Опубликована: Май 27, 2024
Язык: Английский
Neural Computing and Applications, Год журнала: 2024, Номер 36(14), С. 7877 - 7902
Опубликована: Фев. 22, 2024
Abstract Prostate cancer is the one of most dominant among males. It represents leading death causes worldwide. Due to current evolution artificial intelligence in medical imaging, deep learning has been successfully applied diseases diagnosis. However, recent studies prostate classification suffers from either low accuracy or lack data. Therefore, present work introduces a hybrid framework for early and accurate segmentation using learning. The proposed consists two stages, namely stage stage. In stage, 8 pretrained convolutional neural networks were fine-tuned Aquila optimizer used classify patients normal ones. If patient diagnosed with cancer, segmenting cancerous spot overall image U-Net can help diagnosis, here comes importance trained on 3 different datasets order generalize framework. best reported accuracies are 88.91% MobileNet “ISUP Grade-wise Cancer” dataset 100% ResNet152 “Transverse Plane Dataset” precisions 89.22% 100%, respectively. model gives an average AUC 98.46% 0.9778, respectively, “PANDA: Resized Train Data (512 × 512)” dataset. results give indicator acceptable performance
Язык: Английский
Процитировано
19
Neural Computing and Applications, Год журнала: 2023, Номер 35(17), С. 12793 - 12831
Опубликована: Март 8, 2023
Язык: Английский
Процитировано
32
Computer Modeling in Engineering & Sciences, Год журнала: 2024, Номер 139(3), С. 2399 - 2450
Опубликована: Янв. 1, 2024
Sign language, a visual-gestural language used by the deaf and hard-of-hearing community, plays crucial role in facilitating communication promoting inclusivity.Sign recognition (SLR), process of automatically recognizing interpreting sign gestures, has gained significant attention recent years due to its potential bridge gap between hearing impaired world.The emergence continuous development deep learning techniques have provided inspiration momentum for advancing SLR.This paper presents comprehensive up-to-date analysis advancements, challenges, opportunities learning-based recognition, focusing on past five research.We explore various aspects SLR, including data acquisition technologies, datasets, evaluation methods, different types neural networks.Convolutional Neural Networks (CNN) Recurrent (RNN) shown promising results fingerspelling isolated recognition.However, nature poses leading exploration advanced network models such as Transformer model (CSLR).Despite several challenges remain field SLR.These include expanding achieving user independence systems, exploring input modalities, effectively fusing features, modeling co-articulation, improving semantic syntactic understanding.Additionally, developing lightweight architectures mobile applications is practical implementation.By addressing these we can further advance improve hearing-impaired community.
Язык: Английский
Процитировано
7
Neural Computing and Applications, Год журнала: 2024, Номер 36(27), С. 17199 - 17219
Опубликована: Июнь 6, 2024
Abstract Autism Spectrum Disorder (ASD) is a developmental condition resulting from abnormalities in brain structure and function, which can manifest as communication social interaction difficulties. Conventional methods for diagnosing ASD may not be effective the early stages of disorder. Hence, diagnosis crucial to improving patient's overall health well-being. One alternative method autism facial expression recognition since autistic children typically exhibit distinct expressions that aid distinguishing them other children. This paper provides deep convolutional neural network (DCNN)-based real-time emotion system kids. The proposed designed identify six emotions, including surprise, delight, sadness, fear, joy, natural, assist medical professionals families recognizing intervention. In this study, an attention-based YOLOv8 (AutYOLO-ATT) algorithm proposed, enhances model's performance by integrating attention mechanism. outperforms all classifiers metrics, achieving precision 93.97%, recall 97.5%, F1-score 92.99%, accuracy 97.2%. These results highlight potential real-world applications, particularly fields where high essential.
Язык: Английский
Процитировано
7
The European Physical Journal Special Topics, Год журнала: 2025, Номер unknown
Опубликована: Март 27, 2025
Язык: Английский
Процитировано
1
Artificial Intelligence Review, Год журнала: 2022, Номер 56(7), С. 7403 - 7456
Опубликована: Дек. 15, 2022
Язык: Английский
Процитировано
22
Neural Computing and Applications, Год журнала: 2024, Номер 36(22), С. 13381 - 13465
Опубликована: Апрель 20, 2024
Abstract This paper proposes a hybrid Modified Coronavirus Herd Immunity Aquila Optimization Algorithm (MCHIAO) that compiles the Enhanced Optimizer (ECHIO) algorithm and (AO). As one of competitive human-based optimization algorithms, (CHIO) exceeds some other biological-inspired algorithms. Compared to CHIO showed good results. However, gets confined local optima, accuracy large-scale global problems is decreased. On hand, although AO has significant exploitation capabilities, its exploration capabilities are insufficient. Subsequently, novel metaheuristic optimizer, (MCHIAO), presented overcome these restrictions adapt it solve feature selection challenges. In this paper, MCHIAO proposed with three main enhancements issues reach higher optimal results which cases categorizing, enhancing new genes’ value equation using chaotic system as inspired by behavior coronavirus generating formula switch between expanded narrowed exploitation. demonstrates it’s worth contra ten well-known state-of-the-art algorithms (GOA, MFO, MPA, GWO, HHO, SSA, WOA, IAO, NOA, NGO) in addition CHIO. Friedman average rank Wilcoxon statistical analysis ( p -value) conducted on all testing 23 benchmark functions. test well 29 CEC2017 Moreover, tests 10 CEC2019 Six real-world used validate against same twelve classical functions, including 24 unimodal 44 multimodal respectively, exploitative explorative evaluated. The significance technique for functions demonstrated -values calculated rank-sum test, found be less than 0.05.
Язык: Английский
Процитировано
5
Computers, Год журнала: 2024, Номер 13(4), С. 106 - 106
Опубликована: Апрель 22, 2024
In recent years, Sign Language Recognition (SLR) has become an additional topic of discussion in the human–computer interface (HCI) field. The most significant difficulty confronting SLR recognition is finding algorithms that will scale effectively with a growing vocabulary size and limited supply training data for signer-independent applications. Due to its sensitivity shape information, automated based on hidden Markov models (HMMs) cannot characterize confusing distributions observations gesture features sufficiently precise parameters. order simulate uncertainty hypothesis spaces, many scholars provide extension HMMs, utilizing higher-order fuzzy sets generate interval-type-2 HMMs. This expansion helpful because it brings fuzziness conventional HMM mapping under control. neutrosophic are used this work deal indeterminacy practical setting. Existing HMMs consider uncertain information includes indeterminacy. However, model successfully identifies best route between states when there vagueness. three membership functions (truth, indeterminate, falsity grades) more layers autonomy assessing HMM’s uncertainty. approach could be extensive hence seeks solve scalability issue. addition, may function independently signer, without needing gloves or any other input devices. experimental results demonstrate nearly as computationally difficult but similar performance robust variations.
Язык: Английский
Процитировано
4
Computers, Год журнала: 2024, Номер 13(6), С. 153 - 153
Опубликована: Июнь 19, 2024
This article emphasises the urgent need for appropriate communication tools communities of people who are deaf or hard-of-hearing, with a specific emphasis on Arabic Sign Language (ArSL). In this study, we use long short-term memory (LSTM) models in conjunction MediaPipe to reduce barriers effective and social integration communities. The model design incorporates LSTM units an attention mechanism handle input sequences extracted keypoints from recorded gestures. layer selectively directs its focus toward relevant segments sequence, whereas handles temporal relationships encodes sequential data. A comprehensive dataset comprised fifty frequently used words numbers ArSL was collected developing recognition model. comprises many instances gestures by five volunteers. results experiment support effectiveness proposed approach, as achieved accuracies more than 85% (individual volunteers) 83% (combined data). high level precision potential artificial intelligence-powered translation software improve hearing impairments enable them interact larger community easily.
Язык: Английский
Процитировано
4
Bioengineering, Год журнала: 2024, Номер 11(6), С. 629 - 629
Опубликована: Июнь 19, 2024
Prostate cancer is a significant health concern with high mortality rates and substantial economic impact. Early detection plays crucial role in improving patient outcomes. This study introduces non-invasive computer-aided diagnosis (CAD) system that leverages intravoxel incoherent motion (IVIM) parameters for the of prostate (PCa). IVIM imaging enables differentiation water molecule diffusion within capillaries outside vessels, offering valuable insights into tumor characteristics. The proposed approach utilizes two-step segmentation through use three U-Net architectures extracting tumor-containing regions interest (ROIs) from segmented images. performance CAD thoroughly evaluated, considering optimal classifier comparing diagnostic value commonly used apparent coefficient (ADC). results demonstrate combination central zone (CZ) peripheral (PZ) features Random Forest Classifier (RFC) yields best performance. achieves an accuracy 84.08% balanced 82.60%. showcases sensitivity (93.24%) reasonable specificity (71.96%), along good precision (81.48%) F1 score (86.96%). These findings highlight effectiveness accurately segmenting diagnosing PCa. represents advancement methods early PCa, showcasing potential machine learning techniques. developed solution has to revolutionize PCa diagnosis, leading improved outcomes reduced healthcare costs.
Язык: Английский
Процитировано
4