Search

Simple Summary: Lip and oral cavity cancer caused over 177,000 deaths globally in 2020, but patient survival increases with earlier diagnosis. One barrier to early diagnosis is the invasive nature of biopsies needed for diagnosis. Automated diagnosis systems have the potential to perform non-invasive diagnosis by pairing novel imaging data with deep learning models. Given the variability between patients, access to a sufficiently large training database from human subjects limits deep learning applications. We propose a model that maps non-invasive images of oral tissue to a diagnosis by encouraging the model to group normal samples close together (reducing variability between patients). Our model improves non-invasive oral cancer diagnosis through a robust training process that only requires a small amount of data. This work shows how we can address small data challenges through model architecture and training, rather than through the collection of larger databases or manual corrections and normalizations. Background: Multispectral autofluorescence lifetime imaging systems have recently been developed to quickly and non-invasively assess tissue properties for applications in oral cancer diagnosis. As a non-traditional imaging modality, the autofluorescence signal collected from the system cannot be directly visually assessed by a clinician and a model is needed to generate a diagnosis for each image. However, training a deep learning model from scratch on small multispectral autofluorescence datasets can fail due to inter-patient variability, poor initialization, and overfitting. Methods: We propose a contrastive-based pre-training approach that teaches the network to perform patient normalization without requiring a direct comparison to a reference sample. We then use the contrastive pre-trained encoder as a favorable initialization for classification. To train the classifiers, we efficiently use available data and reduce overfitting through a multitask framework with margin delineation and cancer diagnosis tasks. We evaluate the model over 67 patients using 10-fold cross-validation and evaluate significance using paired, one-tailed t-tests. Results: The proposed approach achieves a sensitivity of 82.08% and specificity of 75.92% on the cancer diagnosis task with a sensitivity of 91.83% and specificity of 79.31% for margin delineation as an auxiliary task. In comparison to existing approaches, our method significantly outperforms a support vector machine (SVM) implemented with either sequential feature selection (SFS) (p = 0.0261) or L1 loss (p = 0.0452) when considering the average of sensitivity and specificity. Specifically, the proposed approach increases performance by 2.75% compared to the L1 model and 4.87% compared to the SFS model. In addition, there is a significant increase in specificity of 8.34% compared to the baseline autoencoder model (p = 0.0070). Conclusions: Our method effectively trains deep learning models for small data applications when existing, large pre-trained models are not suitable for fine-tuning. While we designed the network for a specific imaging modality, we report the development process so that the insights gained can be applied to address similar challenges in other non-traditional imaging modalities. A key contribution of this paper is a neural network framework for multi-spectral fluorescence lifetime-based tissue discrimination that performs patient normalization without requiring a reference (healthy) sample from each patient at test time. [ABSTRACT FROM AUTHOR]

Significance: Diagnosis of cancerous and pre-cancerous oral lesions at early stages is critical for the improvement of patient care, to increase survival rates and minimize the invasiveness of tumor resection surgery. Unfortunately, oral precancerous and early-stage cancerous lesions are often difficult to distinguish from oral benign lesions with the existing diagnostic tools used during standard clinical oral examination. In consequence, early diagnosis of oral cancer can be achieved in only about 30% of patients. Therefore, clinical diagnostic technologies for fast, minimally invasive, and accurate oral cancer screening are urgently needed., Aim: This study investigated the use of multispectral autofluorescence imaging endoscopy for the automated and noninvasive discrimination of cancerous and precancerous from benign oral epithelial lesions., Approach: In vivo multispectral autofluorescence endoscopic images of clinically suspicious oral lesions were acquired from 67 patients undergoing tissue biopsy examination. The imaged lesions were classified as precancerous ( n = 4), cancerous ( n = 29), and benign ( n = 34) lesions based on histopathology diagnosis. Multispectral autofluorescence intensity feature maps were generated for each oral lesion and used to train and optimize support vector machine (SVM) models for automated discrimination of cancerous and precancerous from benign oral lesions., Results: After a leave-one-patient-out cross-validation strategy, an optimized SVM model developed with four multispectral autofluorescence features yielded levels of sensitivity and specificity of 85% and 71%, respectively and overall accuracy of 78% in the discrimination of cancerous/precancerous versus benign oral lesions., Conclusion: This study demonstrates the potentials of a computer-assisted detection system based on multispectral autofluorescence imaging endoscopy for the early detection of cancerous and precancerous oral lesions., Competing Interests: Disclosures The authors declare no conflicts of interest.

In contrast to previous studies that focused on classical machine learning algorithms and hand-crafted features, we present an end-to-end neural network classification method able to accommodate lesion heterogeneity for improved oral cancer diagnosis using multispectral autofluorescence lifetime imaging (maFLIM) endoscopy. Our method uses an autoencoder framework jointly trained with a classifier designed to handle overfitting problems with reduced databases, which is often the case in healthcare applications. The autoencoder guides the feature extraction process through the reconstruction loss and enables the potential use of unsupervised data for domain adaptation and improved generalization. The classifier ensures the features extracted are task-specific, providing discriminative information for the classification task. The data-driven feature extraction method automatically generates task-specific features directly from fluorescence decays, eliminating the need for iterative signal reconstruction. We validate our proposed neural network method against support vector machine (SVM) baselines, with our method showing a 6.5%-8.3% increase in sensitivity. Our results show that neural networks that implement data-driven feature extraction provide superior results and enable the capacity needed to target specific issues, such as inter-patient variability and the heterogeneity of oral lesions.Clinical relevance- We improve standard classification algorithms for in vivo diagnosis of oral cancer lesions from maFLIm for clinical use in cancer screening, reducing unnecessary biopsies and facilitating early detection of oral cancer.

We have performed a pilot clinical study, in which multispectral endogenous fluorescence (or autofluorescence) lifetime imaging (FLIM) was performed on clinically suspicious oral lesions of 73 patients undergoing tissue biopsy for oral dysplasia and cancer diagnosis. The results from this pilot study indicated that mild-dysplasia and early stage oral cancer could be detected from benign lesions using a computed aided diagnosis system developed based on biochemical and metabolic biomarkers derived from the endogenous FLIM images. The diagnostic performance of this novel FLIM clinical tool was estimated using a leave-onepatient-out cross-validation approach, which reported levels of sensitivity >90%, specificity >85%, and Area Under the Receiving Operating Curve (ROC-AUC) >0.9.