Advances and challenges in automated malaria diagnosis using digital microscopy imaging with artificial intelligence tools: A review

Frontiers in Microbiology, Journal Year: 2022, Volume and Issue: 13

Published: Nov. 15, 2022

Malaria is an infectious disease caused by parasites of the genus Plasmodium spp. It transmitted to humans bite infected female Anopheles mosquito. most common in resource-poor settings, with 241 million malaria cases reported 2020 according World Health Organization. Optical microscopy examination blood smears gold standard technique for diagnosis; however, it a time-consuming method and well-trained microscopist needed perform microbiological diagnosis. New techniques based on digital imaging analysis deep learning artificial intelligence methods are challenging alternative tool diagnosis diseases. In particular, systems Convolutional Neural Networks image detection emulate visualization expert. Microscope automation provides fast low-cost diagnosis, requiring less supervision. Smartphones suitable option microscopic allowing capture software identification parasites. addition, could be optimal solution malaria, tuberculosis, or Neglected Tropical Diseases endemic areas low resources. The implementation automated using smartphone applications new technologies low-income challenge achieve. Moreover, automating movement microscope slide autofocusing samples hardware would systemize procedure. These diagnostic tools join global effort fight against pandemic other poverty-related

Language: Английский

---

Application of deep learning and machine learning models to improve healthcare in sub-Saharan Africa: Emerging opportunities, trends and implications

Telematics and Informatics Reports, Journal Year: 2023, Volume and Issue: 11, P. 100097 - 100097

Published: Sept. 1, 2023

Deep learning and machine techniques present unmatched opportunities to improve healthcare in sub-Saharan Africa (SSA). However, there is a paucity of literature on AI-based applications deployed care SSA, which makes it challenging organise the research contributions highlight obstacles emerging areas that need be explored future. This study applied PRISMA (Preferred Reporting Items for Systematic Reviews Meta-Analysis) model conduct comprehensive review deep models SSA access while exploring opportunities, trends implications integrating healthcare. reveals AI can analyse derive inferences from massive health data early detection, diagnosis, monitoring chronic disorders, prediction diseases, large-scale public patterns help limit exposure contagious environments. facilitate development targeted interventions patient outcomes all stages treatment, drug monitoring, personalised medicine, control care. Integrating with tremendously assist professionals policymakers disease diagnosis making informed decisions. algorithms bias, poor formats, lack policies frameworks supporting integration data-driven solutions into systems hinder systems. There transparency ethical use crafting support Utilising also researchers workers move towards smart better comprehend future needs

Language: Английский

---

Enhancing parasitic organism detection in microscopy images through deep learning and fine-tuned optimizer

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: March 8, 2024

Abstract Parasitic organisms pose a major global health threat, mainly in regions that lack advanced medical facilities. Early and accurate detection of parasitic is vital to saving lives. Deep learning models have uplifted the sector by providing promising results diagnosing, detecting, classifying diseases. This paper explores role deep techniques detecting various organisms. The research works on dataset consisting 34,298 samples parasites such as Toxoplasma Gondii, Trypanosome, Plasmodium, Leishmania, Babesia, Trichomonad along with host cells like red blood white cells. These images are initially converted from RGB grayscale followed computation morphological features perimeter, height, area, width. Later, Otsu thresholding watershed applied differentiate foreground background create markers for identification interest. transfer VGG19, InceptionV3, ResNet50V2, ResNet152V2, EfficientNetB3, EfficientNetB0, MobileNetV2, Xception, DenseNet169, hybrid model, InceptionResNetV2, employed. parameters these fine-tuned using three optimizers: SGD, RMSprop, Adam. Experimental reveal when RMSprop applied, EfficientNetB0 achieve highest accuracy 99.1% loss 0.09. Similarly, SGD optimizer, InceptionV3 performs exceptionally well, achieving 99.91% 0.98. Finally, applying Adam InceptionResNetV2 excels, 99.96% 0.13, outperforming other optimizers. findings this signify coupled image processing methods generates highly efficient way detect classify

Language: Английский

---

Intelligent diagnostic model for malaria parasite detection and classification using imperative inception-based capsule neural networks

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Aug. 17, 2023

Abstract Malaria is an acute fever sickness caused by the Plasmodium parasite and spread infected Anopheles female mosquitoes. It causes catastrophic illness if left untreated for extended period, delaying exact treatment might result in development of further complications. The most prevalent method now available detecting malaria microscope. Under a microscope, blood smears are typically examined diagnosis. Despite its advantages, this time-consuming, subjective, requires highly skilled personnel. Therefore, automated diagnosis system imperative ensuring accurate efficient treatment. This research develops innovative approach utilizing urgent, inception-based capsule network to distinguish parasitized uninfected cells from microscopic images. diagnostic model incorporates neural networks based on Inception Imperative Capsule networks. inception block extracts rich characteristics images using pre-trained model, such as V3, which facilitates representation learning. Subsequently, dynamic detects parasites classifying them into healthy cells, enabling detection parasites. experiment results demonstrate significant improvement recognition. Compared traditional manual microscopy, proposed more faster. Finally, study demonstrates need provide robust solutions leveraging state-of-the-art technologies combat malaria.

Language: Английский

---

Classification of Malaria Cell Image using Inception-V3 Architecture

JOIV International Journal on Informatics Visualization, Journal Year: 2023, Volume and Issue: 7(2), P. 273 - 273

Published: May 5, 2023

Malaria is a severe global public health problem caused by the bite of infected mosquitoes. It can be cured, but only with early detection and effective, quick treatment. cause conditions if not properly diagnosed treated at an stage. In worst scenario, it death. This study aims focusing on classifying malaria cell images. classified as dangerous disease female Anophles mosquito. As such, leads to mortality when immediate action treatment fails administered. particular, this classify images utilizing Inception-V3 architecture. study, training was conducted 27,558 image data through architecture proposing 3 scenarios. The proposed scenario 1 model applies SGD optimizer generate loss value 0.13 accuracy 0.95; 2 Adam 0.09 0.96; lastly implements RMSprop 0.08 0.97. Applying three scenarios, results apparently indicate that using capable providing best 97% lowest value, compared 2. Further, test confirms in cells effectively.

Language: Английский

---

Artificial Intelligence-Based Approaches for Detection and Classification of Different Classes of Malaria Parasites Using Microscopic Images: A Systematic Review

Archives of Computational Methods in Engineering, Journal Year: 2023, Volume and Issue: 30(8), P. 4781 - 4800

Published: June 23, 2023

Language: Английский

---

Comparative Analysis of the Identification and Categorization of the Malaria Parasite Employing Recent Amalgamated Machine Learning Methodologies

Published: Jan. 24, 2025

Language: Английский

---

A 5G network based conceptual framework for real-time malaria parasite detection from thick and thin blood smear slides using modified YOLOv5 model

Digital Health, Journal Year: 2025, Volume and Issue: 11

Published: Feb. 1, 2025

Objective This paper aims to address the need for real-time malaria disease detection that integrates a faster prediction model with robust underlying network. The study first proposes 5G network-based healthcare system and then develops an automated capable of providing accurate diagnosis, particularly in areas limited diagnostic resources. Methods proposed leverages deep learning-based YOLOv5x algorithm detect parasites thick thin blood smear samples. network architecture was modified by introducing two squeeze-and-excitation (SENet) layers just before Upsample layers. is designed operate over networks efficiently, enabling remote smart solutions. Results demonstrated improved accuracy precision detecting on microscopic slides. inclusion SENet optimized network’s performance, making it suitable Conclusion Our exemplifies how generic one-stage object algorithm, such as YOLOv5x, can be repurposed objects small from visuals cost-effective manner By integrating computational efficiency learning connectivity networks, this significantly enhance capabilities contribute

Language: Английский

---

Deep learning method for malaria parasite evaluation from microscopic blood smear

Artificial Intelligence in Medicine, Journal Year: 2025, Volume and Issue: unknown, P. 103114 - 103114

Published: March 1, 2025

Language: Английский

---

Machine learning evidence towards eradication of malaria burden: A scoping review

Applied Computer Science, Journal Year: 2025, Volume and Issue: 21(1), P. 44 - 69

Published: March 31, 2025

Recent advancements have shown that shallow and deep learning models achieve impressive performance accuracies of over 97% 98%, respectively, in providing precise evidence for malaria control diagnosis. This effectiveness highlights the importance these enhancing our understanding management, which includes critical areas such as control, diagnosis economic evaluation burden. By leveraging predictive systems models, significant opportunities eradicating malaria, empowering informed decision-making facilitating development effective policies could be established. However, global burden is approximated at 95%, there a pressing need its eradication to facilitate achievement SDG targets related good health well-being. paper presents scoping review covering years 2018 2024, utilizing PRISMA-ScR protocol, with articles retrieved from three scholarly databases: Science Direct (9%), PubMed (41%), Google Scholar (50%). After applying exclusion inclusion criteria, final list 61 was extracted review. The results reveal decline research on machine techniques while steady increase approaches has been noted, particularly volume dimensionality data continue grow. In conclusion, clear utilize algorithms through real-time collection, model development, deployment evidence-based recommendations Future directions should focus standardized methodologies effectively investigate both models.

Language: Английский

---