Revolutionizing healthcare: a comparative insight into deep learning’s role in medical imaging DOI Creative Commons
Vivek Kumar Prasad, Ashwin Verma, Pronaya Bhattacharya

и другие.

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

Опубликована: Дек. 4, 2024

Recently, Deep Learning (DL) models have shown promising accuracy in analysis of medical images. Alzeheimer Disease (AD), a prevalent form dementia, uses Magnetic Resonance Imaging (MRI) scans, which is then analysed via DL models. To address the model computational constraints, Cloud Computing (CC) integrated to operate with Recent articles on DL-based MRI not discussed datasets specific different diseases, makes it difficult build model. Thus, article systematically explores tutorial approach, where we first discuss classification taxonomy imaging datasets. Next, present case-study AD using methods. We analyse three distinct models-Convolutional Neural Networks (CNN), Visual Geometry Group 16 (VGG-16), and an ensemble approach-for predictive outcomes. In addition, designed novel framework that offers insight into how various layers interact dataset. Our architecture comprises input layer, cloud-based layer responsible for preprocessing execution, diagnostic issues alerts after successful prediction. According our simulations, CNN outperformed other test 99.285%, followed by VGG-16 85.113%, while lagged disappointing 79.192%. cloud serves as efficient mechanism image processing safeguarding patient confidentiality data privacy.

Язык: Английский

Facial Classification for Autism Spectrum Disorder DOI Creative Commons
Maram Fahaad Almufareh, Samabia Tehsin, Mamoona Humayun

и другие.

Deleted Journal, Год журнала: 2024, Номер 3(3)

Опубликована: Март 29, 2024

Autism spectrum disorder (ASD) is a mental condition that affects people’s learning, communication, and expression in their daily lives. ASD usually makes it difficult to socialize communicate with others, also sometimes shows repetition of certain behaviors. can be cause intellectual disability. big challenge neural development, specially children. It very important identified at an early stage for timely guidance intervention. This research identifies the application deep learning vision transformer (ViT) models classification facial images autistic non-autistic ViT are powerful used image tasks. model applies architectures analyze input patches connect information achieve global-level information. By employing these techniques, this study aims contribute toward detection. showing good results identifying features associated ASD, leading diagnostics. Results show model’s capability distinguishing faces

Язык: Английский

Процитировано

4
CIL-Net: Densely Connected Context Information Learning Network for Boosting Thyroid Nodule Segmentation Using Ultrasound Images DOI
Haider Ali, Mingzhao Wang, Juanying Xie

и другие.

Cognitive Computation, Год журнала: 2024, Номер 16(3), С. 1176 - 1197

Опубликована: Май 1, 2024

Язык: Английский

Процитировано

4
Comparative Study of Object Recognition Utilizing Machine Learning Techniques DOI

Tiyas Sarkar,

Manik Rakhra, Vikrant Sharma

и другие.

Опубликована: Май 9, 2024

Язык: Английский

Процитировано

4
An intelligent deep hash coding network for content-based medical image retrieval for healthcare applications DOI Creative Commons
Lichao Cui, Mingxin Liu

Egyptian Informatics Journal, Год журнала: 2024, Номер 27, С. 100499 - 100499

Опубликована: Июль 5, 2024

The proliferation of medical imaging in clinical diagnostics has led to an overwhelming volume image data, presenting a challenge for efficient storage, management, and retrieval. Specifically, the rapid growth use modalities such as Computed Tomography (CT) X-rays outpaced capabilities conventional retrieval systems, necessitating more sophisticated approaches assist decision-making research. Our study introduces novel deep hash coding-based Content-Based Medical Image Retrieval (CBMIR) framework that uses convolutional neural network (CNN) combined with coding accurate model integrates Dense block-based feature learning network, block, spatial attention block enhance extraction specific imaging. We reduce dimensionality by applying Reconstruction Independent Component Analysis (RICA) algorithm while preserving diagnostic information. achieves mean average precision (mAP) 0.85 on ChestX-ray8, 0.82 TCIA-CT, 0.84 MIMIC-CXR, LIDC-IDRI datasets, times 675 ms, 663 735 748 respectively. Comparisons ResNet DenseNet confirm effectiveness our model, enhancing significantly

Язык: Английский

Процитировано

3
Revolutionizing healthcare: a comparative insight into deep learning’s role in medical imaging DOI Creative Commons
Vivek Kumar Prasad, Ashwin Verma, Pronaya Bhattacharya

и другие.

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

Опубликована: Дек. 4, 2024

Recently, Deep Learning (DL) models have shown promising accuracy in analysis of medical images. Alzeheimer Disease (AD), a prevalent form dementia, uses Magnetic Resonance Imaging (MRI) scans, which is then analysed via DL models. To address the model computational constraints, Cloud Computing (CC) integrated to operate with Recent articles on DL-based MRI not discussed datasets specific different diseases, makes it difficult build model. Thus, article systematically explores tutorial approach, where we first discuss classification taxonomy imaging datasets. Next, present case-study AD using methods. We analyse three distinct models-Convolutional Neural Networks (CNN), Visual Geometry Group 16 (VGG-16), and an ensemble approach-for predictive outcomes. In addition, designed novel framework that offers insight into how various layers interact dataset. Our architecture comprises input layer, cloud-based layer responsible for preprocessing execution, diagnostic issues alerts after successful prediction. According our simulations, CNN outperformed other test 99.285%, followed by VGG-16 85.113%, while lagged disappointing 79.192%. cloud serves as efficient mechanism image processing safeguarding patient confidentiality data privacy.

Язык: Английский

Процитировано

3