Deep Learning and Transfer Learning for Malaria Detection

Jameela, Tayyaba; Athotha, Kavitha; Singh, Ninni; Gunjan, Vinit Kumar; Kahali, Sayan

doi:10.1155/2022/2221728

Cited by 23 publications

(3 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The constraints are high, since a detection with a high sensitivity is required as well as correct species identification. Recent advancements in computer vision, particularly in deep learning algorithms, have shown promise in detecting Plasmodium Falciparum parasites [ 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 ]. Although considerable progress has been made, the various methods developed to date are not sufficiently effective when parasitemia is low and parasites are tiny.…”

Section: Methodsmentioning

confidence: 99%

Investigating the Joint Amplitude and Phase Imaging of Stained Samples in Automatic Diagnosis

Hassini,

Dorizzi,

Thellier

et al. 2023

Sensors

View full text Add to dashboard Cite

The diagnosis of many diseases relies, at least on first intention, on an analysis of blood smears acquired with a microscope. However, image quality is often insufficient for the automation of such processing. A promising improvement concerns the acquisition of enriched information on samples. In particular, Quantitative Phase Imaging (QPI) techniques, which allow the digitization of the phase in complement to the intensity, are attracting growing interest. Such imaging allows the exploration of transparent objects not visible in the intensity image using the phase image only. Another direction proposes using stained images to reveal some characteristics of the cells in the intensity image; in this case, the phase information is not exploited. In this paper, we question the interest of using the bi-modal information brought by intensity and phase in a QPI acquisition when the samples are stained. We consider the problem of detecting parasitized red blood cells for diagnosing malaria from stained blood smears using a Deep Neural Network (DNN). Fourier Ptychographic Microscopy (FPM) is used as the computational microscopy framework to produce QPI images. We show that the bi-modal information enhances the detection performance by 4% compared to the intensity image only when the convolution in the DNN is implemented through a complex-based formalism. This proves that the DNN can benefit from the bi-modal enhanced information. We conjecture that these results should extend to other applications processed through QPI acquisition.

show abstract

Section: Methodsmentioning

confidence: 99%

Investigating the Joint Amplitude and Phase Imaging of Stained Samples in Automatic Diagnosis

Hassini,

Dorizzi,

Thellier

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Many studies ( 9 , 10 ) aimed to compare different deep learning algorithms, mainly convolutional neural networks (CNN) or their derivates, using publicly available databases such as the National Institute of Health Malaria data set ( 11 ) or the Broad Bioimage Benchmark Collection (BBBC) ( 12 ). The more frequent drawbacks observed in these studies were the lack of a test data set, with the results calculated only on the validation data set; no patient-level results but only smear images or cell-sized results; too homogenous staining (not reflecting real diversity in routine practice); and issues related with the segmentation process.…”

Section: Introductionmentioning

confidence: 99%

Automatic patient-level recognition of four Plasmodium species on thin blood smear by a real-time detection transformer (RT-DETR) object detection algorithm: a proof-of-concept and evaluation

Guemas,

Routier,

Ghelfenstein-Ferreira

et al. 2024

Microbiol Spectr

View full text Add to dashboard Cite

Malaria remains a global health problem, with 247 million cases and 619,000 deaths in 2021. Diagnosis of Plasmodium species is important for administering the appropriate treatment. The gold-standard diagnosis for accurate species identification remains the thin blood smear. Nevertheless, this method is time-consuming and requires highly skilled and trained microscopists. To overcome these issues, new diagnostic tools based on deep learning are emerging. This study aimed to evaluate the performances of a real-time detection transformer (RT-DETR) object detection algorithm to discriminate Plasmodium species on thin blood smear images. The algorithm was trained and validated on a data set consisting in 24,720 images from 475 thin blood smears corresponding to 2,002,597 labels. Performances were calculated with a test data set of 4,508 images from 170 smears corresponding to 358,825 labels coming from six French university hospitals. At the patient level, the RT-DETR algorithm exhibited an overall accuracy of 79.4% (135/170) with a recall of 74% (40/54) and 81.9% (95/116) for negative and positive smears, respectively. Among Plasmodium- positive smears, the global accuracy was 82.7% (91/110) with a recall of 90% (38/42), 81.8% (18/22), and 76.1% (35/46) for P. falciparum , P. malariae, and P. ovale/vivax, respectively. The RT-DETR model achieved a World Health Organization (WHO) competence level 2 for species identification. Besides, the RT-DETR algorithm may be run in real-time on low-cost devices such as a smartphone and could be suitable for deployment in low-resource setting areas lacking microscopy experts. IMPORTANCE Malaria remains a global health problem, with 247 million cases and 619,000 deaths in 2021. Diagnosis of Plasmodium species is important for administering the appropriate treatment. The gold-standard diagnosis for accurate species identification remains the thin blood smear. Nevertheless, this method is time-consuming and requires highly skilled and trained microscopists. To overcome these issues, new diagnostic tools based on deep learning are emerging. This study aimed to evaluate the performances of a real-time detection transformer (RT-DETR) object detection algorithm to discriminate Plasmodium species on thin blood smear images. Performances were calculated with a test data set of 4,508 images from 170 smears coming from six French university hospitals. The RT-DETR model achieved a World Health Organization (WHO) competence level 2 for species identification. Besides, the RT-DETR algorithm may be run in real-time on low-cost devices and could be suitable for deployment in low-resource setting areas.

show abstract

“…OOdwin et al developed an algorithm for recognizing unlearned species using the Xception model. Siddiqui et al detected dengue using quicker R-CNN and Inception V2 [11]. An insect classification and identification system was created via AlexNet utilizing a support vector machine and SVM to distinguish between Aedes albopictus and Aedes aegypti Linnaeus, 1762.…”

Section: Introductionmentioning

confidence: 99%

An Innovative Deep Learning Method to Diagnose Mosquito-Borne Illnesses in Blood Image Analysis

Shaimaa Saadoon Mahmood ALrfae

2023

IJRITCC

View full text Add to dashboard Cite

Introduction: Malaria, an infectious illness carried by the bite of infected mosquitoes, is a significant public health concern, especially in Africa. The management of mosquito-human contact is crucial to mitigate its transmission. Artificial intelligence, including machine learning and deep learning techniques, is being utilized to enhance the diagnosis and identification of mosquito species. This advancement aims to facilitate the development of more efficient control measures. Aims and Objective: To analyze the efficiency of three deep learning models in identifying blood-borne diseases by evaluating the macro and micro picture of blood samples. Method: In this retrospective investigation, three deep learning algorithms, namely Convolutional Neural Networks (CNN), MobileNetV2, and ResNet50, were used to identify mosquito-borne illnesses, focusing on malaria. The research used a dataset of 120 blood samples gathered over one year from the hospital's pathology department. The CNN model streamlines preprocessing with multilayer perceptrons, simplifying malaria component extraction. MobileNetV2, a lightweight network, outperforms others with fewer parameters. Its compact blocks in Dense-MobileNet models minimize constraints and computation expenses. ResNet50 resolves degradation issues with a residual structure, preventing overfitting as hidden layers increase. Results: The study evaluated three deep learning models (CNN, MobileNetV2, and ResNet50) for medical classification. The study also demonstrated improved True Positive Rates as False Positive Rates increased, indicating better accurate identification while controlling false positives. ResNet50 consistently outperformed the other models, showcasing its superior performance. The study revealed high precision scores for all models, classifying "Uninfected" and "Infected" cases. ResNet50 exhibited slightly higher precision, indicating its precision-based superiority. Overall, all models demonstrated vital accuracy, and ResNet50 showed exceptional performance. The study found that ResNet50 performs better in True Positive and False Positive Rates. Conclusion: The study has concluded that ResNet50 has shown the best performance in detecting blood-borne diseases.

show abstract

Deep Learning and Transfer Learning for Malaria Detection

Cited by 23 publications

References 36 publications

Investigating the Joint Amplitude and Phase Imaging of Stained Samples in Automatic Diagnosis

Investigating the Joint Amplitude and Phase Imaging of Stained Samples in Automatic Diagnosis

Automatic patient-level recognition of four Plasmodium species on thin blood smear by a real-time detection transformer (RT-DETR) object detection algorithm: a proof-of-concept and evaluation

An Innovative Deep Learning Method to Diagnose Mosquito-Borne Illnesses in Blood Image Analysis

Contact Info

Product

Resources

About