Deep-Learning-Based Automated Identification and Visualization of Oral Cancer in Optical Coherence Tomography Images

Yang, Zihan; Pan, Hongming; Shang, Jun; Zhang, Jun; Liang, Yanmei

doi:10.3390/biomedicines11030802

Cited by 14 publications

(6 citation statements)

References 39 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Interpretability Methods : Research into interpretability methods for deep learning models is critical. Developing techniques to explain the model’s decisions can enhance trust and facilitate its integration into clinical settings ( 33 ).…”

Section: Resultsmentioning

confidence: 99%

Oral squamous cell carcinoma detection using EfficientNet on histopathological images

Albalawi,

Thakur,

Ramakrishna

et al. 2024

Front. Med.

View full text Add to dashboard Cite

IntroductionOral Squamous Cell Carcinoma (OSCC) poses a significant challenge in oncology due to the absence of precise diagnostic tools, leading to delays in identifying the condition. Current diagnostic methods for OSCC have limitations in accuracy and efficiency, highlighting the need for more reliable approaches. This study aims to explore the discriminative potential of histopathological images of oral epithelium and OSCC. By utilizing a database containing 1224 images from 230 patients, captured at varying magnifications and publicly available, a customized deep learning model based on EfficientNetB3 was developed. The model’s objective was to differentiate between normal epithelium and OSCC tissues by employing advanced techniques such as data augmentation, regularization, and optimization.MethodsThe research utilized a histopathological imaging database for Oral Cancer analysis, incorporating 1224 images from 230 patients. These images, taken at various magnifications, formed the basis for training a specialized deep learning model built upon the EfficientNetB3 architecture. The model underwent training to distinguish between normal epithelium and OSCC tissues, employing sophisticated methodologies including data augmentation, regularization techniques, and optimization strategies.ResultsThe customized deep learning model achieved significant success, showcasing a remarkable 99% accuracy when tested on the dataset. This high accuracy underscores the model’s efficacy in effectively discerning between normal epithelium and OSCC tissues. Furthermore, the model exhibited impressive precision, recall, and F1-score metrics, reinforcing its potential as a robust diagnostic tool for OSCC.DiscussionThis research demonstrates the promising potential of employing deep learning models to address the diagnostic challenges associated with OSCC. The model’s ability to achieve a 99% accuracy rate on the test dataset signifies a considerable leap forward in earlier and more accurate detection of OSCC. Leveraging advanced techniques in machine learning, such as data augmentation and optimization, has shown promising results in improving patient outcomes through timely and precise identification of OSCC.

show abstract

Section: Resultsmentioning

confidence: 99%

Oral squamous cell carcinoma detection using EfficientNet on histopathological images

Albalawi,

Thakur,

Ramakrishna

et al. 2024

Front. Med.

View full text Add to dashboard Cite

show abstract

“…Our approach is unique as it employs segmentation strategies to characterize the epithelial layer, rather than to classify disease status. This contrasts with prior in vivo oral OCT studies, which were used to triage healthy, pre-cancerous, and cancerous samples [ 16 , 38 , 39 ] or to classify states of dysplasia [ 38 ]. While these methods achieved high agreement, they are constrained by rigid classification labels, and either have limited transparency in the diagnostic process or employ explainability methods to foster physician trust.…”

Section: Discussionmentioning

confidence: 99%

Three-Dimension Epithelial Segmentation in Optical Coherence Tomography of the Oral Cavity Using Deep Learning

Hill,

Malone,

Liu

et al. 2024

Cancers

View full text Add to dashboard Cite

This paper aims to simplify the application of optical coherence tomography (OCT) for the examination of subsurface morphology in the oral cavity and reduce barriers towards the adoption of OCT as a biopsy guidance device. The aim of this work was to develop automated software tools for the simplified analysis of the large volume of data collected during OCT. Imaging and corresponding histopathology were acquired in-clinic using a wide-field endoscopic OCT system. An annotated dataset (n = 294 images) from 60 patients (34 male and 26 female) was assembled to train four unique neural networks. A deep learning pipeline was built using convolutional and modified u-net models to detect the imaging field of view (network 1), detect artifacts (network 2), identify the tissue surface (network 3), and identify the presence and location of the epithelial–stromal boundary (network 4). The area under the curve of the image and artifact detection networks was 1.00 and 0.94, respectively. The Dice similarity score for the surface and epithelial–stromal boundary segmentation networks was 0.98 and 0.83, respectively. Deep learning (DL) techniques can identify the location and variations in the epithelial surface and epithelial–stromal boundary in OCT images of the oral mucosa. Segmentation results can be synthesized into accessible en face maps to allow easier visualization of changes.

show abstract

“…The first intraoperative application was published by Sunny SP, and OCT significantly differentiated OSCC from dysplastic lesions or healthy tissue, visualizing the microarchitecture of the resected tissues without any changes in the specimen integrity or clinical workflow [ 118 , 119 ]. Furthermore, the automatic identification algorithm for OCT images based on deep learning may provide decision support for the screening and diagnosis of oral cancer [ 120 ].…”

Section: Spectroscopy For the Intraoperative Assessment Of The Resect...mentioning

confidence: 99%

Intraoperative Assessment of Resection Margin in Oral Cancer: The Potential Role of Spectroscopy

Vlocskó,

Piffkó,

Janovszky

2023

Cancers

View full text Add to dashboard Cite

In parallel with the increasing number of oncological cases, the need for faster and more efficient diagnostic tools has also appeared. Different diagnostic approaches are available, such as radiological imaging or histological staining methods, but these do not provide adequate information regarding the resection margin, intraoperatively, or are time consuming. The purpose of this review is to summarize the current knowledge on spectrometric diagnostic modalities suitable for intraoperative use, with an emphasis on their relevance in the management of oral cancer. The literature agrees on the sensitivity, specificity, and accuracy of spectrometric diagnostic modalities, but further long-term prospective, multicentric clinical studies are needed, which may standardize the intraoperative assessment of the resection margin and the use of real-time spectroscopic approaches.

show abstract

Deep-Learning-Based Automated Identification and Visualization of Oral Cancer in Optical Coherence Tomography Images

Cited by 14 publications

References 39 publications

Oral squamous cell carcinoma detection using EfficientNet on histopathological images

Oral squamous cell carcinoma detection using EfficientNet on histopathological images

Three-Dimension Epithelial Segmentation in Optical Coherence Tomography of the Oral Cavity Using Deep Learning

Intraoperative Assessment of Resection Margin in Oral Cancer: The Potential Role of Spectroscopy

Contact Info

Product

Resources

About