In medical imaging, deep learning-based solutions have achieved state-of-the-art performance. However, reliability restricts the integration of deep learning into practical medical workflows since conventional deep learning frameworks cannot quantitatively assess model uncertainty. In this work, we propose to address this shortcoming by utilizing a Bayesian deep network capable of estimating uncertainty to assess oral cancer image classification reliability. We evaluate the model using a large intraoral cheek mucosa image dataset captured using our customized device from high-risk population to show that meaningful uncertainty information can be produced. In addition, our experiments show improved accuracy by uncertainty-informed referral. The accuracy of retained data reaches roughly 90% when referring either 10% of all cases or referring cases whose uncertainty value is greater than 0.3. The performance can be further improved by referring more patients. The experiments show the model is capable of identifying difficult cases needing further inspection.

Oral cancer, a pervasive and rapidly growing malignant disease, poses a significant global health concern. Early and accurate diagnosis is pivotal for improving patient outcomes. Automatic diagnosis methods based on artificial intelligence have shown promising results in the oral cancer field, but the accuracy still needs to be improved for realistic diagnostic scenarios. Vision Transformers (ViT) have outperformed learning CNN models recently in many computer vision benchmark tasks. This study explores the effectiveness of the Vision Transformer and the Swin Transformer, two cutting-edge variants of the transformer architecture, for the mobile-based oral cancer image classification application. The pre-trained Swin transformer model achieved 88.7% accuracy in the binary classification task, outperforming the ViT model by 2.3%, while the conventional convolutional network model VGG19 and ResNet50 achieved 85.2% and 84.5% accuracy. Our experiments demonstrate that these transformer-based architectures outperform traditional convolutional neural networks in terms of oral cancer image classification, and underscore the potential of the ViT and the Swin Transformer in advancing the state of the art in oral cancer image analysis.

Convolutional neural networks have demonstrated excellent performance in oral cancer detection and classification. However, the end-to-end learning strategy makes CNNs hard to interpret, and it can be challenging to fully understand the decision-making procedure. Additionally, reliability is also a significant challenge for CNN based approaches. In this study, we proposed a neural network called the attention branch network (ABN), which combines the visual explanation and attention mechanisms to improve the recognition performance and interpret the decision-making simultaneously. We also embedded expert knowledge into the network by having human experts manually edit the attention maps for the attention mechanism. Our experiments have shown that ABN performs better than the original baseline network. By introducing the Squeeze-and-Excitation (SE) blocks to the network, the cross-validation accuracy increased further. Furthermore, we observed that some previously misclassified cases were correctly recognized after updating by manually editing the attention maps. The cross-validation accuracy increased from 0.846 to 0.875 with the ABN (Resnet18 as baseline), 0.877 with SE-ABN, and 0.903 after embedding expert knowledge. The proposed method provides an accurate, interpretable, and reliable oral cancer computer-aided diagnosis system through visual explanation, attention mechanisms, and expert knowledge embedding.

Oral non-neoplastic and neoplastic lesions have similar clinical manifestations, increasing the risk of inaccurate screening decisions that adversely affect oral cancer (OC) outcomes. Tobacco-use-related changes in the oral soft tissues may affect the accuracy of “smart” oral screening modalities. Because smoking is such a strong predictor of OC risk, it may overwhelm the impact of other variables on algorithm performance. The objective was to evaluate the screening accuracy in tobacco users vs. non-users of a previously developed prototype smartphone and machine-learning algorithm-based oral health screening modality. 318 subjects with healthy mucosa or oral lesions were allocated into either a “tobacco smoker” group or a “tobacco non-smoker” group. Next, intraoral autofluorescence (AFI) and polarized white light images (pWLI), risk factors as well as clinical signs and symptoms were recorded using the prototype screening platform. OC risk status as determined by the algorithm was compared with OC risk evaluation by an oral medicine specialist (gold standard). The screening platform achieved 80.0% sensitivity, 87.5% specificity, 83.67% agreement with specialist screening outcome in tobacco smokers, and 62.1% sensitivity, 82.9% specificity, 73.1% agreement with specialist screening outcome in non-smokers. Tobacco use should be carefully weighted as a variable in the architecture of any imaging-based screening algorithm for OC risk.

SignificanceOral squamous cell carcinoma (OSCC) is exceedingly prevalent and deadly in South Asia, especially the Indian subcontinent, affecting 15 out of every 100,000 people and claiming over 70,000 lives annually. This problem is compounded by inadequate medical infrastructure for screening and cancer care, especially in rural areas.AimWe sought to build and evaluate a theranostic, low-cost, intraoral device to enable simultaneous multichannel optical imaging and photodynamic therapy (PDT) for the detection and treatment of early oral cancer.ApproachThe system’s performance was assessed using buccal mucosal tissue phantoms containing protoporphyrin IX (PpIX), as well as OSCC in vitro tumor models photosensitized with 5-aminolevulinic acid (ALA)-induced PpIX. A murine xenograft model consisting of athymic nude mice implanted with TR146 OSCC cells was used to evaluate the capability of the device to function as a PDT light source with integrated monitoring of PpIX photobleaching as a dosimetry reporter. Experiments were complemented by Monte Carlo modeling of light delivery.ResultsIn tissue phantoms the intraoral device reported a linear correlation up to 50 μM PpIX concentration and the extent of photobleaching increased with increasing light dose up to 100 J/cm2. PpIX fluorescence contrast was found to improve by using ratiometric analysis of PpIX (red) and autofluorescence (green) fluorescence signals. In mouse models that received PDT treatment via the intraoral device, tumor volume decreased significantly, with histological analysis showing necrosis extending to a depth of 3.0 to 3.5 mm after PDT and consistent with Monte Carlo modeling of light delivery.ConclusionA compact, handheld, and low-cost intraoral device with a dental camera form factor is able to carry out simultaneous multichannel white light and fluorescence imaging and image-guided PDT light delivery. The ability to quantify the extent of photobleaching as a dosimetry surrogate points to the potential for real-time treatment monitoring with the same hardware.