not further increase the complexity of the records and workload of health professionals. Conclusion: The use of electronic records benefits the continuity, quality and safety of the care provided and encourages communication between health professionals. As future projects, we intend to make our contribution to professional practice, incorporating the results of this study into clinical practice, through the implementation of a project aimed at creating a platform for recording wounds. [ABSTRACT FROM AUTHOR]

• A promising deep learning based approach was presented to detect cancer metastasis in lymph node images with higher accuracy. • The proposed deep learning framework uses ResNet architectures, transfer learning and 1cycle policy. • The validity of the proposed approach was evaluated on PCAM dataset consisted of 220,025 lymph node images. • The experiments demonstrated that the presented method outperformed most of the current studies. Lymph node metastases are one of the most indicator of some cancer types such as breast, colon and prostate. Breast cancer mostly spreads to lymph nodes in the armpit and it is one of the most common causes of death in women worldwide. Pathologists need more attention and time to diagnose metastasis in digitized lymph node images and also this process tends to be misinterpreted. In this paper, a promising deep learning based approach was presented to detect cancer metastasis in lymph node images with higher accuracy. The proposed deep learning framework uses ResNet architectures, transfer learning and 1cycle policy which is the method of finding the optimal learning rate. The validity of the proposed approach was evaluated on PCAM dataset consisted of 220,025 lymph node images. The experiments demonstrated that the presented method outperformed most of the current studies, achieving accuracy of 98.60% on the PCAM dataset by using ResNet models and fine-tuning techniques effectively. The burden of diagnosis procedure of metastasis can be lessened with the proposed deep learning framework in the clinical applications. [ABSTRACT FROM AUTHOR]

Automatic multi-organ segmentation in medical images is crucial for many clinical applications. The art methods have reported promising results but rely on massive annotated data. However, such data is hard to obtain due to the need for considerable expertise. In contrast, obtaining a single-organ dataset is relatively easier, and many well-annotated ones are publicly available. To this end, this work raises the partially supervised problem: can we use these single-organ datasets to learn a multi-organ segmentation model? In this paper, we propose the P a r t i al- and M utual- P rior incorporated framework (PRIMP) to learn a robust multi-organ segmentation model by deriving knowledge from single-organ datasets. Unlike existing methods that largely ignore the organs' anatomical prior knowledge, our PRIMP is designed with two key prior shared across different subjects and datasets: (1) partial-prior, each organ has its own character (e.g. , size and shape) and (2) mutual-prior, the relative position between different organs follows the comparatively fixed anatomical structure. Specifically, we propose to incorporate partial-prior of each organ by learning from the single-organ statistics, and inject mutual-prior of organs by learning from the multi-organ statistics. By doing so, the model is encouraged to capture organs' anatomical invariance across different subjects and datasets, thus guaranteeing the anatomical reasonableness of the predictions, narrowing down the problem of domain gaps, capturing spatial information among different slices, thereby improving organs' segmentation performance. Experiments on four publicly available datasets (LiTS, Pancreas, KiTS, BTCV) show that our PRIMP can improve the performance on both the multi-organ and single-organ datasets (17.40% and 3.06% above the baseline model on DSC, respectively) and can surpass the comparative approaches. • We propose PRIMP to learn multi-organ segmentation model from single-organ datasets. • PRIMP can effectively capture anatimical priors of single- and multi-organ. • Experiments on four publicly available datasets show PRIMP's effectiveness. [ABSTRACT FROM AUTHOR]

• Extract inter-slice spatial information in the form of 2.5D. • Proposes a lightweight Inception convolution structure with residual connections to significantly reduce the network's parameters. • Employ a combination of BCE and Dice loss to achieve fast convergence and low fluctuations in network training. • Evaluate the proposed method on publicly available datasets, LiTS17 and 3Dircadb. One critical factor that restricts the clinical application of computer-aided liver and tumor segmentation is the method's high complexity and low accuracy. Overcoming this limitation is what we are concerned about in this study. This paper presented a new 2.5D lightweight network for fast and accurate liver and tumor segmentation from CT images. The method is grounded in the U-Net framework, which leverages the techniques from the residual and Inception theories. We first adopted the 2.5D training mode for CNN networks to improve the utilization of spatial information. Then, we designed an improved U-type architecture to substantially reduce the parameters by introducing residual block and InceptionV3, named RIU-Net. Finally, a hybrid loss function combined BCE and Dice is employed to speed up the convergence and improve accuracy. We evaluated the proposed method on two publicly available databases, LiTS17 and 3DIRCADb. The performance of our approach is compared with five closely related techniques. Our result outperforms the others on both accuracy and time cost. Specifically, the total number of parameters is reduced by 70% compared to U-Net. Both quantitative and qualitative results demonstrated the superior applicability of our method and thus proved to be a promising lightweight tool for computer-aided liver and tumor segmentation.. [ABSTRACT FROM AUTHOR]

• A dual attention mechanism is proposed which can highlight the meaningful local patterns contained in the skin lesion regions, enhancing the feature representation and the interpretability of the proposed network at the same time. • An auxiliary learning mechanism which contains auxiliary supervision and KL divergence based regularization is proposed. The KL regularization can make the two auxiliary supervision branches collaborate with each other during training through mutual knowledge transferring, and guide the network to extract the meaningful local pattern features contained in the skin lesion region in a weakly supervised manner. Besides, it brings in strong regularization which makes our proposed network avoid over-fitting when training on the small training data. • The proposed network gained the state-of-the-art performance for skin lesion classification, regardless of binary- or multi-classification. Besides, the robustness and interpretability of the proposed network are strong which can promote its clinical application. Skin lesion varies greatly in appearance, and its classification task suffers from large inter-class similarity and intra-class variation, thus the subtle differences of local pattern contained in the skin lesion regions are critical for its classification. In this paper, we propose a dual attention based network for skin lesion classification with auxiliary learning. The dual attention mechanism includes the spatial attention (SA) and the channel attention (CA) modules. The SA module can focus on the skin lesion region feature with reduced irrelevant artifacts features. In the subsequent CA module, it first captures the non-local based global feature of the skin lesion region and then generates the feature channels reweighting vector, which is used to further refine the meaningful local pattern feature contained in the skin lesion region. Therefore, the performance and the interpretability of the proposed network are enhanced at the same time. The proposed auxiliary learning contains two auxiliary supervision branches and KL regularization. The KL regularization makes the two auxiliary supervision branches collaborate with each other during training through mutual knowledge transferring. The introduced strong regularization can guide the dual attention mechanism to focus on the meaningful local pattern features in a weakly supervised manner and make the network avoid overfitting on small training data. Without extra training data, our proposed network can outperform current competition winners on several datasets, regardless of binary- or multi-classification. The proposed network is robust enough and owns strong interpretability which promotes its clinical application. [ABSTRACT FROM AUTHOR]

• A two-view and multi-scale supervision network is proposed for pancreas segmentation. • A location branch for localization and a segmentation branch for segmentation. • Multi-scale supervision is used to learn multi-scale features. • Effectiveness verified on two pancreas datasets and one spleen dataset. Automatic organ segmentation systems can accelerate the development of computer-aided diagnosis (CAD) in clinical applications. In this paper, we focus on the challenging pancreas segmentation task. The tiny size, poor contrast, and blurred boundaries of the pancreas make it hard to detect. Current approaches emphasize decomposing this task into subtasks (localization and segmentation) and using the same network to solve different tasks. However, they overestimate the generalization ability of their models. In addition, current methods rely too much on the result of localization. To address these challenges, we propose a novel network by two-view feature learning based on attention mechanism and multi-scale supervision, which we term TVMS-Net. For localization, we adopt Attention Gate (AG) to distinguish appearance features of the pancreas in shallow layers. For segmentation, a simple and effective Residual Multi-Scale Dilated Attention (RMSA) module is designed to extract comprehensive inter-channel relationships and multi-scale spatial information. TVMS-Net is supervised in multi-scale to learn specific-level semantic information. Experimental results on two pancreas datasets show that TVMS-Net obtains remarkable performance. Importantly, TVMS-Net also achieves excellent segmentation accuracy on another tiny organ dataset, i.e., the spleen, which justifies the reliability and robustness of our method. [ABSTRACT FROM AUTHOR]

• A novel method for secured ECG transmission to cardiologist for clinical healthcare has been proposed. • 3-DES has been implemented for encryption and water cycle optimization (WCO) algorithm is employed for authentication. • The encrypted and authenticated ECG features are transmitted to cardiologist for analysis via an IoT platform. • The significant of the proposed technique attains better value of results as compared to the existing methods. Telecardiology offers early detection, consultations and disease management in Golden Hours to patients. Transmitting ECG data to a cardiologist without security may incline to misdiagnosis. In this paper, a novel method for secured ECG transmission to cardiologist for clinical healthcare has been proposed. Firstly, ECG datasets are taken from the MIT-BIH database and pre-processed. Savitzky-Golay (SG) filter has been used for the removal of low-frequency noise, while high frequency noise is eliminated using maximal overlap wavelet packet transform (MOWPT) followed by features extraction. For security, triple data encryption standard (3-DES) has been implemented for encryption and water cycle optimization (WCO) algorithm is employed for authentication. Then encrypted and authenticated ECG features are transmitted to cardiologist for analysis via ThingSpeak, an internet of things (IoT) platform. The performance of the proposed method is calculated viz. avalanche effect is 50.12% and mean square error is 0.463, number of pixels change rate (NPCR) is 100, unified averaged changed intensity (UACI) is 39.698 and execution time is 0.003 msec. The significant of the proposed technique provides minimum error with less processing time, better values of avalanche effect, NCPR and UACI as compared to the existing methods. [ABSTRACT FROM AUTHOR]

• A multi-branch network improves the resolution of anisotropic MRI image. • Transforming adjacent slices feature to enhance the resolution of target slice. • Spatial attention adaptively highlights meaningful features. • Hybrid loss encourages the learning of fine contents and structures. Magnetic resonance imaging (MRI) is widely used in clinical applications. However, due to the limitations in signal-to-noise ratio, physical properties of the scanner and scanning time, MRI images are usually acquired in low resolution, which restrains the accuracy of segmentation and recognition tasks. Recently, convolutional neural network (CNN) super-resolution methods have shown great potential in improving the resolution of MRI. Unfortunately, current methods neglect the data continuity and prior information of MRI images. In this paper, we handle the anisotropic 3D brain MRI images SR task as the problem of inserting new slices between adjacent in-plane slices. Then, we propose a novel adjacent slices feature transformer (ASFT) network to utilize the similarity of adjacent slices. Specifically, the backbone of the ASFT network consists of a series of stacked multi-branch features transformation and extraction (MFTE) blocks. In each MFTE block, we construct new spatial attention to focus on features from specific areas in reference branches and use channel attention to enhance the most valuable information. In addition, we propose a hybrid loss function with content and gradient information to refine the pixels and structures of the reconstructed image. We also introduce global residual learning to reduce the difficulty of network training. Experimental results show that the ASFT network gains the PSNR of 43.68 dB, 40.96 dB, and 41.22 dB with the scale factor of × 2 on the public Kirby21, ANVIL-adult, and MSSEG datasets, respectively. When compared with state-of-the-art MRI SR methods, the ASFT network achieves superior quantitative and qualitative performance. [ABSTRACT FROM AUTHOR]

• C-BiLSTM with 1-D kernel is efficient for audio feature extraction. • Boundary regression contributes to more accurate cough detection. • Strong generalization ability is shown facing unknown non-cough events. • High performance of cough detection meets the end of practical clinical application. • The Corp Dataset provides powerful support for further research. Automatic cough detection in the patients' realistic audio recordings is of great significance to diagnose and monitor respiratory diseases, such as COVID-19. Many detection methods have been developed so far, but they are still unable to meet the practical requirements. In this paper, we present a deep convolutional bidirectional long short-term memory (C-BiLSTM) model with boundary regression for cough detection, where cough and non-cough parts need to be classified and located. We added convolutional layers before the LSTM to enhance the cough features and preserve the temporal information of the audio data. Considering the importance of the cough event integrity for subsequent analysis, the novel model includes an embedded boundary regression on the last feature map for both higher detection accuracy and more accurate boundaries. We delicately designed, collected and labelled a realistic audio dataset containing recordings of patients with respiratory diseases, named the Corp Dataset. 168 h of recordings with 9969 coughs from 42 different patients are included. The dataset is published online on the MARI Lab website (https://mari.tongji.edu.cn/info/1012/1030.htm). The results show that the system achieves a sensitivity of 84.13%, a specificity of 99.82% and an intersection-over-union (IoU) of 0.89, which is significantly superior to other related models. With the proposed method, all the criteria on cough detection significantly increased. The open source Corp Dataset provides useful material and a benchmark for researchers investigating cough detection. We propose the state-of-the-art system with boundary regression, laying the foundation for identifying cough sounds in real-world audio data. [ABSTRACT FROM AUTHOR]

COVID-19 infection segmentation has essential applications in determining the severity of a COVID-19 patient and can provide a necessary basis for doctors to adopt a treatment scheme. However, in clinical applications, infection segmentation is performed by human beings, which is time-consuming and generally introduces bias. In this paper, we developed a novel evolvable adversarial framework for COVID-19 infection segmentation. Three generator networks compose an evolutionary population to accommodate the current discriminator, i.e., generator networks evolved with different mutations instead of the single adversarial objective to provide sufficient gradient feedback. Compared with the existing work that enforces a Lipschitz constraint by weight clipping, which may lead to gradient exploding or vanishing, the proposed model also incorporates the gradient penalty into the network, penalizing the discriminator's gradient norm input. Experiments on several COVID-19 CT scan datasets verified that the proposed method achieved superior effectiveness and stability for COVID-19 infection segmentation. [ABSTRACT FROM AUTHOR]