Image processing analysis of oral cancer, oral potentially malignant disorders, and other oral diseases using optical instruments

Fluorescence Visualization-Guided Surgery Improves Local Control for Mandibular Squamous Cell Carcinoma

BACKGROUND

Local recurrence is common in mandibular squamous cell carcinoma (SCC). Fluorescence visualization is a noninvasive technology that can detect oral epithelial dysplasia around mandibular SCC, and it can potentially reduce local recurrence.

PURPOSE

The purpose of this study was to measure and compare local control (LC) between fluorescence visualization-guided surgery (FVS) and conventional surgery for patients with Stages I or II mandibular SCC.

STUDY DESIGN, SETTING, SAMPLE

This retrospective cohort study was conducted at Tokyo Dental College, Chiba Hospital, or Chiba Dental Center. The medical records of mandibular SCC patients from 2000 to 2021 were analyzed. Patients from any sex and 18 years of age or older with complete records who received surgery for mandibular SCC in the early stages were included in this study.

PREDICTOR VARIABLE

The predictor variable was operative treatment and was divided into 2 groups, conventional or FVS.

MAIN OUTCOME VARIABLES

The outcome variable is 5-year LC defined as no recurrence at or within 20 mm of the surgical site.

COVARIATES

Covariates included demographic variables of age, sex, clinical and pathological characteristics, forms of resection, lifestyle, and quality of life.

ANALYSES

Data analysis was performed by carrying out χ2 tests. Survival outcome was performed by the Kaplan-Meier method, which was used to calculate and stratify the log-rank test; P values <.05 indicated statistical significance.

RESULTS

This study sample was composed of 56 subjects with a mean age of 68.5 years old (standard deviation 13.7), and 33 (58.9%) were female. There were 36 (64.3%) and 20 (35.7%) subjects in the conventional and FVS groups. The characteristics and quality of life did not differ significantly between the 2 groups. Five-year LC with FVS was statistically significantly higher than conventional surgery (P = .04, 94.4 vs 77.2%). Multivariate analysis for LC rate only identified FVS (P = .004; hazard ratio = 0.11, 95% confidence interval = 0.46, 0.88).

CONCLUSION AND RELEVANCE

On mandibular SCC, LC was 94.4% in FVS versus 77.2% in conventional surgery. For mandibular SCC at stages I and II, FVS was associated with improved LC.

Machine learning for detection and classification of oral potentially malignant disorders: A conceptual review

Oral potentially malignant disorders represent precursor lesions that may undergo malignant transformation to oral cancer. There are many known risk factors associated with the development of oral potentially malignant disorders, and contribute to the risk of malignant transformation. Although many advances have been reported to understand the biological behavior of oral potentially malignant disorders, their clinical features that indicate the characteristics of malignant transformation are not well established. Early diagnosis of malignancy is the most important factor to improve patients' prognosis. The integration of machine learning into routine diagnosis has recently emerged as an adjunct to aid clinical examination. Increased performances of artificial intelligence AI-assisted medical devices are claimed to exceed the human capability in the clinical detection of early cancer. Therefore, the aim of this narrative review is to introduce artificial intelligence terminology, concepts, and models currently used in oncology to familiarize oral medicine scientists with the language skills, best research practices, and knowledge for developing machine learning models applied to the clinical detection of oral potentially malignant disorders.

Combination of fluorescence visualization and iodine solution-guided surgery for local control of early tongue cancer

The control of enclosed oral epithelial dysplasia is important for the control of oral cancer. Fluorescence visualization and iodine solution are able to detect oral epithelial dysplasia and surrounding oral cancer. The purpose of this study was to clarify the effectiveness of combining fluorescence visualization and iodine solution-guided surgery for early tongue cancer. Participants comprised 264 patients with primary early tongue cancer who underwent surgery. The surgical margin was set at 10 mm outside the clinical tumour, and 5 mm outside the area of fluorescence visualization loss, and 5mm outside the iodine unstained area. The 5-year disease-free survival rate was 87.1% vs 76.1% (P = 0.016) and the 5-year local control rate was 98.6% vs 93.0% (P = 0.008) for combination-guided surgery when compared to conventional surgery. Positive margin rates were 0% for cancer, and 6.5% and 0% for low- and high-grade dysplasia, respectively, with combination-guided surgery (P = 0.257). Multivariate analysis revealed that combination-guided surgery (odds ratio 0.140, 95% confidence interval 0.045-0.437; P < 0.001) and intraoperative frozen section examination (odds ratio 0.302; 95% confidence interval 0.115-0.791; P = 0.015) were significantly associated with local control. The combination of fluorescence visualization and iodine solution are effective in selecting surgical margins for early tongue cancer.

The Effectiveness of Population Mass Screening to Oral Cancer: A Simulation Study

Background: Mass screening of high-risk populations for oral cancer has proven to be effective in reducing oral cancer mortality. However, the magnitude of the effectiveness of the various screening scenarios has rarely been addressed. Methods: We developed a simulation algorithm for a prospective cohort under various oral cancer screening scenarios. A hypothetical cohort of 8 million participants aged ≥30 years with cigaret smoking and/or betel quid chewing habits was constructed based on parameters extracted from studies on oral cancer screening. The results of a population-based screening program in Taiwan and a randomized controlled trial in India were used to validate the fitness; then, the effectiveness of the model was determined by changing the screening parameters. Results: There was a reduction in the risk of advanced oral cancer by 40% (relative risk [RR] = 0.60, 95% confidence interval [CI]:0.59-0.62) and oral cancer mortality by 29% (RR = 0.71, 95% CI: 0.69-0.73) at the 6-year follow-up in a screening scenario similar to the biennial screening in Taiwan, with a 55.1% attendance rate and 92.6% referral rate. The incremental effect in reducing advanced oral cancer was approximately 5% with a short 1-year screening frequency, and the corresponding reduction in mortality was, on average, 6.5%. The incremental reduction in advanced oral cancer per 10% increase in the compliance rate was 3% to 4%, while only 1% to 2% reduction was noted per 10% increase in the referral rate. The effectiveness of screening in reducing advanced oral cancer was 5% to 6% less when both betel quid chewing and alcohol drinking habits were present. Conclusion: Our computer simulation model demonstrated the effect of screening on the reduction in oral cancer mortality under various scenarios. The results provide screening policymakers with the necessary guidance to implement screening programs to save lives.

Artificial intelligence for oral cancer diagnosis: What are the possibilities?

Oral cancer could be prevented. The primary strategy is based on prevention. Most patients with oral cancer present to the hospital network with advanced staging and a low chance of cure. This condition may be related to physicians' difficulty of making an early diagnosis. With the advancement of information technology, artificial intelligence (AI) holds great promise in terms of assisting in diagnosis. Few machine learning algorithms have been developed for this purpose to date. In this paper, we will discuss the possibilities for diagnosing oral cancer using AI as a tool, as well as the implications for the population. A set of photographic images of oral lesions has been segmented, indicating not only the area of the lesion but also the class of lesion associated with it. Different neural network architectures were trained with the goal of fine segmentation (pixel by pixel), classification of image crops, and classification of whole images based on the presence or absence of a lesion. The accuracy results are acceptable, opening up possibilities not only for identifying lesions but also for classifying the pathology associated with them.

Evaluation of Oral Mucosal Lesions Using the IllumiScan® Fluorescence Visualisation Device: Distinguishing Squamous Cell Carcinoma

We evaluated whether fluorescence intensity (FI) and its coefficient of variation (CV) can be used to diagnose squamous cell carcinoma (SCC) through IllumiScan®, an oral mucosa fluorescence visualisation (FV) device. Overall, 190 patients with oral mucosal lesions (OMLs; SCC, 59; non-SCC OMLs, 131) and 49 patients with normal oral mucosa (NOM) were enrolled between January 2019 and March 2021. The FI of the images was analysed using image analysis software. After establishing regions of interest for SCC, non-SCC, and NOM, the average FI, standard deviation (SD), and CV were compared. There was a significant difference in the average FI for all pairs of comparisons. The SD was not significantly different between the SCC and NOM groups (p = 0.07). The CV differed significantly for NOM (p < 0.001) and non-SCC groups (p < 0.001) relative to the SCC group but was not different between NOM and non-SCC groups (p = 0.15). Univariate analysis of SCC and non-SCC groups showed significant differences for all factors, except age. However, multivariate analysis showed a significant intergroup difference only in the CV (p = 0.038). Therefore, analysing the CV in FV images of OML may be useful for the diagnosis of oral cancer.

Non-invasive imaging of oral potentially malignant and malignant lesions: A systematic review and meta-analysis

Non-invasive (NI) imaging techniques have been developed to overcome the limitations of invasive biopsy procedures, which is the gold standard in diagnosis of oral dysplasia and Oral Squamous Cell Carcinoma (OSCC). This systematic review and meta- analysis was carried out with an aim to investigate the efficacy of the NI-imaging techniques in the detection of dysplastic oral potentially malignant disorders (OPMDs) and OSCC. Records concerned in the detection of OPMDs, Oral Cancer were identified through search in PubMed, Science direct, Cochrane Library electronic database (January 2000 to October 2020) and additional manual searches. Out of 529 articles evaluated for eligibility, 56 satisfied the pre-determined inclusion criteria, including 13 varying NI-imaging techniques. Meta-analysis consisted 44 articles, wherein majority of the studies reported Autofluorescence (AFI-38.6%) followed by Chemiluminescence (CHEM), Narrow Band Imaging (NBI) (CHEM, NBI-15.9%), Fluorescence Spectroscopy (FS), Diffuse Reflectance Spectroscopy (DRS), (FS, DRS-13.6%) and 5aminolevulinic acid induced protoporphyrin IX fluorescence (5ALA induced PPIX- 6.8%). Higher sensitivities (Sen) and specificities (Spe) were obtained using FS (Sen:74%, Spe:96%, SAUC=0.98), DRS (Sen:79%, Spe:86%, SAUC = 0.91) and 5 ALA induced PPIX (Sen:91%, Spe:78%, SAUC = 0.98) in the detection of dysplastic OPMDs from non-dysplastic lesions(NDLs). AFI, FS, DRS, NBI showed higher sensitivities and SAUC (>90%) in differentiating OSCC from NDLs. Analysed NI-imaging techniques suggests the higher accuracy levels in the diagnosis of OSCC when compared to dysplastic OPMDs. 5 ALA induced PPIX, DRS and FS showed evidence of superior accuracy levels in differentiation of dysplastic OPMDs from NDLs, however results need to be validated in a larger number of studies.

Detecting Oral Cancer: The Potential of Artificial Intelligence

BACKGROUND

Physical inspection is a simple way to diagnose oral cancer. Most cases of oral cancer, on the contrary, are diagnosed late, resulting in needless mortality and morbidity. While screening high-risk populations appear to be helpful, these people are often found in areas with minimal access to health care. In this paper, we have reviewed several aspects related to oral cancer such as its cause, the risk factors associated with it, India's oral cancer situation at the moment, various screening methods, and the ability of artificial intelligence in the detection and classification purpose. Oral cancer results can be enhanced by combining imaging and artificial intelligence approaches for better detection and diagnosis.

OBJECTIVE

This paper aims to cover the various oral cancer screening detection techniques that use Artificial Intelligence (AI).

METHODS

In this paper, we have covered the imaging methods that are used in screening oral cancer and after that the potential of AI for the detection of oral cancer.

CONCLUSION

This paper covers some of the main concepts regarding oral cancer and various AI methods used to detect it.

Performance of deep convolutional neural network for classification and detection of oral potentially malignant disorders in photographic images

Oral potentially malignant disorders (OPMDs) are a group of conditions that can transform into oral cancer. The purpose of this study was to evaluate convolutional neural network (CNN) algorithms to classify and detect OPMDs in oral photographs. In this study, 600 oral photograph images were collected retrospectively and grouped into 300 images of OPMDs and 300 images of normal oral mucosa. CNN-based classification models were created using DenseNet-121 and ResNet-50. The detection models were created using Faster R-CNN and YOLOv4. The image data were randomly selected and assigned as training, validating, and testing data. The testing data were evaluated to compare the performance of the CNN models with the diagnosis results produced by oral and maxillofacial surgeons. DenseNet-121 and ResNet-50 were found to produce high efficiency in diagnosis of OPMDs, with an area under the receiver operating characteristic curve (AUC) of 95%. Faster R-CNN yielded the highest detection performance, with an AUC of 74.34%. For the CNN-based classification model, the sensitivity and specificity were 100% and 90%, respectively. For the oral and maxillofacial surgeons, these values were 91.73% and 92.27%, respectively. In conclusion, the DenseNet-121, ResNet-50 and Faster R-CNN models have potential for the classification and detection of OPMDs in oral photographs.

Algorithm mediated early detection of oral cancer from image analysis

OBJECTIVE

To develop Automatic Oral Cancer Detection algorithm for identification and differentiation of premalignant lesions from buccal cavity images for early detection of oral cancer, which may reduce related fatalities in developing countries.

STUDY DESIGN

The oral cavity images of normal, erythroplakia, and leukoplakia (20 images of each) were collected and processed using MATLAB image processing tools. First, maximum red value was used to differentiate between normal and abnormal. Second, mean red value was used for the selection of a processing path through YCbCr. Third, gray-level co-occurrence matrix (GLCM) based features were used to make final decisions. Images have been randomly divided and shuffled between training and test set to rigorously train the algorithm.

RESULTS

With 100% efficiency, normal images were separated from abnormal images in the first step by applying R value distribution with a cutoff R value, 11,900. Further, images with a mean R value >200 and <200 were processed by segmentation of Y plane and Cr plane, respectively. For the final decision, abnormal images were analyzed through the GLCM using the entropy feature as one of the key indicators, which can apply to the differentiation decision with 89% efficiency.

CONCLUSIONS

The developed algorithm can successfully differentiate premalignant lesions from normal. A graphic user interface was developed, which displays outcomes with reasonable accuracy.

Use of autofluorescence and fluorescent probes as a potential diagnostic tool for oral cancer: A systematic review

INTRODUCTION

The prognosis of patients with Oral squamous cell carcinoma (OSCC) are directly related to the stage of development of the tumor at the time of diagnosis, but it is estimated an average delay in diagnosis of 2-5 months. New non-invasive techniques for the early diagnosis of OSCC are being developed, such as methodologies to detect spectral changes of tumor cells. We conducted a systematic review to analyze the potential use of autofluorescence and/or fluorescent probes for OSCC diagnosis.

MATERIAL AND METHODS

Four databases (PubMed, Scopus, Embase and Web of Science) were used as research sources. Protocol was registered with PROSPERO. It was included studies that evaluated tissue autofluorescence and/or used fluorescent probes as a method of diagnosing and/or treatment of oral cancer in humans.

RESULTS

Forty-five studies were selected for this systematic review, of which 28 dealt only with autofluorescence, 18 on fluorescent probes and 1 evaluated both methods. The VELscope® was the most used device for autofluorescence, exhibiting sensitivity (33%-100%) and specificity (12%-88.6%). 5-Aminolevulinic acid (5-ALA) was the most used fluorescent probe, exhibiting high sensitivity (90%-100%) and specificity (51.3%-96%). Hypericin, rhodamine 6 G, rhodamine 610, porphyrin and γ-glutamyl hydroxymethyl rhodamine green have also been reported.

CONCLUSION

Thus, the autofluorescence and fluorescent probes can provide an accurate diagnosis of oral cancer, assisting the dentist during daily clinical activity, but it is not yet possible to suggest that this method may replace histopathological examination.

Countermeasure and opportunistic screening systems for oral cancer

OBJECTIVES

Tokyo Dental College started oral cancer screening in cooperation with a local dental association in 1992. Reveal the usefulness of Countermeasure and Opportunistic Screening Systems for Oral Cancer. The actual results of countermeasure and opportunistic oral cancer screening systems are reported.

MATERIALS AND METHODS

Countermeasure screening for the public was performed in each region, and opportunistic screening was performed in a general dental clinic of a cooperating physician.

RESULTS

In countermeasure screening, 19,721 persons were checked from 1992 to 2018; the gender ratio was 1:3. The close examination rate was 4.45%. The detection rates of oral cancer and oral potentially malignant disorders were 0.13% and 1.85%, respectively. In opportunistic screening, 29,912 persons were checked from 2006 to 2018; the gender ratio was 2:3. The close examination rate was 2.33%. The detection rates of oral cancer and oral potentially malignant disorders were 0.08% and 2.15%, respectively. The close examination rate was significantly lower in opportunistic screening than in countermeasure screening. The oral cancer detection rates and the positive predictive value for cancer were equivalent. In addition, the detection rate of oral potentially malignant disorders was significantly higher in opportunistic screening than in countermeasure screening.

CONCLUSION

Oral cancer detection rates were equivalent between countermeasure and opportunistic screenings, and opportunistic screening were more effective on number of participants and the close examination rate, and the detection rate of oral potentially malignant disorders.

Non-Invasive Early Detection of Oral Cancers Using Fluorescence Visualization with Optical Instruments

Simple Summary

Oral cancer has a high mortality rate. Then, oral cancer screening is needed for early detection and treatment. Fluorescence visualization is non-invasive, convenient, and in real-time, and examinations can be repeated. Our study aimed to show the usefulness of oral cancer screening with fluorescence visualization. A total of 502 patients were performed using fluorescence visualization that was analyzed using subjective and objective evaluation. Results of this study, subjective evaluation for detection oral cancer was high sensitivity and low specificity, while objective evaluation using imaging processing analysis was high sensitivity and high specificity. Therefore, oral cancer screening using fluorescence visualization is useful for the detection of oral cancer. The widespread use of this screening can reduce the mortality rate of oral cancer.

Abstract

Background: Oral cancer screening is important for early detection and early treatment, which help improve survival rates. Biopsy is the gold standard for a definitive diagnosis but is invasive and painful, while fluorescence visualization is non-invasive, convenient, and real-time, and examinations can be repeated using optical instruments. The purpose of this study was to clarify the usefulness of fluorescence visualization in oral cancer screening. Methods: A total of 502 patients, who were examined using fluorescence visualization with optical instruments in our hospitals between 2014 and 2019, were enrolled in this study. The final diagnosis was performed by pathological examination. Fluorescence visualization was analyzed using subjective and objective evaluations. Results: Subjective evaluations for detecting oral cancer offered 96.8% sensitivity and 48.4% specificity. Regarding the objective evaluations, sensitivity and specificity were 43.7% and 84.6% for mean green value, 55.2% and 67.0% for median green value, 82.0% and 44.2% for coefficient of variation of value, 59.6% and 45.3% for skewness, and 85.1% and 75.8% for value ratio. For the sub-analysis of oral cancer, all factors on objective and subjective evaluation showed no significant difference. Conclusions: Fluorescence visualization with subjective and objective evaluation is useful for oral cancer screening.

Strategies to improve diagnosis and risk assessment for oral cancer patients

To realise the benefits of new diagnostic techniques for the prediction of oral cancer, further validation and multicentre analyses are needed to determine their clinical impact in contemporary practice.

Improvement of oral cancer screening quality and reach: The promise of artificial intelligence

Oral cancer is easily detectable by physical (self) examination. However, many cases of oral cancer are detected late, which causes unnecessary morbidity and mortality. Screening of high-risk populations seems beneficial, but these populations are commonly located in regions with limited access to health care. The advent of information technology and its modern derivative artificial intelligence (AI) promises to improve oral cancer screening but to date, few efforts have been made to apply these techniques and relatively little research has been conducted to retrieve meaningful information from AI data. In this paper, we discuss the promise of AI to improve the quality and reach of oral cancer screening and its potential effect on improving mortality and unequal access to health care around the world.