Automated classification of cells into multiple classes in epithelial tissue of oral squamous cell carcinoma using transfer learning and convolutional neural network

Neural Networks, Год журнала: 2020, Номер 128, С. 47 - 60

Опубликована: Май 7, 2020

Язык: Английский

---

AUTOMATED DIAGNOSIS OF EPILEPSY USING CWT, HOS AND TEXTURE PARAMETERS

International Journal of Neural Systems, Год журнала: 2013, Номер 23(03), С. 1350009 - 1350009

Опубликована: Фев. 19, 2013

Epilepsy is a chronic brain disorder which manifests as recurrent seizures. Electroencephalogram (EEG) signals are generally analyzed to study the characteristics of epileptic In this work, we propose method for automated classification EEG into normal, interictal and ictal classes using Continuous Wavelet Transform (CWT), Higher Order Spectra (HOS) textures. First CWT plot was obtained then HOS texture features were extracted from these plots. Then statistically significant fed four classifiers namely Decision Tree (DT), K-Nearest Neighbor (KNN), Probabilistic Neural Network (PNN) Support Vector Machine (SVM) select best classifier. We observed that SVM classifier with Radial Basis Function (RBF) kernel function yielded results an average accuracy 96%, sensitivity 96.9% specificity 97% 23.6 s duration data. Our proposed technique can be used automatic seizure monitoring software. It also assist doctors cross check efficacy their prescribed drugs.

Язык: Английский

---

Automatic Detection of Oral Squamous Cell Carcinoma from Histopathological Images of Oral Mucosa Using Deep Convolutional Neural Network

International Journal of Environmental Research and Public Health, Год журнала: 2023, Номер 20(3), С. 2131 - 2131

Опубликована: Янв. 24, 2023

Worldwide, oral cancer is the sixth most common type of cancer. India in 2nd position, with highest number patients. To population patients, contributes to almost one-third total count. Among several types cancer, and dominant one squamous cell carcinoma (OSCC). The major reason for tobacco consumption, excessive alcohol unhygienic mouth condition, betel quid eating, viral infection (namely human papillomavirus), etc. early detection OSCC, its preliminary stage, gives more chances better treatment proper therapy. In this paper, author proposes a convolutional neural network model, automatic experimental purposes, histopathological images are considered. proposed model compared analyzed state-of-the-art deep learning models like VGG16, VGG19, Alexnet, ResNet50, ResNet101, Mobile Net Inception Net. achieved cross-validation accuracy 97.82%, which indicates suitability approach classification data.

Язык: Английский

---

Classifying histopathological images of oral squamous cell carcinoma using deep transfer learning

Heliyon, Год журнала: 2023, Номер 9(3), С. e13444 - e13444

Опубликована: Фев. 6, 2023

Oral cancer is a prevalent malignancy that affects the oral cavity in region of head and neck. The study malignant lesions an essential step for clinicians to provide better treatment plan at early stage cancer. Deep learning based computer-aided diagnostic system has achieved success many applications can accurate timely diagnosis lesions. In biomedical image classification, getting large training dataset challenge, which be efficiently handled by transfer as it retrieves general features from natural images adapted directly new dataset. this work, achieve effective deep system, classifications Squamous Cell Carcinoma (OSCC) histopathology are performed using two proposed approaches. first approach, identify best appropriate model differentiate between benign cancers, assisted convolutional neural networks (DCNNs), considered. To handle challenge small further increase efficiency model, pretrained VGG16, VGG19, ResNet50, InceptionV3, MobileNet, fine-tuned half layers leaving others frozen. second baseline DCNN architecture, trained scratch with 10 convolution proposed. addition, comparative analysis these models carried out terms classification accuracy other performance measures. experimental results demonstrate ResNet50 obtains substantially superior than selected well 96.6%, precision recall values 97% 96%, respectively.

Язык: Английский

---

Machine learning concepts applied to oral pathology and oral medicine: A convolutional neural networks' approach

Journal of Oral Pathology and Medicine, Год журнала: 2023, Номер 52(2), С. 109 - 118

Опубликована: Янв. 4, 2023

Abstract Introduction Artificial intelligence models and networks can learn process dense information in a short time, leading to an efficient, objective, accurate clinical histopathological analysis, which be useful improve treatment modalities prognostic outcomes. This paper targets oral pathologists, medicinists, head neck surgeons provide them with theoretical conceptual foundation of artificial intelligence‐based diagnostic approaches, special focus on convolutional neural networks, the state‐of‐the‐art deep learning. Methods The authors conducted literature review, network's foundations functionality were illustrated based unique interdisciplinary point view. Conclusion development computer vision methods for pattern recognition image analysis cancer has potential aid diagnosis prediction.

Язык: Английский

---

An ensemble deep learning model with empirical wavelet transform feature for oral cancer histopathological image classification

International Journal of Data Science and Analytics, Год журнала: 2024, Номер unknown

Опубликована: Фев. 5, 2024

Язык: Английский

---

Automated oral cancer identification using histopathological images: A hybrid feature extraction paradigm

Micron, Год журнала: 2011, Номер 43(2-3), С. 352 - 364

Опубликована: Окт. 7, 2011

Язык: Английский

---

Automated oral squamous cell carcinoma identification using shape, texture and color features of whole image strips

Tissue and Cell, Год журнала: 2019, Номер 63, С. 101322 - 101322

Опубликована: Дек. 3, 2019

Язык: Английский

---

Machine-Learning Applications in Oral Cancer: A Systematic Review

Applied Sciences, Год журнала: 2022, Номер 12(11), С. 5715 - 5715

Опубликована: Июнь 4, 2022

Over the years, several machine-learning applications have been suggested to assist in various clinical scenarios relevant oral cancer. We offer a systematic review identify, assess, and summarize evidence for reported uses areas of cancer detection prevention, prognosis, pre-cancer, treatment, quality life. The main algorithms applied context corresponded SVM, ANN, LR, comprising 87.71% total published articles field. Genomic, histopathological, image, medical/clinical, spectral, speech data were used most often predict four application found this review. In conclusion, our study has shown that are useful diagnosis, prevention potentially malignant lesions (pre-cancer) therapy. Nevertheless, we strongly recommended these methods daily practice.

Язык: Английский

---

Diagnosis of Hashimoto’s thyroiditis in ultrasound using tissue characterization and pixel classification

Proceedings of the Institution of Mechanical Engineers Part H Journal of Engineering in Medicine, Год журнала: 2013, Номер 227(7), С. 788 - 798

Опубликована: Апрель 16, 2013

Hashimoto's thyroiditis is the most common type of inflammation thyroid gland, and accurate diagnosis would be helpful to better manage disease process predict failure. Most published computer-based techniques that use ultrasound images for are limited by lack procedure standardization because individual investigators various initial settings. This article presents a computer-aided diagnostic technique uses grayscale features classifiers provide more objective reproducible classification normal thyroiditis-affected cases. In this paradigm, we extracted based on entropy, Gabor wavelet, moments, image texture, higher order spectra from 100 images. Significant were selected using t-test. The resulting feature vectors used build following three tenfold stratified cross validation technique: support vector machine, k-nearest neighbor, radial basis probabilistic neural network. Our results show combination 12 coupled with machine classifier polynomial kernel 1 linear gives highest accuracy 80%, sensitivity 76%, specificity 84%, positive predictive value 83.3% detection thyroiditis. proposed system novel have not yet been explored diagnosis. Even though only presented preliminary encouraging warrant analysis such powerful larger databases.

Язык: Английский

---