Brain tumor detection and screening using artificial intelligence techniques: Current trends and future perspectives

Raghavendra, U.; Gudigar, Anjan; Paul, Aritra; Goutham, T.S.; Inamdar, Mahesh Anil; Hegde, Ajay; Devi, Aruna; Ooi, Chui Ping; Deo, Ravinesh C.; Barua, Prabal Datta; Molinari, Filippo; Ciaccio, Edward J.; Acharya, U. Rajendra

doi:10.1016/j.compbiomed.2023.107063

Cited by 24 publications

(7 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AI algorithms trained on extensive datasets of medical images, possess the capacity to autonomously analyze intricate patterns, potentially revolutionizing the diagnostic process. 6,7 In recent studies for stroke classification, Ammar et al compared five deep learning models (ReNet50, VGG-16, Xception, InceptionV3, and InceptionResNetV2), with VGG-16 demonstrating the highest performance accuracy at 96.0% for intracranial hemorrhage. 8 Worachotsueptrakun compared four deep learning models (AlexNet, VGG-16, GoogleNet, and ResNet), with GoogleNet exhibiting the best performance, achieving accuracy, precision, recall, and F1-score of 92.00%, 94.00%, 83.96%, and 88.70%,…”

Section: Stoke Classification Techniquementioning

confidence: 99%

See 1 more Smart Citation

Comparative analysis of deep learning techniques for accurate stroke detection

Kaewlek,

Sitinwan,

Lueangaroon

et al. 2024

JAMS

View full text Add to dashboard Cite

Background: The traditional diagnosis of strokes through computed tomography (CT) heavily relies on radiologists’ expertise for accurate interpretation. However, the increasing demand for this critical task exceeds the available radiologist workforce, necessitating innovative solutions. This research addresses this challenge by introducing deep learning techniques to enhance the initial screening of stroke cases, thereby augmenting radiologists’ diagnostic capabilities. Objective: This study aims to compare four techniques for classifying stroke lesions in CT images. Materials and methods: Four distinct models-CNN-2-Model, LeNet, GoogleNet, and VGG-16-were trained using a dataset comprising 1,636 CT images, including 1,111 normal brain images and 525 stroke images. Seventy percent of the images were used to train the most effective deep learning model, and subsequently, these images were utilized to evaluate the performance of each model. The evaluation involved assessing accuracy, precision, sensitivity, specificity, F1 score, false positive rate, and AUC. Results: The evaluation process included a comprehensive statistical analysis of the models’ prediction results. The findings revealed that VGG-16 emerged as the top-performing deep learning model, achieving an impressive accuracy of 0.969, precision of 0.952, sensitivity of 0.952, specificity of 0.978, F1 score of 0.952, false positive rate of 0.022, and AUC of 0.965. Conclusion: In conclusion, deep learning techniques, particularly the VGG-16 model, demonstrate significant promise in enhancing the accuracy of stroke lesion classification in CT images. These findings underscore the potential of leveraging advanced technologies to address the growing challenges in stroke diagnosis and pave the way for more efficient and accessible healthcare solutions.

show abstract

Section: Stoke Classification Techniquementioning

confidence: 99%

“…The performance of the model was analyzed. The accuracy, positive predictive value (precision), sensitivity (recall), specificity, F-1 score, and false positive rate were calculated using Equations ( 1) to (6). The receiver operating characteristic curve (ROC) and area under the curve (AUC) were also evaluated.…”

Section: Comparative Analysis For Stroke Detectionmentioning

confidence: 99%

Comparative analysis of deep learning techniques for accurate stroke detection

Kaewlek,

Sitinwan,

Lueangaroon

et al. 2024

JAMS

View full text Add to dashboard Cite

show abstract

“…As an example, if a neural network learns to detect items in pictures, that are like lesions on the skin, this information may be utilized to identify other objects in radiological diagnostics [5].This approach in deep learning is called transfer learning, or Transfer Learning (TL). The present work is devoted to studies of the effectiveness of TL methods in the detection of brain tumors based on the analysis of MRI images using pre-trained deep convolutional neural networks [6]. To conduct computer experiments and compare performance, neural networks with pre-trained weights were used on the ImageNet database (a data set of 13 million high-resolution labeled images belonging to more than 20,000 categories) [7,8].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Transfer Methods for Biomedical Classification of Images

Al-Janabi

2024

Ijirme

View full text Add to dashboard Cite

Research using computers has been carried out on the effectiveness of applying deep learning transfer methods to solve the problem of identifying human brain tumors using MRI imaging. Various deep learning and fine-tuning methodologies of models have been proposed and implemented. The deep convolution networks MobileNetV2, VGG-16, Xception and ResNet-50, trained on the ImageNet image set, were used as basic models. A deep convolutional neural network 2D-CNN has also been developed and trained. A computer study of the performance indicators revealed that the fine-tuning method was effective On an enlarged data set, the Xception model outperformed other deep learning models in terms of accuracy: the clarity with which brain tumors are classified using MRI images was 94%, precision 97.7%, recall 94.01%, f1 score 96%, AUC 96.90%.

show abstract

“…Enzyme-linked immunosorbent assay (ELISA) [ 9 ], polymerase chain reaction (PCR) [ 10 ], clustered regularly interspaced short palindromic repeats Cas9 (CRISPR-Cas9) [ 11 ], loop-mediated isothermal amplification (LAMP), time-resolved fluorescence spectroscopy (TR-FS), radioimmunoassay (RIA), and electrophoresis [ 12 ] have been used for the detection of cancer biomarkers [ 13 ]. In addition, emerging technologies such as artificial intelligence, long read sequencing, microarrays, DNA methylation, and liquid biopsy are also committed to the development and high throughput profiling of many biomarkers to strengthen cancer management and improve early screening [ 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

Molybdenum Disulfide as Tunable Electrochemical and Optical Biosensing Platforms for Cancer Biomarker Detection: A Review

Qin,

Zhang,

2023

Biosensors

View full text Add to dashboard Cite

Cancer is a common illness with a high mortality. Compared with traditional technologies, biomarker detection, with its low cost and simple operation, has a higher sensitivity and faster speed in the early screening and prognosis of cancer. Therefore, extensive research has focused on the development of biosensors and the construction of sensing interfaces. Molybdenum disulfide (MoS2) is a promising two-dimensional (2D) nanomaterial, whose unique adjustable bandgap shows excellent electronic and optical properties in the construction of biosensor interfaces. It not only has the advantages of a high catalytic activity and low manufacturing costs, but it can also further expand the application of hybrid structures through different functionalization, and it is widely used in various biosensors fields. Herein, we provide a detailed introduction to the structure and synthesis methods of MoS2, and explore the unique properties and advantages/disadvantages exhibited by different structures. Specifically, we focus on the excellent properties and application performance of MoS2 and its composite structures, and discuss the widespread application of MoS2 in cancer biomarkers detection from both electrochemical and optical dimensions. Additionally, with the cross development of emerging technologies, we have also expanded the application of other emerging sensors based on MoS2 for early cancer diagnosis. Finally, we summarized the challenges and prospects of MoS2 in the synthesis, functionalization of composite groups, and applications, and provided some insights into the potential applications of these emerging nanomaterials in a wider range of fields.

show abstract

Brain tumor detection and screening using artificial intelligence techniques: Current trends and future perspectives

Cited by 24 publications

References 100 publications

Comparative analysis of deep learning techniques for accurate stroke detection

Comparative analysis of deep learning techniques for accurate stroke detection

Deep Learning Transfer Methods for Biomedical Classification of Images

Molybdenum Disulfide as Tunable Electrochemical and Optical Biosensing Platforms for Cancer Biomarker Detection: A Review

Contact Info

Product

Resources

About