Detection and isolation of brain tumors in cancer patients using neural network techniques in MRI images DOI Creative Commons
Mahdi Mir, Zaid Saad Madhi, Ali Hamid AbdulHussein

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Окт. 7, 2024

MRI imaging primarily focuses on the soft tissues of human body, typically performed prior to a patient's transfer surgical suite for medical procedure. However, utilizing images tumor diagnosis is time-consuming process. To address these challenges, new method automatic brain was developed, employing combination image segmentation, feature extraction, and classification techniques isolate specific region interest in an corresponding tumor. The proposed this study comprises five distinct steps. Firstly, pre-processing conducted, various filters enhance quality. Subsequently, thresholding applied facilitate segmentation. Following extraction performed, analyzing morphological structural properties images. Then, selection carried out using principal component analysis (PCA). Finally, artificial neural network (ANN). In total, 74 unique features were extracted from each image, resulting dataset 144 observations. Principal employed select top 8 most effective features. Artificial Neural Networks (ANNs) leverage comprehensive data selective knowledge. Consequently, approach evaluated compared with alternative methods, significant improvements precision, accuracy, F1 score. demonstrated notable increases 99.3%, 97.3%, 98.5% Sensitivity These findings highlight efficiency accurately segmenting classifying

Язык: Английский

Efficient differential privacy enabled federated learning model for detecting COVID-19 disease using chest X-ray images DOI Creative Commons

Rawia Ahmed,

Praveen Kumar Reddy Maddikunta,

Thippa Reddy Gadekallu

и другие.

Frontiers in Medicine, Год журнала: 2024, Номер 11

Опубликована: Июнь 3, 2024

The rapid spread of COVID-19 pandemic across the world has not only disturbed global economy but also raised demand for accurate disease detection models. Although many studies have proposed effective solutions early and prediction with Machine Learning (ML) Deep learning (DL) based techniques, these models remain vulnerable to data privacy security breaches. To overcome challenges existing systems, we introduced Adaptive Differential Privacy-based Federated (DPFL) model predicting from chest X-ray images which introduces an innovative adaptive mechanism that dynamically adjusts levels on real-time sensitivity analysis, improving practical applicability (FL) in diverse healthcare environments. We compared analyzed performance this distributed a traditional centralized model. Moreover, enhance by integrating FL approach stopping achieve efficient minimal communication overhead. ensure without compromising utility accuracy, evaluated under various noise scales. Finally, discussed strategies increasing model’s accuracy while maintaining robustness as well privacy.

Язык: Английский

Процитировано

2
URCA: Uncertainty-based region clipping algorithm for semi-supervised medical image segmentation DOI

Chendong Qin,

Yongxiong Wang, Jiapeng Zhang

и другие.

Computer Methods and Programs in Biomedicine, Год журнала: 2024, Номер 254, С. 108278 - 108278

Опубликована: Июнь 12, 2024

Язык: Английский

Процитировано

2
Redefining retinal vessel segmentation: empowering advanced fundus image analysis with the potential of GANs DOI Creative Commons
Badar Almarri, B. Naveen Kumar, H. Pai

и другие.

Frontiers in Medicine, Год журнала: 2024, Номер 11

Опубликована: Окт. 21, 2024

Retinal vessel segmentation is a critical task in fundus image analysis, providing essential insights for diagnosing various retinal diseases. In recent years, deep learning (DL) techniques, particularly Generative Adversarial Networks (GANs), have garnered significant attention their potential to enhance medical analysis. This paper presents novel approach by harnessing the capabilities of GANs. Our method, termed GANVesselNet, employs specialized GAN architecture tailored intricacies structures. dual-path network employed, featuring an Auto Encoder-Decoder (AED) pathway and UNet-inspired pathway. unique combination enables efficiently capture multi-scale contextual information, improving accuracy segmentation. Through extensive experimentation on publicly available datasets, including STARE DRIVE, GANVesselNet demonstrates remarkable performance compared traditional methods state-of-the-art approaches. The proposed exhibits superior sensitivity (0.8174), specificity (0.9862), (0.9827) segmenting vessels dataset, achieves commendable results DRIVE dataset with (0.7834), (0.9846), (0.9709). Notably, previously unseen data, underscoring its real-world clinical applications. Furthermore, we present qualitative visualizations generated segmentations, illustrating network’s proficiency accurately delineating vessels. summary, this introduces powerful By capitalizing advanced GANs incorporating architecture, offers quantum leap accuracy, opening new avenues enhanced analysis improved decision-making.

Язык: Английский

Процитировано

2
Fingertip Video Dataset for Non-Invasive Diagnosis of Anemia Using ResNet-18 Classifier DOI Creative Commons

Humera Sabir,

Kifayat Ullah Khan,

Omer Ishaq

и другие.

IEEE Access, Год журнала: 2024, Номер 12, С. 68880 - 68892

Опубликована: Янв. 1, 2024

Hemoglobin is the iron containing protein in red blood cells which carries oxygen from lungs to rest of body tissues. Accurate measurement hemoglobin essential for diagnosing anemia, a condition characterized by deficiency cells. This particularly vital before initiating transfusions thalassemia patients. Non-invasive estimation levels can be achieved through photoplethysmography (PPG)-based methods. PPG an optical method measure volume changes successive heart beats. signals obtained fingertip videos using light source and photodetector. SmartphonePPG utilizes smartphone's flashlight as its camera photodetector acquire signals, offering affordable portable point-of-care tool. Despite ubiquity smartphones, their cameras often contain noise, making feature selection characteristics challenging. While PPG-based methods are invaluable, lack real-world datasets poses significant challenge maximizing benefits technology. In this paper, we introduce dataset comprising 1-minute video recordings 150 anemic patients, camera. The dataset, publicly accessible research purposes a , covers age range 6 months 32 years, with diverse values (4.3 gm/dL - 12.4 gm/dL). Utilizing propose deep learning-based technique employing ResNet-18 architecture estimate levels. approach eliminates need manual extraction overcoming limitation existing smartphonePPG-based systems. Our model achieves level RMSE 0.81-1.39 when compared gold standard laboratory method, Complete Blood Count (CBC) test reports.In contrast, HemaApp, state-of-the-art utilizing machine classifier (SVM), yields 1.7 on our dataset. accuracy simplicity position it promising alternative non-invasive

Язык: Английский

Процитировано

1
Multi-threshold image segmentation using a boosted whale optimization: case study of breast invasive ductal carcinomas DOI
Jinge Shi, Yi Chen, Zhennao Cai

и другие.

Cluster Computing, Год журнала: 2024, Номер 27(10), С. 14891 - 14949

Опубликована: Авг. 7, 2024

Язык: Английский

Процитировано

1
Advancements in Tumor Diagnostics through Carbon Dot‐Assisted Photoacoustic Imaging DOI Open Access

Rajan Patyal,

Khushboo Warjurkar,

Vinay Sharma

и другие.

Advanced Optical Materials, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 7, 2024

Abstract Serendipitously discovered, carbon dots (CDs) have attracted significant attention as a potential contrast agent for photoacoustic imaging (PAI) in the biomedical sector. CDs play an essential role PAI, contributing significantly to early detection of diseases and monitoring treatment progress, particularly tumor imaging. This review emphasizes domain highlighting their characteristics like biocompatibility, enhanced spatial resolution, optical absorption NIR region, facile surface functionalization tumor‐ targeted The study explores use enhancing resolution PAI improved visualization tumors organs such breast, cervical, liver, gastrointestinal, skin, cardiovascular system, nervous others. Challenges associated with optimizing signal‐to‐noise ratio ensuring stability under physiological conditions, also been discussed.

Язык: Английский

Процитировано

1
GC-MS Analysis and ADMET Properties of Tom Brown Weaning Meal and its Subsequent Effect on Liver Enzymes and Learning/Memory Parameters in Male Wistar Rats: A Docking and in Vivo Study DOI Creative Commons
Ekementeabasi Aniebo Umoh,

Agnes Igimi Odey,

Condoleezza Bohneji Mbu

и другие.

Natural Product Communications, Год журнала: 2024, Номер 19(11)

Опубликована: Ноя. 1, 2024

Background: The paucity of information on the effect Tom Brown's weaning meal liver and learning memory functions necessitated this study. Methods: Fifteen rats were acclimatized for a week used They divided into control, combined, Brown experimental groups. Rat Chow, Chow/Tombrown, Feed alone given to animals accordingly. At end four-week feeding period, enzymes (AST, ALT, ALP) parameters assessed. GC-MS ADMET properties done its ligands. Eleven Ligands with zero violations using Lipinski rule five (ROF) docked netrin, AST, ALT. Results: ALP results groups presented as mean ± SEM 67.89 3.15 Iu/L, 71.68 1.30 Iu/l, 73.65 0.89 Iu/l; 129.81 1.77 129.51 1.84 130.94 1.31 Iu/L; 22.10 1.24 23.28 0.61 22.48 1.29 respectively. There was no significant difference among in or other assessed study (P > 0.05). 5-hydroxymethyl furfural carpaine ligands better docking score. Conclusions: non-significant values long-term is evident having these parameters. are possible compounds that could enhance leraning/memory from results. However, they had low peak areas GCMS result not seen.

Язык: Английский

Процитировано

1
Challenges issues and future recommendations of deep learning techniques for SARS-CoV-2 detection utilising X-ray and CT images: a comprehensive review DOI Creative Commons
Md Shofiqul Islam, Fahmid Al Farid, F. M. Javed Mehedi Shamrat

и другие.

PeerJ Computer Science, Год журнала: 2024, Номер 10, С. e2517 - e2517

Опубликована: Дек. 24, 2024

The global spread of SARS-CoV-2 has prompted a crucial need for accurate medical diagnosis, particularly in the respiratory system. Current diagnostic methods heavily rely on imaging techniques like CT scans and X-rays, but identifying these images proves to be challenging time-consuming. In this context, artificial intelligence (AI) models, specifically deep learning (DL) networks, emerge as promising solution image analysis. This article provides meticulous comprehensive review imaging-based diagnosis using up May 2024. starts with an overview covering basic steps learning-based data sources, pre-processing methods, taxonomy techniques, findings, research gaps performance evaluation. We also focus addressing current privacy issues, limitations, challenges realm diagnosis. According taxonomy, each model is discussed, encompassing its core functionality critical assessment suitability detection. A comparative analysis included by summarizing all relevant studies provide overall visualization. Considering best deep-learning detection, conducts experiment twelve contemporary techniques. experimental result shows that MobileNetV3 outperforms other models accuracy 98.11%. Finally, elaborates explores potential future directions methodological recommendations advancement.

Язык: Английский

Процитировано

1
Potential roles of PIWI-interacting RNAs in breast cancer, a new therapeutic strategy DOI
Hongpeng Zhang, Yanshu Li

Pathology - Research and Practice, Год журнала: 2024, Номер 257, С. 155318 - 155318

Опубликована: Апрель 17, 2024

Язык: Английский

Процитировано

0
ASATrans: Adaptive spatial aggregation transformer for cervical nuclei segmentation on rough edges DOI Creative Commons
Hualin Sun, HU Sheng-yao

PLoS ONE, Год журнала: 2024, Номер 19(7), С. e0307206 - e0307206

Опубликована: Июль 12, 2024

The main characteristic of cervical cytopathy is reflected in the edge shape nuclei. Existing computer-aided diagnostic techniques can clearly segment individual nuclei, but cannot rough edges adherent nucleus. Therefore, we propose an effective method (ASATrans) to accurately nuclei by exploring adaptive spatial aggregation methods. ASATrans creates a Multi-Receptive Embedding Layer that samples patches using diverse-scale kernels. This approach provides cross-scale features each embedding, preventing semantic corruption might arise from mapping disparate analogous underlying representations. Furthermore, design Adaptive Pixel Adjustment Block introducing long-range dependency and aggregation. achieved through stratification process into distinct groups. Each group given exclusive sampling volume modulation scale, fostering collaborative learning paradigm combines local global dependencies. feature extraction achieves adaptability, mitigates interference unnecessary pixels, allows for better segmentation Extensive experiments on two datasets (HRASPP Dataset, ISBI Dataset), demonstrating our proposed outperforms other state-of-the-art methods large margin.

Язык: Английский

Процитировано

0