Resource-constrained Environments Research Articles

Navigating the Challenges in Setting Up a Sustainable Open-Heart Surgery Unit in a Resource-Constrained Environment in Northern Nigeria: Model and Strategies.

Cardiac surgery requiring cardiopulmonary bypass had been unavailable in Northern Nigeria and the federal capital territory of Nigeria regularly. Several attempts in the past at setting up this service in a self-sustaining manner in Northern Nigeria had failed. This paper is a contrasting response to an earlier publication that emphasized the less-than-desirable role played by international cardiac surgery missions in the evolution of a sustainable open-heart surgery program in Nigeria. The cardiothoracic unit of Federal Medical Centre, Abuja, was established on March 1, 2021, but could not conduct safe open-heart surgery. The model and strategies employed in commencing open-heart surgeries, including the choice of personnel training within the country and focused collaboration with foreign missions, are discussed. We also report the first seven patients to undergo cardiac surgery under cardiopulmonary bypass in our government-run hospital as well as the transition from foreign missions to local team operations. Seven patients were operated on within the first six months of setting up with high levels of skill transfer and local team participation, culminating in one of the operations entirely carried out by the local team of personnel. All outcomes were good at an average of one-year follow-up. In resource-constrained government-run hospitals, a functional, safe cardiac surgery unit can be set up by implementing well-planned strategies to mitigate encountered peculiar challenges. Furthermore, with properly harnessed foreign missions, a prior-trained local team of personnel can achieve independence and become a self-sustaining cardiac surgery unit within the shortest possible time.

To Exit or Not to Exit: Cost-Effective Early-Exit Architecture Based on Markov Decision Process

Recently, studies on early-exit mechanisms have emerged to reduce the computational cost during the inference process of deep learning models. However, most existing early-exit architectures simply determine early exiting based only on a target confidence level in the prediction, without any consideration of the computational cost. Such an early-exit criterion fails to balance accuracy and cost, making it difficult to use in various environments. To address this problem, we propose a novel, cost-effective early-exit architecture in which an early-exit criterion is designed based on the Markov decision process (MDP). Since the early-exit decisions within an early-exit model are sequential, we model them as an MDP problem to maximize accuracy as much as possible while minimizing the computational cost. Then, we develop a cost-effective early-exit algorithm using reinforcement learning that solves the MDP problem. For each input sample, the algorithm dynamically makes early-exit decisions considering the relative importance of accuracy and computational cost in a given environment, thereby balancing the trade-off between accuracy and cost regardless of the environment. Consequently, it can be used in various environments, even in a resource-constrained environment. Through extensive experiments, we demonstrate that our proposed architecture can effectively balance the trade-off in different environments, while the existing architectures fail to do so since they focus only on reducing their cost while preventing the degradation of accuracy.

Malaria vaccines: a new era of prevention and control.

Malaria killed over 600,000 people in 2022, a death toll that has not improved since 2015. Additionally, parasites and mosquitoes resistant to existing interventions are spreading across Africa and other regions. Vaccines offer hope to reduce the mortality burden: the first licensed malaria vaccines, RTS,S and R21, will be widely deployed in 2024 and should substantially reduce childhood deaths. In this Review, we provide an overview of the malaria problem and the Plasmodium parasite, then describethe RTS,S and R21 vaccines (the first vaccines for any human parasitic disease), summarizing their benefits and limitations. We explore next-generation vaccines designed using new knowledge of malaria pathogenesis and protective immunity, which incorporate antigens and platforms to elicit effective immune responses against different parasite stages in human or mosquito hosts. We describe a decision-making process that prioritizes malaria vaccine candidates for development in a resource-constrained environment. Future vaccines might improve upon the protective efficacy of RTS,S or R21 for children, or address the wider malaria scourge by preventing pregnancy malaria, reducing the burden of Plasmodium vivax or accelerating malaria elimination.

AD plasma biomarkers are stable for an extended period at -20°C: implications for resource-constrained environments.

Standard procedures for measuring Alzheimer's disease (AD) plasma biomarkers include storage at -80°C. This is challenging in countries lacking research infrastructure, such -80°C freezer. To investigate stability of AD biomarkers from plasma stored at -20°C, we compared aliquots stored at -80°C and others at -20°C for two, four, six, fifteen, and thirty-five weeks. pTau181, Aβ42, Aβ40, NfL, and GFAP were measured for each timepoint. pTau181 and Aβ42/Aβ40 ratios showed minimal variation for up to 15 weeks. NfL and GFAP had higher variability. This finding of 15-week stability at -20°C enables greater participation in AD biomarker studies in resource constrained environments.

“A life-changing experience and the beginning of a lifelong commitment”: experiences and impact of Global Health Community Service-Learning in undergraduate dental curriculum in Canada

BackgroundGlobal Health Community Service-Learning (GHCSL) can have a profound professional and personal impact on learners. This pedagogy provides understanding of unfamiliar environments and challenges learners to step out of their comfort zones, adapt to new cultures, and navigate unique situations. Yet, there are relatively few studies exploring the experiences of learners participating in community service-learning placements in global regions as part of undergraduate dental curriculum. This study aimed to explore the experiences and impact of the GHCSL program in East Africa among undergraduate dental learners at the Schulich School of Dentistry.MethodsEight undergraduate dental learners were enrolled in GHCSL pilot placements. Placement agreements were established with Makerere University in Kampala, Uganda, and the University of Rwanda in Kigali, Rwanda for the summer of 2022. Stakeholders from both institutions were engaged in the development and implementation of these placements. Learners were required to engage in weekly reflection through a ‘storytelling and incident-based narrative’ while carrying out their placement. A qualitative study design was employed, and an inductive interpretive approach was utilized to thematically analyze the learners’ reflective essays.ResultsFive major themes emerged from the learners’ reflective essays: (1) experiential clinical learning; (2) cultural humility and social awareness; (3) awareness of contrasting healthcare systems; (4) commitment to service; and (5) personal and professional growth. Most learners reflected on their engagement with diverse communities, being exposed to unique patient cases, and witnessing the adaptability exuded in resource-constrained environments. These experiences presented the learners with an opportunity to develop cultural humility and gain a newfound motivation to mitigate global oral health disparities in populations beyond that of their local communities. Learners also reflected on enhanced social awareness experiences and the awareness of contrasting healthcare systems in Canada and their placements, which encouraged the development of empathy, communication, and compassion skills, as well as an understanding of the disproportionate burden of conditions in low-resource settings.ConclusionThe reflective essays concluded that the GHCSL placements had a positive impact on the learners, encouraging many to develop a heightened desire for lifelong learning to address oral health disparities within a global context.

Research on Student Classroom Behavior Detection Based on the Real-Time Detection Transformer Algorithm

With the rapid development of artificial intelligence and big data technology, intelligent education systems have become a key research focus in the field of modern educational technology. This study aims to enhance the intelligence level of educational systems by accurately detecting student behavior in the classroom using deep learning techniques. We propose a method for detecting student classroom behavior based on an improved RT DETR (Real-Time Detection Transformer) object detection algorithm. By combining actual classroom observation data with AI-generated data, we create a comprehensive and diverse student behavior dataset (FSCB-dataset). This dataset not only more realistically simulates the classroom environment but also effectively addresses the scarcity of datasets and reduces the cost of dataset construction. The study introduces MobileNetV3 as a lightweight backbone network, reducing the model parameters to one-tenth of the original while maintaining nearly the same accuracy. Additionally, by incorporating learnable position encoding and dynamic upsampling techniques, the model significantly improves its ability to recognize small objects and complex scenes. Test results on the FSCB-dataset show that the improved model achieves significant improvements in real-time performance and computational efficiency. The lightweight network is also easy to deploy on mobile devices, demonstrating its practicality in resource-constrained environments.

Enhancing YOLOv5 for superior PCB defect detection: lightweight heads and small target precision

Abstract This paper introduces an improved YOLOv5 network designed to achieve enhanced performance in the field of printed circuit board (PCB) defect detection. We incorporate a lightweight detection head, reducing the number of network parameters and improving computational efficiency. This not only facilitates operation in resource-constrained environments but also maintains detection accuracy. To better address the specific requirements of PCB defect detection, we introduce a small target detection head. By adding a detection head specifically designed for small targets, we enhance the network’s ability to perceive subtle defects, thereby improving detection accuracy. Finally, we validate our proposed improvements through experiments on a large-scale PCB defect dataset. Compared to the traditional YOLOv5 network, our model exhibits significant improvements in both accuracy and efficiency.

Research on the Training and Application Methods of a Lightweight Agricultural Domain-Specific Large Language Model Supporting Mandarin Chinese and Uyghur

In the field of Natural Language Processing (NLP), the lack of support for minority languages, especially Uyghur, the scarcity of Uyghur language corpora in the agricultural domain, and the lightweight nature of large language models remain prominent issues. This study proposes a method for constructing a bilingual (Uyghur and Chinese) lightweight specialized large language model for the agricultural domain. By utilizing a mixed training approach of Uyghur and Chinese, we extracted Chinese corpus text from agricultural-themed books in PDF format using OCR (Optical Character Recognition) technology, converted the Chinese text corpus into a Uyghur corpus using a rapid translation API, and constructed a bilingual mixed vocabulary. We applied the parameterized Transformer model algorithm to train the model for the agricultural domain in both Chinese and Uyghur. Furthermore, we introduced a context detection and fail-safe mechanism for the generated text. The constructed model possesses the ability to support bilingual reasoning in Uyghur and Chinese in the agricultural domain, with higher accuracy and a smaller size that requires less hardware. It (our work) addresses issues such as the scarcity of Uyghur corpora in the agricultural domain, mixed word segmentation and word vector modeling in Uyghur for widespread agricultural languages, model lightweighting and deployment, and the fragmentation of non-relevant texts during knowledge extraction from small-scale corpora. The lightweight design of the model reduces hardware requirements, facilitating deployment in resource-constrained environments. This advancement promotes agricultural intelligence, aids in the development of specific applications and minority languages (such as agriculture and Uyghur), and contributes to rural revitalization.

Wheat Leaf Disease Detection: A Lightweight Approach with Shallow CNN Based Feature Refinement

Improving agricultural productivity is essential due to rapid population growth, making early detection of crop diseases crucial. Although deep learning shows promise in smart agriculture, practical applications for identifying wheat diseases in complex backgrounds are limited. In this paper, we propose CropNet, a hybrid method that utilizes Red, Green, and Blue (RGB) imaging and a transfer learning approach combined with shallow convolutional neural networks (CNN) for further feature refinement. To develop our customized model, we conducted an extensive search for the optimal deep learning architecture. Our approach involves freezing the pre-trained model for feature extraction and adding a custom trainable CNN layer. Unlike traditional transfer learning, which typically uses trainable dense layers, our method integrates a trainable CNN, deepening the architecture. We argue that pre-trained features in transfer learning are better suited for a custom shallow CNN followed by a fully connected layer, rather than being fed directly into fully connected layers. We tested various architectures for pre-trained models including EfficientNetB0 and B2, DenseNet, ResNet50, MobileNetV2, MobileNetV3-Small, and Inceptionv3. Our approach combines the strengths of pre-trained models with the flexibility of custom architecture design, offering efficiency, effective feature extraction, customization options, reduced overfitting, and differential learning rates. It distinguishes itself from classical transfer learning techniques, which typically fine-tune the entire pre-trained network. Our aim is to provide a lightweight model suitable for resource-constrained environments, capable of delivering outstanding results. CropNet achieved 99.80% accuracy in wheat disease detection with reduced training time and computational cost. This efficient performance makes CropNet promising for practical implementation in resource-constrained agricultural settings, benefiting farmers and enhancing production.

Innovative sanitation solutions: evaluating the performance of faecal sludge treatment plant in humanitarian settings: A case study from Rohingya camps in Bangladesh

The Rohingya crisis has precipitated one of the most significant humanitarian emergencies, resulting in over 900,000 refugees residing in Cox's Bazar, Bangladesh. The rapid influx of refugees has posed substantial challenges in providing essential services, particularly sanitation, in a densely populated and resource-constrained environment. This study examines the performance of faecal sludge treatment plants (FSTPs) implemented in the Rohingya camps, focusing on key performance indicators such as treatment efficiency, operational challenges, health outcomes, and resource recovery potential. Analysis reveals that FSTPs have played a crucial role in reducing pathogen levels in treated sludge, thereby mitigating the risk of waterborne diseases and environmental contamination. Despite these successes, operational challenges such as frequent desludging, infrastructure maintenance, and the logistical complexities of waste collection and transport persist. The study highlights the importance of addressing these challenges through regular maintenance, infrastructure upgrades, and innovative logistical solutions. Health outcomes have shown improvement with a reduction in sanitation-related diseases, underscoring the health benefits of effective FSTP implementation. Additionally, resource recovery from treated sludge presents a sustainable approach, with potential applications in agriculture and energy production. Community engagement has emerged as a critical factor in the success of these interventions. Effective community involvement in the planning, implementation, and maintenance of sanitation facilities has been instrumental in ensuring their acceptance and proper use. The study emphasizes the need for continued community education and participation to sustain these benefits. In conclusion, while FSTPs have significantly improved sanitation conditions in the Rohingya camps, ongoing efforts are required to enhance operational efficiency, address logistical challenges, and foster community involvement. The findings provide valuable insights for the design and implementation of innovative sanitation solutions in similar humanitarian settings worldwide.

Robust Forest Fire Detection using Deep Convolutional Neural Networks

Forest fires pose significant threats to ecosystems, wildlife, and human lives, necessitating proactive measures for early detection and rapid response. This paper presents FireGuard, an efficient model for forest fire detection using deep convolutional neural networks (CNNs). Leveraging the power of deep learning and image processing techniques, FireGuard analyzes aerial imagery and satellite data to detect signs of smoke and fire outbreaks in forested areas. The model utilizes a lightweight CNN architecture optimized for real-time performance and resource-constrained environments, making it suitable for deployment on unmanned aerial vehicles (UAVs), surveillance cameras, and satellite platforms. Experimental results demonstrate the effectiveness of FireGuard in accurately identifying forest fires with high precision and recall, outperforming traditional methods and existing deep learning models. By providing early warnings of potential fire incidents, FireGuard enables timely intervention by fire fighting agencies, thereby mitigating the impact of forest fires and preserving natural habitats.

Beyond borders: an appraisal of the global congenital cardiac surgery workforce.

Congenital heart disease (CHD) is the most common congenital abnormality worldwide. Patients in resource-constrained environments experience higher levels of mortality and morbidity from CHD. Many studies have quantified the incidence of CHD in resource-constrained environments and compared these to availability of surgeons per population. However, no study to date has compiled and created a repository of the global paediatric cardiac surgical workforce. We aimed to quantify the number and details of the congenital cardiac surgeons globally, compare the population of under 15's in each respective countries, and address the workforce gaps that currently exist within paediatric cardiac surgery. Cardiothoracic Surgery Network (CTSNet) was searched in 2021 to extrapolate the current paediatric cardiac surgical workforce. The data was evaluated with the current literature to address current gaps in workforce planning and sustainability to produce this appraisal. A total of 4,027 congenital cardiac surgeons were identified from CTSNet with 75% residing in high-income countries (HICs) or upper-middle-income countries (UMICs) despite these income groups only accounting for 16% of the world population. Despite similar incidence and prevalence of CHD globally, we found an unequal distribution in the availability of congenital cardiac surgeons worldwide. The disparity in the availability of surgeons between HICs and low-middle-income countries (LMICs) is a global health issue that will require serous thought and planning to resolve. The high proportion of preventable deaths from CHD cases is a regrettable figure that governments and medical organisations should further strive to decrease. Education and proactive investments in training up local teams in LMICs will allow for sustainability in global congenital cardiac surgery.

Healthcare professionals' willingness to utilize a mobile health application for adverse drug reaction reporting in a limited resource setting: An input for digital health, 2023

BackgroundAdverse drug reactions (ADRs) are a significant public health concern, particularly in limited resource settings where underreporting is prevalent due to various challenges. Mobile health applications (mHealth apps) offer a promising solution to enhance pharmacovigilance by facilitating easier and more efficient ADR reporting. However, despite the increasing availability and use of mHealth apps, there is a lack of evidence on healthcare professionals' willingness to adopt them for ADR reporting in resource-constrained environments. Therefore, this study aimed to assess the willingness of healthcare professionals in Ethiopia to utilize mobile health applications for adverse drug reaction reporting and identify associated factors. MethodsWe carried out a cross-sectional study involving 422 healthcare professionals working in institutional settings. We gathered data through a pretested questionnaire that participants completed themselves. We inputted the data using Epi Data V.4.6 and analyzed it using SPSS V.26. Our analysis involved conducting multivariable logistic regression to identify the factors influencing the likelihood of healthcare professionals using mobile applications to report adverse drug reactions. ResultsThe study involved 389 healthcare professionals. Approximately 301 (77.4 %) of them expressed willingness to utilize mobile applications for reporting adverse drug reactions. The willingness to utilize mobile applications was significantly associated with the type of mobile phone (smart: AOR 3.56; 95 % CI 2.15–5.67), basic computer training (AOR 4.43; 95 % CI 2.27–8.64), mobile health-related training (AOR 1.96; 95 % CI 1.01–3.79), attitude (AOR 4.01; 95 % CI 2.19–7.35), perceived ease of use (AOR 2.91; 95 % CI 1.59–5.23), and perceived usefulness (AOR 2.10; 95 % CI 1.15–3.85). ConclusionsOverall, there was a high proportion of healthcare professionals willing to use mobile devices for reporting drug adverse reactions. Their willingness correlated with factors such as the type of mobile phone, perceived ease of use, attitude, training, and perceived usefulness of mobile applications. With the increasing use of smartphones, motivation among healthcare professionals is rising. Basic computer and mHealth-related training are crucial for enhancing the acceptability of such applications and should be incorporated into future implementations. Taking these factors into account could offer insights into the design and implementation of mobile applications for adverse drug reactions in Ethiopia.

Anonymous data sharing scheme for resource-constrained internet of things environments

With the rapid development of Internet of Things (IoT) technology in industrial, agricultural, medical and other fields, IoT terminal devices face security and privacy challenges when sharing data. Among them, ensuring data confidentiality, achieving dual-side privacy protection, and performing reliable data integrity verification are basic requirements. Especially in resource-constrained environments, limitations in the storage, computing, and communication capabilities of devices increase the difficulty of implementing these security safeguards. To address this problem, this paper proposes a resource-constrained anonymous data-sharing scheme (ADS-RC) for the IoT. In ADS-RC, we use elliptic curve operations to replace computation-intensive bilinear pairing operations, thereby reducing the computational and communication burden on end devices. We combine an anonymous verifiable algorithm and an attribute encryption algorithm to ensure double anonymity and data confidentiality during the data-sharing process. To deal with potential dishonest behavior, this solution supports the revocation of malicious user permissions. In addition, we designed a batch data integrity verification algorithm and stored verification evidence on the blockchain to ensure the security and traceability of data during transmission and storage. Through experimental verification, the ADS-RC scheme achieves reasonable efficiency in correctness, security and efficiency, providing a new solution for data sharing in resource-constrained IoT environments.

Efficient and Secure Data Aggregation for Resource-Constrained IoT Environments

Efficient and Secure Data Aggregation for Resource-Constrained IoT Environments

Lightweight Deep Learning for Resource-Constrained Environments: A Survey

Over the past decade, the dominance of deep learning has prevailed across various domains of artificial intelligence, including natural language processing, computer vision, and biomedical signal processing. While there have been remarkable improvements in model accuracy, deploying these models on lightweight devices, such as mobile phones and microcontrollers, is constrained by limited resources. In this survey, we provide comprehensive design guidance tailored for these devices, detailing the meticulous design of lightweight models, compression methods, and hardware acceleration strategies. The principal goal of this work is to explore methods and concepts for getting around hardware constraints without compromising the model’s accuracy. Additionally, we explore two notable paths for lightweight deep learning in the future: deployment techniques for TinyML and Large Language Models. Although these paths undoubtedly have potential, they also present significant challenges, encouraging research into unexplored areas.

Research on Fault Attacks of Lightweight Cryptographic Algorithms

Fault attacks exploit fault injection techniques to compromise cryptographic systems, compromising device operations and leaking secret keys. Lightweight block cipher algorithms are suitable for resource-constrained environments, ensuring security while reducing hardware and software overhead. This paper summarizes the current research status of fault attacks on some lightweight cryptographic algorithms, including block ciphers, stream ciphers, etc. These studies are crucial for addressing system vulnerabilities, protecting personal privacy, and enhancing the stability of embedded systems. The aim of this paper is to provide comprehensive references for cryptographers, assisting in selecting suitable algorithms and defense strategies.

Relationship between Access to Clean Water and Child Growth and Development in Kenya

Purpose: The aim of the study was to analyze the relationship between access to clean water and child growth and development in Kenya. Methodology: This study adopted a desk methodology. A desk study research design is commonly known as secondary data collection. This is basically collecting data from existing resources preferably because of its low cost advantage as compared to a field research. Our current study looked into already published studies and reports as the data was easily accessed through online journals and libraries. Findings: Access to clean water significantly impacts child growth and development in Kenya by reducing waterborne diseases like diarrhea, improving hygiene practices, and ensuring adequate hydration. Children with reliable access to clean water experience better nutritional outcomes and enhanced immune function, crucial for their physical growth and cognitive development. Unique Contribution to Theory, Practice and Policy: Ecological systems theory (Bronfenbrenner), health belief model (Rosenstock) & social determinants of health (SDH) theory may be used to anchor future studies on relationship between access to clean water and child growth and development in Kenya. Expand the implementation of effective water purification technologies and community water supply projects in resource-constrained environments. Advocate for integrated policies that prioritize water security alongside nutrition-sensitive interventions. Policies should support multi-sectoral collaboration between health, water, and sanitation sectors to address the complex interplay between water access and child nutrition.

DSEUNet: A lightweight UNet for dynamic space grouping enhancement for skin lesion segmentation

A prevalent cancer, skin cancer, requires precise segmentation for effective diagnosis and treatment. Despite the availability of numerous effective, lightweight neural network models, there remains a significant gap in their application in resource-limited settings, particularly in mobile healthcare. This study aims to bridge this gap by developing a model combining high efficiency and robust performance in such environments. This study introduces the Dynamic Spatial Group Enhanced Network (DSEUNet), significantly reducing the computational loads and parameter quantities while ensuring competitive segmentation accuracy. Our essential contribution is developing a model that excels in segmentation accuracy and addresses computational and parameter efficiency challenges in mobile healthcare applications. In extensive trials on the ISIC2017 and ISIC2018 datasets, DSEUNet not only surpasses traditional models like UNet in accuracy but does so with far fewer parameters and reduced computational demand. This research contributes to developing more accessible and efficient tools for medical image analysis, particularly in resource-constrained environments. As far as we know, this is the first model with 23.306 KB parameters and 9.497 M FLOPs.

Distributed Maritime Operations and the Medical Challenge: How a Medical Common Operating Picture Can Help Bridge the Gap.

This work explores the challenges of delivering medical care in the geographically dispersed and resource-constrained environment of Distributed Maritime Operations (DMO) and Expeditionary Advanced Base Operations (EABO). Traditional medical planning approaches may struggle to adapt to the vast operational space, extended evacuation times, and limited medical force present in these scenarios. The concept of a Medical Common Operating Picture (COP) emerges as a potential solution. By providing a shared view of the medical situation across the theater, encompassing logistics, personnel, and patient data, a medical COP has the potential to facilitate medical command and control (MED C2) in DMO/EABO. The implementation of a medical COP has the potential to optimize resource allocation, enhance situational awareness, streamline medical evacuation, and reduce healthcare provider moral injury in large-scale combat operations. A medical COP will allow medical planners to make informed decisions on triage, resupply, and evacuation, ensuring the best use of limited medical resources. This is done by leveraging a comprehensive understanding of the medical landscape, enabling informed clinical and operational decision-making by humanitarian and combat personnel respectively. A fully realized medical COP system will enable a dynamic theater evacuation policy, balancing the conflicting needs of patient care at higher echelons with the operational expediency of returning servicemembers to their operational units, thereby maximizing evacuation effectiveness. It will further enable medical personnel to perform dynamic casualty triage based on operational realities, mitigating potential ethical dilemmas. Implementing such a medical COP system will require overcoming communication limitations to facilitate data exchange and potentially integrating clinical decision support tools for real-time data analysis and recommendations. It will also require the rapid adoption of modernized operational medicine documentation solutions by medical assets within the operational forces. Ultimately, this work suggests that a medical COP has the potential to bridge the gap between traditional medical planning and the unique demands of DMO/EABO, ultimately optimizing casualty care, maximizing resource efficiency, and preserving the fighting force.