Learning from the few: Fine-grained approach to pediatric wrist pathology recognition on a limited dataset.

Wrist pathologies, particularly fractures common among children and adolescents, present a critical diagnostic challenge. While X-ray imaging remains a prevalent diagnostic tool, the increasing misinterpretation rates highlight the need for more accurate analysis, especially considering the lack of specialized training among many surgeons and physicians. Recent advancements in deep convolutional neural networks offer promise in automating pathology detection in trauma X-rays. However, distinguishing subtle variations between pediatric wrist pathologies in X-rays remains challenging. Traditional manual annotation, though effective, is laborious, costly, and requires specialized expertise. In this paper, we address the challenge of pediatric wrist pathology recognition with a fine-grained approach, aimed at automatically identifying discriminative regions in X-rays without manual intervention. We refine our fine-grained architecture through ablation analysis and the integration of LION. Leveraging Grad-CAM, an explainable AI technique, we highlight these regions. Despite using limited data, reflective of real-world medical study constraints, our method consistently outperforms state-of-the-art image recognition models on both augmented and original (challenging) test sets. Our proposed refined architecture achieves an increase in accuracy of 1.06% and 1.25% compared to the baseline method, resulting in accuracies of 86% and 84%, respectively. Moreover, our approach demonstrates the highest fracture sensitivity of 97%, highlighting its potential to enhance wrist pathology recognition.

Modulating Nonlinear Acoustic Response of Phospholipid-Coated Microbubbles with pH for Ultrasound Imaging.

Activatable microbubble contrast agents for contrast-enhanced ultrasound have a potential role for measuring physiologic and pathologic states in deep tissues, including tumor acidosis. In this study, we describe a novel observation of increased harmonic oscillation of phosphatidylcholine microbubbles (PC-MBs) in response to lower ambient pH using a clinical ultrasound scanner. MB echogenicity and nonlinear echoes were monitored at neutral and acidic pH using B-mode and Cadence contrast pulse sequencing (CPS), a harmonic imaging technique at 7.0 and 1.5 MHz. A 3-fold increase in harmonic signal intensity was observed when the pH of PC-MB suspensions was decreased from 7.4 to 5.5 to mimic normal and pathophysiological levels that can be encountered in vivo. This pH-mediated activation is tunable based on the chemical structure and length of phospholipids composing the MB shell. It is also reliant on the presence of phosphate groups, as the use of lipids without phosphate instead of phospholipids completely abrogated this phenomenon. The increased harmonic signal likely is the result of increased MB oscillation caused by a decrease of the interfacial tension induced at a lower pH, altering the lipid conformation. While relative signal changes are interpreted clinically as mostly related to blood flow, pH effects could be significant contributors, particularly when imaging tumors. While our observation can be used clinically, it requires further research to isolate the effect of pH from other variables. These findings could pave the way toward for the development of new smart ultrasound contrast agents that expand the clinical utility of contrast-enhanced ultrasound.

The enlightening role of explainable artificial intelligence in medical & healthcare domains: A systematic literature review.

In domains such as medical and healthcare, the interpretability and explainability of machine learning and artificial intelligence systems are crucial for building trust in their results. Errors caused by these systems, such as incorrect diagnoses or treatments, can have severe and even life-threatening consequences for patients. To address this issue, Explainable Artificial Intelligence (XAI) has emerged as a popular area of research, focused on understanding the black-box nature of complex and hard-to-interpret machine learning models. While humans can increase the accuracy of these models through technical expertise, understanding how these models actually function during training can be difficult or even impossible. XAI algorithms such as Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) can provide explanations for these models, improving trust in their predictions by providing feature importance and increasing confidence in the systems. Many articles have been published that propose solutions to medical problems by using machine learning models alongside XAI algorithms to provide interpretability and explainability. In our study, we identified 454 articles published from 2018-2022 and analyzed 93 of them to explore the use of these techniques in the medical domain.

SentiUrdu-1M: A large-scale tweet dataset for Urdu text sentiment analysis using weakly supervised learning.

Low-resource languages are gaining much-needed attention with the advent of deep learning models and pre-trained word embedding. Though spoken by more than 230 million people worldwide, Urdu is one such low-resource language that has recently gained popularity online and is attracting a lot of attention and support from the research community. One challenge faced by such resource-constrained languages is the scarcity of publicly available large-scale datasets for conducting any meaningful study. In this paper, we address this challenge by collecting the first-ever large-scale Urdu Tweet Dataset for sentiment analysis and emotion recognition. The dataset consists of a staggering number of 1, 140, 821 tweets in the Urdu language. Obviously, manual labeling of such a large number of tweets would have been tedious, error-prone, and humanly impossible; therefore, the paper also proposes a weakly supervised approach to label tweets automatically. Emoticons used within the tweets, in addition to SentiWordNet, are utilized to propose a weakly supervised labeling approach to categorize extracted tweets into positive, negative, and neutral categories. Baseline deep learning models are implemented to compute the accuracy of three labeling approaches, i.e., VADER, TextBlob, and our proposed weakly supervised approach. Unlike the weakly supervised labeling approach, the VADER and TextBlob put most tweets as neutral and show a high correlation between the two. This is largely attributed to the fact that these models do not consider emoticons for assigning polarity.

Light-DehazeNet: A Novel Lightweight CNN Architecture for Single Image Dehazing.

Due to the rapid development of artificial intelligence technology, industrial sectors are revolutionizing in automation, reliability, and robustness, thereby significantly increasing quality and productivity. Most of the surveillance and industrial sectors are monitored by visual sensor networks capturing different surrounding environment images. However, during tempestuous weather conditions, the visual quality of the images is reduced due to contaminated suspended atmospheric particles that affect the overall surveillance systems. To tackle these challenges, this article presents a computationally efficient lightweight convolutional neural network referred to as Light-DehazeNet (LD-Net) for the reconstruction of hazy images. Unlike other learning-based approaches, which separately measure the transmission map and the atmospheric light, our proposed LD-Net jointly estimates both the transmission map and the atmospheric light using a transformed atmospheric scattering model. Furthermore, a color visibility restoration method is proposed to evade the color distortion in the dehaze image. Finally, we conduct extensive experiments using synthetic and natural hazy images. The quantitative and qualitative evaluation on different benchmark hazy datasets verify the superiority of the proposed method over other state-of-the-art image dehazing techniques. Moreover, additional experimentation validates the applicability of the proposed method in the object detection tasks. Considering the lightweight architecture with minimal computational cost, the proposed system is encouraged to be incorporated as an integral part of the vision-based monitoring systems to improve the overall performance.

Towards Smart Home Automation Using IoT-Enabled Edge-Computing Paradigm.

Smart home applications are ubiquitous and have gained popularity due to the overwhelming use of Internet of Things (IoT)-based technology. The revolution in technologies has made homes more convenient, efficient, and even more secure. The need for advancement in smart home technology is necessary due to the scarcity of intelligent home applications that cater to several aspects of the home simultaneously, i.e., automation, security, safety, and reducing energy consumption using less bandwidth, computation, and cost. Our research work provides a solution to these problems by deploying a smart home automation system with the applications mentioned above over a resource-constrained Raspberry Pi (RPI) device. The RPI is used as a central controlling unit, which provides a cost-effective platform for interconnecting a variety of devices and various sensors in a home via the Internet. We propose a cost-effective integrated system for smart home based on IoT and Edge-Computing paradigm. The proposed system provides remote and automatic control to home appliances, ensuring security and safety. Additionally, the proposed solution uses the edge-computing paradigm to store sensitive data in a local cloud to preserve the customer's privacy. Moreover, visual and scalar sensor-generated data are processed and held over edge device (RPI) to reduce bandwidth, computation, and storage cost. In the comparison with state-of-the-art solutions, the proposed system is 5% faster in detecting motion, and 5 ms and 4 ms in switching relay on and off, respectively. It is also 6% more efficient than the existing solutions with respect to energy consumption.

An Adaptive Game-Based Learning Strategy for Children Road Safety Education and Practice in Virtual Space.

Virtual reality (VR) has been widely used as a tool to assist people by letting them learn and simulate situations that are too dangerous and risky to practice in real life, and one of these is road safety training for children. Traditional video- and presentation-based road safety training has average output results as it lacks physical practice and the involvement of children during training, without any practical testing examination to check the learned abilities of a child before their exposure to real-world environments. Therefore, in this paper, we propose a 3D realistic open-ended VR and Kinect sensor-based training setup using the Unity game engine, wherein children are educated and involved in road safety exercises. The proposed system applies the concepts of VR in a game-like setting to let the children learn about traffic rules and practice them in their homes without any risk of being exposed to the outside environment. Thus, with our interactive and immersive training environment, we aim to minimize road accidents involving children and contribute to the generic domain of healthcare. Furthermore, the proposed framework evaluates the overall performance of the students in a virtual environment (VE) to develop their road-awareness skills. To ensure safety, the proposed system has an extra examination layer for children's abilities evaluation, whereby a child is considered fit for real-world practice in cases where they fulfil certain criteria by achieving set scores. To show the robustness and stability of the proposed system, we conduct four types of subjective activities by involving a group of ten students with average grades in their classes. The experimental results show the positive effect of the proposed system in improving the road crossing behavior of the children.

Formulation effects on paclitaxel transfer and uptake in the human placenta.

Diagnosis and treatment of breast cancer in pregnancy can result in morbidity and mortality for the mother and fetus. Many new paclitaxel nanoformulations commercially available worldwide for breast cancer treatment are being adopted due to favorable dosing regimens and side effect profiles, but their transplacental transport and resultant fetal exposure remain unknown. Here, we examine three formulations: Taxol (paclitaxel dissolved in Kolliphor EL and ethanol); Abraxane (albumin nanoparticle); and Genexol-PM (polymeric micelle). In the ex vivo dually perfused human placental cotyledon, placental accumulation of Genexol-PM is higher than Taxol, and both nanoformulations have lower maternal concentrations of paclitaxel over time. In vitro studies of these formulations and fluorescent nanoparticle analogs demonstrate that Genexol-PM allows paclitaxel to overcome P-glycoprotein efflux, but Abraxane behaves as a free drug formulation. We anticipate that these findings will impact future development of rational and safe treatment strategies for pregnancy-associated breast cancer and other diseases.

WET: Word embedding-topic distribution vectors for MOOC video lectures dataset.

In this article, we present a dataset containing word embeddings and document topic distribution vectors generated from MOOCs video lecture transcripts. Transcripts of 12,032 video lectures from 200 courses were collected from Coursera learning platform. This large corpus of transcripts was used as input to two well-known NLP techniques, namely Word2Vec and Latent Dirichlet Allocation (LDA) to generate word embeddings and topic vectors, respectively. We used Word2Vec and LDA implementation in the Gensim package in Python. The data presented in this article are related to the research article entitled "Integrating word embeddings and document topics with deep learning in a video classification framework" [1]. The dataset is hosted in the Mendeley Data repository [2].

Cross-Cultural Polarity and Emotion Detection Using Sentiment Analysis and Deep Learning on COVID-19 Related Tweets.

How different cultures react and respond given a crisis is predominant in a society's norms and political will to combat the situation. Often, the decisions made are necessitated by events, social pressure, or the need of the hour, which may not represent the nation's will. While some are pleased with it, others might show resentment. Coronavirus (COVID-19) brought a mix of similar emotions from the nations towards the decisions taken by their respective governments. Social media was bombarded with posts containing both positive and negative sentiments on the COVID-19, pandemic, lockdown, and hashtags past couple of months. Despite geographically close, many neighboring countries reacted differently to one another. For instance, Denmark and Sweden, which share many similarities, stood poles apart on the decision taken by their respective governments. Yet, their nation's support was mostly unanimous, unlike the South Asian neighboring countries where people showed a lot of anxiety and resentment. The purpose of this study is to analyze reaction of citizens from different cultures to the novel Coronavirus and people's sentiment about subsequent actions taken by different countries. Deep long short-term memory (LSTM) models used for estimating the sentiment polarity and emotions from extracted tweets have been trained to achieve state-of-the-art accuracy on the sentiment140 dataset. The use of emoticons showed a unique and novel way of validating the supervised deep learning models on tweets extracted from Twitter.

HPLC Method Development for Quantification of Doxorubicin in Cell Culture and Placental Perfusion Media.

Assessment of drug transport across the placenta is important in understanding the effect of drugs on placental and fetal health. These phenomena can be studied in both in vitro cell lines and ex vivo placental perfusions. We have successfully developed a sensitive yet simple high performance liquid chromatography (HPLC) method coupled with fluorescence detection to determine the concentration of doxorubicin (DXR) in cell culture media for transport studies in human trophoblast cells (BeWo, b30 clone) and in fetal media for placental perfusion experiments. The method was developed based on a protein precipitation technique and was validated in both media types for linearity, intra-day, and inter-day precision and accuracy. The relationship of peak area to concentration was linear with R2 values of 0.99 or greater obtained over the concentration range of 1.5 to 15,000 ng/mL. Despite the high concentrations of albumin in fetal perfusion media (30 mg/mL), the lower limits of detection and quantification for DXR were found to be 1.5 and 5 ng/mL, respectively. This analytical method may be used to study the transport of DXR across BeWo cells and human placenta during placental perfusion studies.

Determination of the Transplacental Transfer of Paclitaxel and Antipyrine by High Performance Liquid Chromatography Coupled with Photodiode Array Detector.

Ex vivo placental perfusion experiments are important in understanding the quantity and mechanisms of xenobiotic transport to the fetus during pregnancy. Our study demonstrates that paclitaxel and antipyrine concentrations in placental perfusion medium containing physiological concentrations of human serum albumin during pregnancy (30 mg/mL) can be quantified by RP-HPLC and UV detection. A liquid-liquid extraction method was developed for the quantification of paclitaxel and celecoxib (internal standard) from perfusion medium. Antipyrine, which is a necessary marker in placental perfusions for determining the validity of experiments and calculating the clearance index of xenobiotics, was also analyzed by HPLC and UV detection. Antipyrine concentrations were determined by HPLC after precipitating the perfusion medium in acetonitrile and separating the precipitated proteins by centrifugation. Concentrations were fitted to linear regressions with R2 values approaching 1. Lower limits of detection for paclitaxel and antipyrine were 100 ng/mL and 200 ng/mL, respectively. Both methods demonstrated high intra-day and inter-day precision and trueness. Additionally, the use of these methods was demonstrated in a placental perfusion experiment using Taxol® (paclitaxel dissolved in Cremophor-EL). The fetal transfer rate of Taxol was 6.6% after 1 hour.

Cytotoxicity of Endocytosis and Efflux Inhibitors in the BeWo Cell Line.

AIMS: The purpose of this study was to determine the cell viability and cytotoxicity of various endocytosis and efflux inhibitors which can be used to determine transport and uptake mechanisms in the BeWo (b30 clone) human placental trophoblast cell line. Ethanol and dimethylsulfoxide (DMSO) were also studied since they are often used as cosolvents for administration of these inhibitors. METHODOLOGY: The water-soluble tetrazolium-1 (WST-1) assay was used to quantify cell viability and the lactate dehydrogenase (LDH) assay was used to determine cytotoxicity. RESULTS: By the WST-1 assay, reduced cell viability was observed following 4 hours of exposure to chlorpromazine (10 µg/mL), colchicine (1 mM), filipin (3 µg/mL), gentamicin (2 mM), GF120918 (1 µM), methyl-ß-cyclodextrin (5 mM), and verapamil (100 µM). By the LDH assay, however, no cytotoxicity was observed after 4 hours of exposure to the aforementioned compounds. Amiloride (500 µM), ethanol (up to 0.1% v/v), and DMSO (up to 0.1% v/v) did not reduce cell viability nor induce cytotoxicity. CONCLUSION: This information is valuable when selecting potential inhibitors of endocytosis and efflux and the selection of time points for mechanistic studies.

Placental control of drug delivery.

The placenta serves as the interface between the maternal and fetal circulations and regulates the transfer of oxygen, nutrients, and waste products. When exogenous substances are present in the maternal bloodstream-whether from environmental contact, occupational exposure, medication, or drug abuse-the extent to which this exposure affects the fetus is determined by transport and biotransformation processes in the placental barrier. Advances in drug delivery strategies are expected to improve the treatment of maternal and fetal diseases encountered during pregnancy.

Transport of digoxin-loaded polymeric nanoparticles across BeWo cells, an in vitro model of human placental trophoblast.

BACKGROUND: Fetal arrhythmias can lead to fetal congestive heart failure and hydrops fetalis. Digoxin (the first-line treatment) has low transplacental permeability and high risk of maternal side effects. Biodegradable digoxin-loaded PEGylated poly(lactic-co-glycolic acid) nanoparticles may increase digoxin transport across BeWo b30 cell monolayers (an in vitro model of trophoblast in human placenta) by reducing the drug's interaction with P-gp. Results/methodology: The nanoparticles showed high encapsulation efficiency and sustained release over 48 h. Transport studies revealed significantly increased permeability across BeWo cell layers of digoxin-loaded nanoparticles when compared with free digoxin. P-gp inhibition also increased the permeability of digoxin, but not digoxin-loaded nanoparticles. CONCLUSION: This represents a novel treatment strategy for fetal cardiovascular disease which may improve maternal and fetal outcomes.

Significant gaps in awareness of familial hypercholesterolemia among physicians in selected Asia-Pacific countries: a pilot study.

BACKGROUND: Familial hypercholesterolemia (FH) is a dominantly inherited disorder characterized by high plasma cholesterol levels and a very high risk of early heart disease. The prevalence of FH is estimated to be at least 1:500, with at least 3.6 million individuals in the Asia-Pacific region. OBJECTIVE: To assess awareness, knowledge, and perception of FH among practicing physicians in Japan, South Korea, and Taiwan. METHODS: Physicians from 3 economically developed Asian countries were requested to anonymously complete a structured Internet-based survey regarding FH. This survey sought responses on the clinical description, inheritance, prevalence, cardiovascular disease risk, practices, and opinions on screening. RESULTS: Of 230 physicians surveyed, 47% were aware of the heritability, 27% of the prevalence, and 13% of the risk of cardiovascular disease relating to FH. The majority (70%) perceived themselves to have an above-moderate familiarity with FH. Primary care physicians (59%) and lipid specialists (41%) were perceived as the best providers for caring for FH, including cascade screening services, with a lesser role perceived for cardiologists, endocrinologists, and no significant role for nursing staff. Only 35% of physicians were aware of specialist clinical services for lipid disorders in their geographic area. CONCLUSION: Extensive education and training programs are required to complement the implementation of region-specific models of care for FH in Asia. Further enhancement of existing lipid services and facilities are also warranted to optimise service models.

Influences of nanomaterials on the barrier function of epithelial cells.

Recent advances in nanotechnology have led to exciting opportunities in medicine, energy, manufacturing, and other fields. Nevertheless, it is important to adequately assess the potential impacts of nanomaterial exposure. This chapter focuses on the interactions of nanomaterials with epithelial barriers in the lungs, intestine, kidneys, skin, and placenta. Methods for determining transepithelial electrical resistance and paracellular permeability are described. Effects on cell viability and barrier integrity depend on the chemical nature of the nanomaterial, nanoparticle size, surface coatings, and concentration. Disruption of tight junctions can affect permeability and interfere with normal regulatory processes of the epithelial barrier. Future research is needed to better understand the possibilities and the limits of novel approaches in nanotechnology.

Mipomersen, an apolipoprotein B synthesis inhibitor, reduces atherogenic lipoproteins in patients with severe hypercholesterolemia at high cardiovascular risk: a randomized, double-blind, placebo-controlled trial.

OBJECTIVES: This study sought to examine the efficacy and safety of mipomersen for reducing atherogenic lipids and lipoproteins in patients with hypercholesterolemia. BACKGROUND: Many patients on lipid-lowering therapies remain unable to achieve target low-density lipoprotein (LDL) cholesterol levels. Mipomersen, an antisense oligonucleotide inhibitor of apolipoprotein B, reduces LDL cholesterol and atherogenic lipoproteins. METHODS: This randomized, double-blind, multicenter study enrolled 158 patients with baseline LDL cholesterol levels ≥100 mg/dl with, or at high risk for, coronary heart disease who were receiving maximally tolerated lipid-lowering therapy. Patients received weekly subcutaneous mipomersen 200 mg (n = 105) or placebo (n = 52) for 26 weeks, with a 24-week follow-up period. Randomization was stratified by type 2 diabetes status. RESULTS: Sixty mipomersen and 44 placebo patients completed treatment. Mean baseline LDL cholesterol levels were 122.7 and 122.6 mg/dl in the placebo and mipomersen patients, respectively. Mipomersen reduced LDL cholesterol by -36.9% compared with placebo at -4.5% (p < 0.001). Target LDL cholesterol <100 mg/dl was attained in 76% of mipomersen and 38% of placebo patients. Mipomersen also significantly reduced apolipoprotein B (-38%) and lipoprotein(a) (-24%) (p < 0.001). Common adverse events included injection site reactions (78% with mipomersen, 31% with placebo) and flu-like symptoms (34% with mipomersen, 21% with placebo). Elevations in transaminases and liver fat also occurred in some patients, and these levels returned toward baseline after treatment cessation. CONCLUSIONS: Mipomersen significantly reduced LDL cholesterol, apolipoprotein B, and lipoprotein(a) in patients with hypercholesterolemia with, or at risk for, coronary heart disease not controlled by existing therapies. (Safety and Efficacy of Mipomersen [ISIS 301012] as Add-On Therapy in High Risk Hypercholesterolemic Patients; NCT00770146).

Training of medical and paramedical personnel in burn care and prevention.

This paper discusses the requirements for training in burn care within a resource limited environment, what is currently practiced and goes on to suggest a strategy for effective delivery of education and training.

A novel, cost effective escharotomy simulator and trainee assessment.

Over the last two decades, virtual reality, haptics, simulators, robotics, and other "advanced technologies" have emerged as important innovations in medical learning and practice. In the 21st century, however, it is important to continue to develop simple teaching aids which are available to large audiences in low and middle-income countries. We present a simple 'escharotomy simulator' which has been well received, resulting in an increase in knowledge, and an increase in confidence to carry out the procedure.