Machine learning based on magnetic resonance imaging and clinical parameters helps predict mesenchymal-epithelial transition factor expression in oral tongue squamous cell carcinoma: a pilot study

OBJECTIVES

This study aimed to develop machine learning models to predict phosphorylated mesenchymal-epithelial transition factor (p-MET) expression in oral tongue squamous cell carcinoma (OTSCC) using magnetic resonance imaging (MRI)-derived texture features and clinical features.

METHODS

Thirty-four patients with OTSCC were retrospectively collected. Texture features were derived from preoperative MR images, including T2WI, apparent diffusion coefficient mapping, and contrast-enhanced (ce)-T1WI. Dimension reduction was performed consecutively with reproducibility analysis and an information gain algorithm. Five machine learning methods-AdaBoost, logistic regression (LR), naïve Bayes (NB), random forest (RF), and support vector machine (SVM)-were adopted to create models predicting p-MET expression. Their performance was assessed with fivefold cross-validation.

RESULTS

In total, 22 and 12 cases showed low and high p-MET expression, respectively. After dimension reduction, 3 texture features (ADC-Minimum, ce-T1WI-Imc2, and ce-T1WI-DependenceVariance) and 2 clinical features (depth of invasion [DOI] and T-stage) were selected with good reproducibility and best correlation with p-MET expression levels. The RF model yielded the best overall performance, correctly classifying p-MET expression status in 87.5% of OTSCCs with an area under the receiver operating characteristic curve of 0.875.

CONCLUSION

Differences in p-MET expression in OTSCCs can be noninvasively reflected in MRI-based texture features and clinical parameters. Machine learning can potentially predict biomarker expression levels, such as MET, in patients with OTSCC.

The application of radiomics machine learning models based on multimodal MRI with different sequence combinations in predicting cervical lymph node metastasis in oral tongue squamous cell carcinoma patients

BACKGROUND

The purpose of this study was to explore preliminary the performance of radiomics machine learning models based on multimodal MRI to predict the risk of cervical lymph node metastasis (CLNM) for oral tongue squamous cell carcinoma (OTSCC) patients.

METHODS

A total of 400 patients were enrolled in this study and divided into six groups according to the different combinations of MRI sequences. Group I consisted of patients with T1-weighted images (T1WI) and FS-T2WI (fat-suppressed T2-weighted images), group II consisted of patients with T1WI, FS-T2WI, and contrast enhanced MRI (CE-MRI), group III consisted of patients with T1WI, FS-T2WI, and T2-weighted images (T2WI), group IV consisted of patients with T1WI, FS-T2WI, CE-MRI, and T2WI, group V consisted of patients with T1WI, FS-T2WI, T2WI, and apparent diffusion coefficient map (ADC), and group VI consisted of patients with T1WI, FS-T2WI, CE-MRI, T2WI, and ADC. Machine learning models were constructed. The performance of the models was compared in each group.

RESULTS

The machine learning model in group IV including T1WI, FS-T2WI, T2WI, and CE-MRI presented best prediction performance, with AUCs of 0.881 and 0.868 in the two sets. The models with CE-MRI performed better than the models without CE-MRI(I vs. II, III vs. IV, V vs. VI).

CONCLUSIONS

The radiomics machine learning models based on CE-MRI showed great accuracy and stability in predicting the risk of CLNM for OTSCC patients.

Magnetic resonance imaging accuracy in assessing depth of invasion in tongue squamous cell carcinomas and predicting cervical nodal metastasis

OBJECTIVE

The aim of this study was to evaluate magnetic resonance imaging (MRI) accuracy in assessing the depth of invasion (DOI) compared to pathological DOI in oral tongue squamous cell carcinoma (SCC) and to determine whether MRI-measured DOI can predict lymph node metastasis in the cervical region.

PATIENTS AND METHODS

This retrospective study comprised 36 patients diagnosed with oral tongue SCC who underwent head and neck MRI 1-30 days before surgery and were surgically treated at King Fahad Medical City between January 2017 and November 2022. Relevant information was collected from the patients' records, and the data were analyzed to determine the radiological-histopathological correlations for the DOI and ascertain the cutoff point for nodal metastasis.

RESULTS

A value for Pearson's correlation coefficient between MRI-measured and pathological DOI was 0.86, indicating that these measures were highly associated and consistent with each other. The MRI-measured DOI coronal view (CV) was slightly overestimated than the pathological DOI by 1.72 mm. The cutoff values for the MRI-measured DOI CV and pathological DOI that indicated nodal metastasis were 7.08 mm and 9.04 mm, respectively.

CONCLUSIONS

Preoperative MRI is a valuable tool to accurately stage oral tongue SCC by measuring the depth of tumor invasion.

Machine learning-based MRI radiomics for assessing the level of tumor infiltrating lymphocytes in oral tongue squamous cell carcinoma: a pilot study

BACKGROUND

To investigate the value of machine learning (ML)-based magnetic resonance imaging (MRI) radiomics in assessing tumor-infiltrating lymphocyte (TIL) levels in patients with oral tongue squamous cell carcinoma (OTSCC).

METHODS

The study included 68 patients with pathologically diagnosed OTSCC (30 with high TILs and 38 with low TILs) who underwent pretreatment MRI. Based on the regions of interest encompassing the entire tumor, a total of 750 radiomics features were extracted from T2-weighted (T2WI) and contrast-enhanced T1-weighted (ceT1WI) imaging. To reduce dimensionality, reproducibility analysis by two radiologists and collinearity analysis were performed. The top six features were selected from each sequence alone, as well as their combination, using the minimum-redundancy maximum-relevance algorithm. Random forest, logistic regression, and support vector machine models were used to predict TIL levels in OTSCC, and 10-fold cross-validation was employed to assess the performance of the classifiers.

RESULTS

Based on the features selected from each sequence alone, the ceT1WI models outperformed the T2WI models, with a maximum area under the curve (AUC) of 0.820 versus 0.754. When combining the two sequences, the optimal features consisted of one T2WI and five ceT1WI features, all of which exhibited significant differences between patients with low and high TILs (all P < 0.05). The logistic regression model constructed using these features demonstrated the best predictive performance, with an AUC of 0.846 and an accuracy of 80.9%.

CONCLUSIONS

ML-based T2WI and ceT1WI radiomics can serve as valuable tools for determining the level of TILs in patients with OTSCC.

Ultrasonography use for tongue cancer management: A scoping review

BACKGROUND

Tongue cancer is associated with debilitating diseases and poor prognostic outcomes. The use of imaging techniques like ultrasonography to assist in the clinical management of affected patients is desirable, but its reliability remains debatable. Therefore, the aim of this study is to investigate the importance of ultrasound use for the clinicopathological management of tongue cancer.

METHODS

A scoping review was carried out using specific search strategies in the following electronic databases: PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar. Collected data included bibliographical information, study design, ultrasound equipment, the aim of the ultrasonography use, the timing of ultrasound use during oncological treatment (pre-, trans-, and/or post-operatively), and the advantages and disadvantages of the use of the ultrasound.

RESULTS

A total of 47 studies were included in this review after following the selection process. The majority of the studies investigated the use of ultrasound pre-operatively for the investigation of lymph node metastases or to determine the tumor thickness and depth of invasion. The sensitivity, specificity, and accuracy of ultrasound to determine clinical lymph node metastases ranged from 47% to 87.2%, from 84.3% to 95.8%, and from 70% to 86.2%, respectively. The sensitivity and specificity to determine the microscopic depth of invasion were 92.3% and from 70.6% to 82.1%, respectively.

CONCLUSION

Ultrasonography seems to be a reliable imaging technique for the investigation of important prognostic parameters for tongue cancer, including depth of invasion and lymph node metastases.

Incidental Tongue Cancer Detected With Total-Body 13 N-NH 3 PET/CT

ABSTRACT

A 68-year-old man with chest tightness underwent cardiac blood perfusion imaging on total-body 13 N-NH 3 PET/CT. Incidentally, mildly increased 13 N-NH 3 activity was observed in the left side of the body of the tongue. Pathological diagnosis proved to be mucosal squamous cell carcinoma.

Depth of Invasion in Carcinoma Tongue: Evaluation of Clinical and Imaging Techniques

BACKGROUND

"Depth of invasion" is an additional index incorporated in 8th AJCC staging system for oral cavity squamous cell carcinoma based on its prognostic significance. Pre-operative assessment by clinical palpation and imaging modalities has been used with limitations. The aim of the study is to compare different techniques including clinical palpation, ultrasound, and magnetic resonance imaging with histopathology for assessment of depth of tumor invasion.

MATERIALS

Fifty patients of carcinoma tongue (T1-T3) were enrolled. Clinical palpation, Ultrasound tongue, and Magnetic resonance imaging were used to assess depth of tumor invasion. Microscopic depth of invasion was considered as reference. Statistical analysis was done to assess the level of agreement, reliability, and internal consistency. ROC analysis was done to find the "Area Under Curve" for microscopic depth versus ultrasound, MRI, and gross histopathological "depth of invasion".

RESULTS

Ultrasound tongue showed highest "area under curve", Intra class correlation (ICC:0.786) with a good consistency (Cronbach's Alpha:0.880) with histological reference compared to MRI(ICC:0.689;CA:0.816). Clinical palpation showed weak agreement (Kappa:0.43) for assessing depth. To observe the concordance between ultrasound and microscopic depth, Lin's Concordance Correlation Coefficient (CCC = 0.782) was calculated with 95% limits of agreement. Lin's concordance correlation between ultrasound and microscopic depth showed a good agreement.

CONCLUSIONS

Ultrasound tongue is a reliable imaging modality for pre-operative T staging by assessing tumor "depth of invasion" in carcinoma tongue patients with good internal consistency as per 8th AJCC staging system.

LEVEL OF EVIDENCE

2 (CEBM-Level of Evidence-2.1) Laryngoscope, 134:215-221, 2024.

Long-term control of lung metastasis of tongue cancer treated effectively with cetuximab combination chemotherapy

Generally, systemic chemotherapy is indicated for oral squamous cell carcinoma with distant metastasis and has a poor prognosis. Recently, the advent of molecular targeted drugs, such as cetuximab and immune checkpoint inhibitors, has dramatically improved prognosis, though controlling distant metastasis remains challenging. We report a case of tongue cancer in which lung metastases disappeared in the long term. A 60-year-old Japanese male with squamous cell carcinoma of the tongue underwent preoperative chemoradiotherapy and surgery including subtotal glossectomy, bilateral modified radical neck dissection, and immediate reconstruction with an anterolateral thigh flap. One month after surgery, multiple nodules less than 10 mm in diameter appeared in both lungs on CT imaging. Multiple lung metastases were diagnosed with no local recurrence or regional lymph node metastasis. The patient continues to receive a 4-week treatment course of chemotherapy that included cetuximab every 3 months and the lung metastases were markedly reduced in size or had disappeared. No local recurrence or newly emerged metastases were observed. The patient has been doing well for nine years since the appearance of the lung metastases.

Efficacy of radiological depth of invasion measurements on magnetic resonance images acquired at different magnetic field strengths and imaging sequences in predicting cervical lymph node metastasis and other outcomes in tongue cancer

OBJECTIVES

We investigated the correlation between magnetic resonance imaging (MRI) parameters and tumor pathological depth of invasion (pDOI), between pDOI and radiological DOI (rDOI), between rDOI and duration between biopsy and MRI, and between rDOI and duration between MRI and surgery to determine the efficacy of rDOI in identifying small lesions and other conditions.

STUDY DESIGN

We examined 36 adult patients who had been diagnosed histopathologically with cancer of the tongue and had undergone a glossectomy. Using 1.5 Tesla (T) and 3.0T MRI, we measured rDOI at the deepest infiltration point on 4 MRI sequences. We calculated the correlations between rDOI and the variables examined by Spearman rho analysis and evaluated the diagnostic performance of rDOI by receiver operating characteristic curve analysis.

RESULTS

Axial T2-weighted images using 1.5T MRI provided the closest approximation of pDOI. Although the correlation between rDOI and pDOI was significant, rDOI showed poor or acceptable discrimination in identifying small lesions and other conditions. There were no significant correlations between rDOI and the time between biopsy and MRI or between MRI and surgery.

CONCLUSIONS

The correlation between rDOI and pDOI is significant, but rDOI is ineffective in predicting malignancy and other conditions. Axial T2-weighted images using 1.5T MRI provide the closest approximation of pDOI.

Radiomics analysis of intraoral ultrasound images for prediction of late cervical lymph node metastasis in patients with tongue cancer

BACKGROUND

We investigated the predictability of late cervical lymph node metastasis using radiomics analysis of ultrasonographic images of tongue cancer.

METHODS

We selected 120 patients with tongue cancer who underwent intraoral ultrasonography, 30 of which had late cervical lymph node metastasis. Radiomics analysis was used to extract and quantify the image features. Bootstrap forest (BF), support vector machine (SVM), and neural tanh boost (NTB) were used as the machine learning models, and receiver operating characteristic curve analysis was conducted to determine diagnostic performance.

RESULTS

The sensitivity, specificity, accuracy, and AUC in the validation group were, respectively, 0.600, 0.967, 0.875, and 0.923 for the BF model; 0.700, 0.967, 0.900, and 0.950 for the SVM model; and 0.900, 0.967, 0.950, and 0.967 for NTB model.

CONCLUSIONS

Radiomics analysis and machine learning models using ultrasonographic images of pretreated tongue cancer could predict late cervical lymph node metastasis with high accuracy.

Prognostic significance of the harmonized maximum standardized uptake value of 18F-FDG-PET/CT in patients with resectable oral tongue squamous cell carcinoma: a multicenter study

OBJECTIVES

To investigate the usefulness of harmonized 18F-FDG-PET/CT parameters for predicting the postoperative recurrence and prognosis of oral tongue squamous cell carcinoma (OTSCC).

METHODS

We retrospectively analyzed the cases of 107 OTSCC patients who underwent surgical resection at four institutions in Japan in 2010-2016 and evaluated the harmonized PET parameters of the maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) for the primary tumor as the pSUVmax, pMTV, and pTLG. For lymph node metastasis, we used harmonized PET parameters of nodal-SUVmax, nodal-total MTV (tMTV), and nodal-total TLG (tTLG). The associations between the harmonized PET parameters and the patients' relapse-free survival (RFS) and overall survival (OS) were evaluated by the Kaplan-Meier method and Cox proportional hazard regression analysis for model 1 (preoperative stage) and model 2 (preoperative  +  postoperative stages).

RESULTS

The harmonized SUVmax values were significantly lower than those before harmonization (p=0.012). The pSUVmax was revealed as a significant preoperative risk factor for RFS and OS. Nodal-SUVmax, nodal-tMTV, and nodal-tTLG were significant preoperative risk factors for OS. The combination of pSUVmax + nodal-SUVmax significantly stratified the patients into a low-risk group (pSUVmax <3.97 + nodal-SUVmax <2.85 or ≥2.85) and a high-risk group (pSUVmax ≥3.97 + nodal-SUVmax <2.85 or pSUVmax ≥3.97 + nodal-SUVmax ≥2.85) for recurrence and prognosis (RFS: p=0.001; OS: p<0.001).

CONCLUSIONS

The harmonized pSUVmax is a significant prognostic factor for the survival of OTSCC patients. The combination of pSUVmax and nodal-SUVmax identified OTSCC patients at high risk for recurrence and poor prognosis at the preoperative stage.

Imaging nodal volume and survival in oral tongue cancer with cervical lymph node metastasis

OBJECTIVES

Oral tongue cancer is the most prevalent type of oral cavity cancer and presents the worst prognosis. With the use of TNM staging system, only the size of primary tumor and lymph node are considered. However, several studies have considered the primary tumor volume as a possible significant prognostic factor. Our study, therefore, aimed to explore the role of nodal volume from imaging as a prognostic implication.

METHODS

Medical records and imaging (either from Computed Tomography (CT) or Magnetic Resonance Imaging (MRI) scan) of 70 patients diagnosed with oral tongue cancer with cervical lymph node metastasis between January 2011 and December 2016 were retrospectively reviewed. The pathological lymph node was identified, and nodal volume was measured using the Eclipse radiotherapy planning system and was further analysed for its prognostic implications, particularly on overall survival, disease-free survival, and distant metastasis-free survival.

RESULTS

From A Receiver Operating Characteristic (ROC) curve analysis, the optimal cut-off value of the nodal volume was 3.95 cm3, to predict the disease prognosis, in terms of overall survival and metastatic-free survival (p ≤ 0.001 and p = 0.005, respectively), but not the disease-free survival (p = 0.241). For the multivariable analysis, the nodal volume, but not TNM staging, was a significant prognostic factor for distant metastasis.

CONCLUSIONS

In patients with oral tongue cancer and cervical lymph node metastasis, the presence of an imaging nodal volume of ≥3.95 cm3 was a poor prognostic factor for distant metastasis. Therefore, the lymph node volume may have a potential role to adjunct with the current staging system to predict the disease prognosis.

LEVEL OF EVIDENCE

2b.

Intraoral Ultrasound in Early Stage Tongue Carcinoma

Preoperative assessment of depth of invasion (DOI) is critical in the surgical management of early stage carcinoma tongue. Intraoral ultrasound (IOUS) has been recently described as a cost-effective alternative to MRI in the accurate measurement of DOI. We describe the technique of IOUS in different types of lesions (surface, ulcerative, and exophytic); and provide key imaging pearls for the routine use of this novel ultrasound application.

Intra-operative intaoral ultrasonograpy ensures free deep resection margins in carcinoma tongue: A pilot study

ABSTRACT

Resection of primary in carcinoma tongue is challenging as the tumor often spreads submucosal and deep in muscles. There are various Intraoperative tools available to guide tumor resection margins. We studied the use of intraoperative ultrasound-guided resection for obtaining free deep resection margins in cases of oral carcinoma tongue. To assess the feasibility of resection of tongue tumor using intraoral ultrasound intraoperatively. 5 patients of oral tongue squamous cell carcinoma were included in this prospective pilot study. Intraoral ultrasound-guided resection of the primary tongue tumor was done. The surgeon moved the probe over the tumor and focussed on the point of maximum invasion by the tumor. A 26 G needle was passed in the tongue at a point that was 10 mm away from the deepest point of infiltration. It marked the deepest plane of resection. Following the above marking, a resection of tumor was done. Deep surgical resection margin was found to be >5 mm (free) in all the USG-guided surgically resected primary tongue tumors on histopathological examination. We have found intraoperative ultrasonography a useful tool for the surgical resection of primary tongue tumours.

Prediction of Prognosis of Tongue Squamous Cell Carcinoma Based on Clinical MR Imaging Data Modeling

Objective: Tongue squamous cell carcinoma (TSCC) is one of the most common and poor prognosis head and neck tumors. The purpose of this study is to establish a model for predicting TSCC prognosis based on clinical and MR radiomics data and to develop a nomogram. Methods: A retrospective analysis was performed on the clinical and imaging data of 211 patients with pathologically confirmed TSCC who underwent radical surgery at xx hospital from February 2011 to January 2020. Patients were divided into a study group (recurrence, metastasis, and death, n = 76) and a control group (normal survival, n = 135) according to 1 to 6 years of follow-up. A training set and a test set were established based on a ratio of 7:3 and a time point. In the training set, 3 prediction models (clinical data model, imaging model, and combined model) were established based on the MR radiomics score (Radscore) combined with clinical features. The predictive performance of these models was compared using the Delong curve, and the clinical net benefit of the model was tested using the decision curve. Then, the external validation of the model was performed in the test set, and a nomogram for predicting TSCC prognosis was developed. Results: Univariate analysis confirmed that betel nut consumption, spicy hot pot or pickled food, unclean oral sex, drug use, platelet/lymphocyte ratio (PLR), neutrophil/lymphocyte ratio (NLR), depth of invasion (DOI), low differentiation, clinical stage, and Radscore were factors that affected TSCC prognosis (P < .05). In the test set, the combined model based on these factors had the highest predictive performance for TSCC prognosis (area under curve (AUC) AUC: 0.870, 95% CI [0.761-0.942]), which was significantly higher than the clinical model (AUC: 0.730, 95% CI [0.602-0.835], P = .033) and imaging model (AUC: 0.765, 95% CI [0.640-0.863], P = .074). The decision curve also confirmed the higher clinical net benefit of the combined model, and these results were validated in the test set. The nomogram developed based on the combined model received good evaluation in clinical application. Conclusion: MR-LASSO extracted texture parameters can help improve the performance of TSCC prognosis models. The combined model and nomogram provide support for postoperative clinical treatment management of TSCC.

Application value of indocyanine green fluorescence in sentinel lymph node biopsy for early-stage tongue cancer and oropharyngeal cancer

OBJECTIVES

Occult cervical lymph node metastasis is the most important reason for recurrence of early-stage tongue cancer and oropharyngeal cancer. Cervical sentinel lymph node (SLN) biopsy may help to identify them. Pigment dyes and radionuclide were used to label SLN. Both of them had shortage. This study aims to investigate the application and clinical value of indocyanine green fluorescence imaging in cervical SLN biopsy for patients with early-stage tongue cancer and oropharyngeal cancer.

METHODS

Retrospective analysis was conducted on 23 patients with early tongue cancer and oropharyngeal cancer, who received surgical treatment and used indocyanine green as a tracer to find SLN in Hunan Cancer Hospital from April to October 2021. The detection rate of SLN was calculated and the distribution of SLN in different regions of the neck was analyzed.

RESULTS

SLN was successfully identified in 22 of 23 patients, with a detection rate of 95.65%. Among these 22 patients, 3 patients were found to have cancer metastasis, and the rate of occult lymph node metastasis was 13.63%. No pathologically positive lymph nodes were detected in SLN-negative patients, and thus the positive predictive rate was 100%. For patients with primary lesions located in the anterior 2/3 of the tongue, the constituent ratios of SLN in neck area I, II, III, and IV were 15.15%, 71.72%, 13.13%, and 0, respectively. For patients with primary lesions located in base of the tongue, the constituent ratios of SLN in neck area I, II, III, and IV were 0, 44.44%, 44.44%, and 11.12%, respectively.

CONCLUSIONS

Indocyanine green fluorescence imaging has a high detection rate with accurate positive prediction in the anterior cervical SLN biopsy in patients with early-stage tongue cancer and oropharyngeal cancer. Meanwhile, it can also reflect the lymphatic drainage of tumors located at different primary sites, which has high clinical value.

A Prognostic Model Generated from an Apparent Diffusion Coefficient Ratio Reliably Predicts the Outcomes of Oral Tongue Squamous Cell Carcinoma

This study aimed to develop an apparent diffusion coefficient (ADC) ratio-based prognostic model to predict the recurrence and disease-free survival (DFS) of oral tongue squamous cell carcinoma (OTSCC). A total of 188 patients with cT1-2 oral tongue squamous cell carcinoma were enrolled retrospectively. Clinical and laboratory data were extracted from medical records. The ADC values were measured at the regions of interest of the tumor and non-tumor tissues of the MRI images, and the ADC ratio was used for comparison between the patient with recurrence (n = 83 case, 44%) and patients without recurrence (n = 105 cases, 56%). Cox proportional hazards models were generated to analyze the risk factors of cancer recurrence. A nomogram was developed based on significant risk factors to predict 1-, 5- and 10-year DFS. The receiver operator characteristic (ROC) curves of predictors in the multivariable Cox proportional hazards prognostic model were generated to predict the recurrence and DFS. The integrated areas under the ROC curve were calculated to evaluate discrimination of the models. The ADC ratio, tumor thickness and lymph node ratio were reliable predictors in the final prognostic model. The final model had a 71.1% sensitivity and an 81.0% specificity. ADC ratio was the strongest predictor of cancer recurrence in prognostic performance. Discrimination and calibration statistics were satisfactory with C-index above 0.7 for both model development and internal validation. The calibration curve showed that the 5- and 10-year DFS predicted by the nomogram agreed with actual observations.

High-Frequency Intraoral Ultrasound for Preoperative Assessment of Depth of Invasion for Early Tongue Squamous Cell Carcinoma: Radiological-Pathological Correlations

The eighth edition of the TNM classification officially introduced "depth of invasion" (DOI) as a criterion for determining the T stage in tongue squamous cell carcinoma. The DOI is a well-known independent risk factor for nodal metastases. In fact, several experts strongly suggest elective neck dissection for tongue cancer with a DOI > 4 mm due to the high risk of early and occult nodal metastases. Imaging plays a pivotal role in preoperative assessments of the DOI and, hence, in planning the surgical approach. Intraoral ultrasound (IOUS) has been proposed for early-stage SCC of the oral tongue as an alternative to magnetic resonance imaging (MRI) for local staging. The aim of this work is to investigate the accuracy of IOUS in the assessment of the DOI in early oral SCC (CIS, pT1, and pT2). A total of 41 patients with tongue SCCs (CIS-T2) underwent a preoperative high-frequency IOUS. An IOUS was performed using a small-size, high-frequency hockey-stick linear probe. The ultrasonographic DOI (usDOI) was retrospectively compared to the pathological DOI (pDOI) as the standard reference. In patients who underwent a preoperative MRI, their usDOI, magnetic resonance DOI (mriDOI), and pDOI were compared. Specificity and sensitivity for the IOUS to predict a pDOI > 4 mm and to differentiate invasive and noninvasive tumors were also evaluated. A high correlation was found between the pDOI and usDOI, pDOI and mriDOI, and usDOI and mriDOI (Spearman's ρ = 0.84, p < 0.0001, Spearman's ρ = 0.79, p < 0.0001, and Spearman's ρ = 0.91, p < 0.0001, respectively). A Bland-Altman plot showed a high agreement between the usDOI and pDOI, even though a mean systematic error was found between the usDOI and pDOI (0.7 mm), mriDOI and pDOI (1.6 mm), and usDOI and mriDOI (-0.7 mm). The IOUS was accurate at determining the T stage (p < 0.0001). The sensitivity and specificity for the IOUS to predict a pDOI ≥4 mm were 92.31% and 82.14%, respectively, with an AUC of 0.87 (p < 0.0001). The specificity, sensitivity, negative predictive value (NPV), and positive predictive value (PPV) for the IOUS to predict an invasive cancer were 100%, 94.7%, 60%, and 100%, respectively. The AUC was 0.8 (95% CI 0.646-0.908, p < 0.0001). The IOUS was accurate in a preoperative assessment of a pDOI and T stage, and can be proposed as an alternative to MRI in the preoperative staging of tongue SCC.

Post-Treatment Neck Dissection of Tonsillar and Base of Tongue Squamous Cell Carcinoma in the Era of PET-CT, HPV, and p16

Human-papillomavirus (HPV)-positive tonsillar and base of tongue carcinomas (TSCC/BOTSCC) are rising in incidence and treatments with radiotherapy, chemoradiotherapy (RT/CRT), and neck dissections (NDs) have several side effects. Therefore, an improved selection of patients needing salvage NDs would be beneficial. We examined the prevalence and localisations of viable tumour cells in neck lymph nodes in patients post-RT/CRT, identified by fluorodeoxyglucose positron-emission tomography with computer-tomography (FDG PET-CT), with a focus on HPV-associated tumours. Patients with 217 TSCC/BOTSCC with tumours assessed for HPV-DNA and p16^INK4a undergoing FDG PET-CT 12 weeks after treatment and/or an ND were included. The FDG PET-CT data were compared with the findings in the pathology report after the ND. In total, 36/217 (17%) patients were selected for an ND due to positive findings in post-treatment FDG PET-CT. Of these, 35/36 were HPV-associated, 10/36 (28%) had viable tumour cells in the pathology reports of the neck specimen, and 8/10 (80%) were consistent with the FDG PET-CT findings, while 2/36 (5%) were missed by FDG PET-CT. We conclude that FDG PET-CT 12 weeks after RT/CRT is useful, but not completely reliable for finding all the metastases of HPV-associated TSCC/BOTSCC. Nonetheless, our data indicate that an ND could be more selectively guided by FDG PET-CT.

Near-infrared aggregation-induced emission nanodots for early diagnosis of tongue squamous cell carcinoma and sentinel lymph nodes mapping

Fluorescence imaging has been widely used in the biomedical field owing to its merits of high sensitivity, superb accuracy, high biosafety, etc. However, despite the well performance of fluorescent materials...

Comparison of Whiskbroom and Pushbroom darkfield elastic light scattering spectroscopic imaging for head and neck cancer identification in a mouse model

The early detection of head and neck cancer is a prolonged challenging task. It requires a precise and accurate identification of tissue alterations as well as a distinct discrimination of cancerous from healthy tissue areas. A novel approach for this purpose uses microspectroscopic techniques with special focus on hyperspectral imaging (HSI) methods. Our proof-of-principle study presents the implementation and application of darkfield elastic light scattering spectroscopy (DF ELSS) as a non-destructive, high-resolution, and fast imaging modality to distinguish lingual healthy from altered tissue regions in a mouse model. The main aspect of our study deals with the comparison of two varying HSI detection principles, which are a point-by-point and line scanning imaging, and whether one might be more appropriate in differentiating several tissue types. Statistical models are formed by deploying a principal component analysis (PCA) with the Bayesian discriminant analysis (DA) on the elastic light scattering (ELS) spectra. Overall accuracy, sensitivity, and precision values of 98% are achieved for both models whereas the overall specificity results in 99%. An additional classification of model-unknown ELS spectra is performed. The predictions are verified with histopathological evaluations of identical HE-stained tissue areas to prove the model’s capability of tissue distinction. In the context of our proof-of-principle study, we assess the Pushbroom PCA-DA model to be more suitable for tissue type differentiations and thus tissue classification. In addition to the HE-examination in head and neck cancer diagnosis, the usage of HSI-based statistical models might be conceivable in a daily clinical routine.

Comparison between computed tomography and ultrasound for presurgical evaluation of oral tongue squamous cell carcinoma tumor thickness

PURPOSE

To compare the accuracy of oral tongue squamous cell carcinoma (OTSCC) tumor thickness (TT) measured on CT to intraoperative ultrasound (US) and histopathology.

METHODS AND MATERIALS

Twenty-six patients with OTSCC who underwent tumor resection by a single surgeon with simultaneous intraoperative US between 3/2016 and 4/2019 were prospectively identified, and their data reviewed. TT was independently measured in 19 patients who underwent preoperative CT (cTT) by two neuroradiologists blinded to US and histological results. The confidence level of interpretation of cTT was recorded by each reader using a 5-point Likert scale. The degree of dental artifact on CT was also scored. cTT was compared to TT measured on intraoperative US (uTT) and histopathologic assessment of TT (hTT).

RESULTS

OTSCC was visualized on CT in 52% (10/19) and 63% (12/19) of cases for readers 1 and 2, respectively. Mean Likert score was 0.42 for reader 1 and 0.73 for reader 2. Mean cTT of OTSCCs was 5.8 mm +/- 1.7 mm (n = 11). In comparison, mean uTT and hTT were 7.6 mm±3.5 mm and 7.1 +/- 4.2 mm, respectively. The Pearson coefficient (95% confidence interval) was 0.10 (-0.53-0.66) between cTT and hTT (n = 11) and 0.93 (0.74-0.98) between uTT and hTT.

CONCLUSIONS

Preoperative CT is not reliable for assessment of TT in OTSCC compared to US and histopathology, particularly for OTSCC under 10 mm. US offers a practical complementary imaging tool with a unique role for primary tumor assessment that can aid in pre-operative planning, especially for small tumors.

The utility of intra-oral ultrasound in improving deep margin clearance of oral tongue cancer resections

OBJECTIVES

To investigate the potential utility of intra-oral ultrasound (IOUS) in guiding deep margin clearance and measuring depth of invasion (DOI) of oral tongue carcinomas (OTC).

MATERIALS AND METHODS

Retrospective chart review of consecutive patients with T1-T3 OTC who underwent intraoperative ultrasound-guided resection and a comparator group that had undergone resection without the use of IOUS both by a single surgeon. Data was extracted from operative, pathology and radiology reports. Deep margins and DOI were reviewed by a dedicated head and neck pathologist. Correlation between histologic and ultrasound DOI was assessed using Pearson correlation.

RESULTS

A total of 23 patients were included in the study cohort with a comparator group of 21 patients in the control group. None of the patients in the study cohort had a positive (cut-through) deep margin and the mean deep margin clearance was 8.5 ± 4.9 and 6.7 ± 3.8 for the IOUS and non-IOUS groups respectively (p-value 0.18) showing a non-significant improvement in the IOUS group. As a secondary outcome, there was a strong correlation between histologic and ultrasound DOI (0.9449).

CONCLUSION

Ultrasound appears to be a potentially effective tool in guiding OTC resections. In this small series, IOUS facilitated deep margin clearance and resulted in a non-statistically significant increase in deep margin clearance. Intraoral ultrasound can accurately measure lesional DOI.

Subtraction iodine imaging with area detector CT to improve tumor delineation and measurability of tumor size and depth of invasion in tongue squamous cell carcinoma

Purpose

Tumor size and depth of invasion (DOI) are mandatory assessments for tumor classification in tongue cancer but are often non-assessable on CT due to dental artifacts. This study investigated whether subtraction iodine imaging (SII) would improve tumor delineation and measurability.

Materials and methods

Fifty-seven consecutive patients with tongue cancer, who underwent scanning with a 320-row area detector CT with contrast administration and were treated with surgical resection, were retrospectively evaluated. CT was reconstructed with single-energy projection-based metallic artifact reduction (sCT). SII was generated by subtracting the pre-contrast volume scans from the post-contrast volume scans using a high-resolution deformable registration algorithm. MRI scans were also evaluated for comparing the ability of measurements. Two radiologists visually graded the tumor delineation using a 5-point scale. Tumor size and DOI were measured wherever possible. The tumor delineation score was compared using the Wilcoxon signed-rank method. Spearman’s correlations between imaging and pathological measurements were calculated. Intraclass correlation coefficients of measurements between readers were estimated.

Results

The tumor delineation score was greater on sCT-plus-SII than on sCT alone (medians: 3 and 1, respectively; p < 0.001), with higher number of detectable cases observed with sCT-plus-SII (36/57 [63.2%]) than sCT alone (21/57 [36.8%]). Tumor size and DOI measurability were higher with sCT-plus-SII (29/57 [50.9%]) than with sCT alone (17/57 [29.8%]). MRI had the highest detectability (52/57 [91.2%]) and measurability (46/57 [80.7%]). Correlation coefficients between radiological and pathological tumor size and DOI were similar for sCT (0.83–0.88), sCT-plus-SII (0.78–0.84), and MRI (0.78–0.90). Intraclass correlation coefficients were higher than 0.95 for each modality.

Conclusions

SII improves detectability and measurability of tumor size and DOI in patients with oral tongue squamous cell carcinoma, thus increasing the diagnostic potential. SII may also be beneficial for cases unevaluable on MRI due to artifacts or for patients with contraindications to MRI.

The accuracy of magnetic resonance imaging to measure the depth of invasion in oral tongue cancer: a systematic review and meta-analysis

The accuracy of magnetic resonance imaging (MRI)-derived depth of invasion (DOI) compared to histopathological DOI is still controversial. A meta-analysis was performed to address this controversy and further investigate the best imaging sequence to measure DOI of tongue squamous cell carcinomas (SCC). A comprehensive literature search of five electronic databases was conducted. Stata/SE was used to establish a continuous variable model to assess the consistency between MRI-derived DOI and histopathological DOI. IBM SPSS Statistics 22.0 was used to evaluate the correlation between MRI-derived DOI and histopathological DOI. The meta-analysis showed that the weighted mean difference (WMD) of DOI measured by MRI had an acceptable overestimation compared with that measured by histopathology (WMD 1.64 mm; P < 0.001). In the subgroup analyses, there was no difference between T1-weighted imaging (T1WI) and histopathological values (WMD 0.77 mm; P = 0.273), while T2-weighted imaging (T2WI) had a major overestimation (WMD 2.09 mm; P < 0.001). The overall inter-class correlation coefficient (ICC) between MRI-derived DOI and histopathological DOI was 0.869 (95% CI 0.837-0.895), and was 0.923 (95% CI 0.894-0.944) in the T1WI subgroup and 0.790 (95% CI 0.718-0.845) in the T2WI subgroup. MRI is an accurate modality for evaluating the DOI in oral tongue SCC, and T1WI showed relatively higher validity than T2WI for DOI measurements.

Correlation Of Preoperative Volume Of Oral Tongue Squamous Cell Carcinoma (SCC) On CT Scan With Postsurgical Tumour Size

BACKGROUND

Computed tomography (CT) radiological investigation is commonly used worldwide as a reliable method for determining the existence of oral tongue squamous cell carcinoma. This research was carried out to determine the correlation between the pre-surgical results of the CT scan and the size of the post-surgical tumour.

METHODS

An analytical crosssectional study was conducted at the radiology department of Jinnah Post Graduate Medical Centre (JPMC), Karachi, Pakistan from May to October 2020. All patients aged 18 years or above of either gender having malignant tumours of the tongue already proven on biopsy were consecutively enrolled. Pre-surgery CT scan images were performed. The preoperative volume on CT scan then correlated with the post-surgical tumour size findings. The tumour thickness level of anterio-posterior (AP), transverse dimension (TS) and craniocaudal (CC) on pre-surgery CT scan and post-surgery tumour size were the outcome variables.

RESULTS

Of 84 patients, the mean age of the patients was 48.38 ±11.40 years. There were 56 (66.7%) males and 28 (33.3%) females. A moderate positive correlation of AP tumour size on CT scan was observed with post-surgical AP tumour size (r=0.671, p-value <0.001). Similarly, a moderate positive correlation of TS tumour size was observed with post-operative TS tumour size (r=0.692, p-value <0.001). While, a strong positive correlation of CC tumour size was observed with post-operative surgical tumour size (r=0.787, p-value <0.001).

CONCLUSIONS

This study has reported a strong positive correlation of tumour thickness level on CT scan with post-surgical findings for determining the oral tongue squamous cell carcinoma.

Effect of postoperative radiotherapy for free flap volume changing after tongue reconstruction

OBJECTIVE

The purpose of this study was to evaluate the effect of postoperative radiotherapy (RT) on temporal volume changes with a cutaneous free flap (CF) and a myocutaneous free flap (MCF).

METHODS

The subjects were 24 men and 11 women (mean age, 50.5 ± 15.5 years) with tongue or floor of mouth cancer. Twenty-seven cases of CF and eight cases of MCF were selected. The flap volume change of the reconstructed tongue was calculated using computed tomography (CT) images taken immediately and at one year postoperatively using the DICOM image processing software OsiriX®.

RESULTS

The reduction rate in flap volume at one year postoperatively was 82.0 ± 15% in CF without RT, 70.3 ± 26.1% in CF with RT, 88.5 ± 14.7% in MCF without RT, and 99.5 ± 16% in MCF with RT. The MCF volume was significantly higher compared to the CF volume. Although postoperative RT reduced the CF volume by 30%, there was only a slight reduction in the MCF volume.

CONCLUSIONS

We evaluated the effect of postoperative RT on volume reduction in 35 cases of the reconstructed tongue with CF and MCF using a computer-assisted volume rendering technique. In this study, the effect of RT on volume reduction was different between the CF and MCF.

Complexity of vocal tract shaping in glossectomy patients and typical speakers: A principal component analysis

The glossectomy procedure, involving surgical resection of cancerous lingual tissue, has long been observed to affect speech production. This study aims to quantitatively index and compare complexity of vocal tract shaping due to lingual movement in individuals who have undergone glossectomy and typical speakers using real-time magnetic resonance imaging data and Principal Component Analysis. The data reveal that (i) the type of glossectomy undergone largely predicts the patterns in vocal tract shaping observed, (ii) gross forward and backward motion of the tongue body accounts for more change in vocal tract shaping than do subtler movements of the tongue (e.g., tongue tip constrictions) in patient data, and (iii) fewer vocal tract shaping components are required to account for the patients' speech data than typical speech data, suggesting that the patient data at hand exhibit less complex vocal tract shaping in the midsagittal plane than do the data from the typical speakers observed.

Squamous cell carcinoma of the tongue with cardiac metastasis on 18F-FDG PET/CT: A case report and literature review

INTRODUCTION

The most common malignancies metastasizing to the heart are cancers of the lung, breast, mesothelioma, melanoma, leukemia, and lymphoma. Cardiac metastasis from a tongue cancer is a rare finding and only a few cases have been reported previously in the literature. In this case report and literature review, we discuss the main clinical features of patients with cardiac metastases secondary to a tongue cancer and imaging modalities performed, especially the 18F-Fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT).

PATIENT CONCERNS

This is a case of a 39-year-old woman who in April 2018 was diagnosed with an invasive well differentiated squamous cell carcinoma of the movable tongue. She underwent a left hemiglossectomy followed by a revision of hemiglossectomy and ipsilateral selective neck lymph nodes dissection levels II to III because of pathological margins. An early inoperable clinical recurrence was diagnosed and she received radiochemotherapy with good clinical and metabolic response. She remained asymptomatic thereafter.

DIAGNOSIS

In January 2020, a pre-scheduled 18F-FDG PET/CT showed a diffuse cardiac involvement. In February 2020, a biopsy of the lesion revealed a metastatic squamous cell carcinoma.

INTERVENTIONS

She was deemed to not be a cardiac surgical candidate and treated by palliative chemotherapy: taxol-carboplatin associated with cetuximab then cetuximab alone because of adverse effects. A re-evaluation imaging performed in April 2020 evidenced a progression of the cardiac involvement, which led to switch chemotherapy by immunotherapy with nivolumab.

OUTCOMES

This patient had a very poor prognosis and succumbed to major heart failure 4 months after the diagnosis of cardiac metastasis.

CONCLUSION

In this case report, 18F-FDG PET/CT proved to be useful in detecting cardiac metastasis and changed the therapeutic management of the patient. It suggests that patients with tongue malignancies in a context of poor initial prognosis should be followed-up early by 18F-FDG PET/CT with HFLC diet to facilitate detection of recurrence.

Tongue Tumor Detection in Hyperspectral Images Using Deep Learning Semantic Segmentation

OBJECTIVE

The utilization of hyperspectral imaging (HSI) in real-time tumor segmentation during a surgery have recently received much attention, but it remains a very challenging task.

METHODS

In this work, we propose semantic segmentation methods, and compare them with other relevant deep learning algorithms for tongue tumor segmentation. To the best of our knowledge, this is the first work using deep learning semantic segmentation for tumor detection in HSI data using channel selection, and accounting for more spatial tissue context, and global comparison between the prediction map, and the annotation per sample. Results, and Conclusion: On a clinical data set with tongue squamous cell carcinoma, our best method obtains very strong results of average dice coefficient, and area under the ROC-curve of [Formula: see text], and [Formula: see text], respectively on the original spatial image size. The results show that a very good performance can be achieved even with a limited amount of data. We demonstrate that important information regarding tumor decision is encoded in various channels, but some channel selection, and filtering is beneficial over the full spectra. Moreover, we use both visual (VIS), and near-infrared (NIR) spectrum, rather than commonly used only VIS spectrum; although VIS spectrum is generally of higher significance, we demonstrate NIR spectrum is crucial for tumor capturing in some cases.

SIGNIFICANCE

The HSI technology augmented with accurate deep learning algorithms has a huge potential to be a promising alternative to digital pathology or a doctors' supportive tool in real-time surgeries.

Machine learning-based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma

OBJECTIVES

To develop and compare several machine learning models to predict occult cervical lymph node (LN) metastasis in early-stage oral tongue squamous cell cancer (OTSCC) from preoperative MRI texture features.

MATERIALS AND METHODS

We retrospectively enrolled 116 patients with early-stage OTSCC (cT1-2N0) who had been surgically treated by tumor excision and elective neck dissection (END). For each patient, we extracted 86 texture features from T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (ceT1WI), respectively. Dimension reduction was performed in three consecutive steps: reproducibility analysis, collinearity analysis, and information gain algorithm. Models were created using six machine learning methods, including logistic regression (LR), random forest (RF), naïve Bayes (NB), support vector machine (SVM), AdaBoost, and neural network (NN). Their performance was assessed using tenfold cross-validation.

RESULTS

Occult LN metastasis was pathologically detected in 42.2% (49/116) of the patients. No significant association was identified between node status and patients' gender, age, or clinical T stage. Dimension reduction steps selected 6 texture features. The NB model gave the best overall performance, which correctly classified the nodal status in 74.1% (86/116) of the carcinomas, with an AUC of 0.802.

CONCLUSION

Machine learning-based MRI texture analysis offers a feasible tool for preoperative prediction of occult cervical node metastasis in early-stage OTSCC.

KEY POINTS

• A machine learning-based MRI texture analysis approach was adopted to predict occult cervical node metastasis in early-stage OTSCC with no evidence of node involvement on conventional images. • Six texture features from T2WI and ceT1WI of preoperative MRI were selected to construct the predictive model. • After comparing six machine learning methods, naïve Bayes (NB) achieved the best performance by correctly identifying the node status in 74.1% of the patients, using tenfold cross-validation.

Can MRI-derived depth of invasion predict nodal recurrence in oral tongue cancer?

OBJECTIVES

To evaluate the prognostic value of preoperative radiological findings for nodal recurrence in clinically node-negative (cN0) patients with oral tongue squamous cell carcinoma (SCC).

METHODS

The study population consisted of 52 patients with cT1-2N0 oral tongue SCC classified according to the 7th edition of the Union for International Cancer Control (UICC) staging system. The subjects had undergone preoperative radiological examinations, including magnetic resonance imaging (MRI) and ¹⁸F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/computed tomography. All patients were treated with local resection and watchful waiting for neck management. Using an unpaired t test, Pearson's chi-squared test, and the Kaplan-Meier method, the MRI-derived depth of invasion (DOI), the standardized uptake value (SUV) on FDG-PET, and the T stage according to the 7th and 8th UICC were assessed as prognostic factors.

RESULTS

The MRI-derived DOI was recorded as ≤ 5 mm in 24 patients and > 5 mm in 28 patients. During the follow-up period, nine patients exhibited nodal recurrence, with the MRI-derived DOI being significantly higher in patients with positive than in those with negative (p = 0.011). The SUV was not significant. Five-year cumulative nodal recurrence probabilities were 4.5% for patients with an MRI-derived DOI ≤ 5 mm, while it was 32.1% for > 5 mm (p = 0.013). Although the T classifications were not significant, none of our patients whose T stage according to the 8th UICC was T1 suffered nodal recurrence.

CONCLUSIONS

MRI-derived DOI can predict nodal recurrence, while preoperative information may assist in treatment planning for oral tongue SCC.

Operative tactics in floor of mouth and tongue cancer resection - the importance of imaging and planning

Complete tumour resection (R0 margin) is an axiom of surgical oncology. Oral cancer ablation is challenging, due to anatomical, functional, and aesthetic considerations. R0 margin is strongly linked to better survival outcomes with great variation in the R0 % across units. This is commonly attributed to disease biology. Without disputing the importance of biological characteristics, we contend that image-based anatomical surgical planning has an important role to play in achieving complete resection. Here, we present our approach utilising cross-sectional imaging, anatomical characteristics and spatial awareness in planning resections for floor of mouth (FOM) and oral tongue cancers. We highlight the challenge of controlling the deep tumour margin lingual to mandible due to anterior vector constraints and emphasise the importance of resecting the genial muscles in a planned fashion and that any rim resection should be obliquely sagittal. In resecting lateral FOM tumours, assessing extension to the parapharyngeal fat is crucial; and mandibular rim resection at a sagittal plane below the mylohyoid line is often required. Assessing the proximity of the contralateral neurovascular pedicle, pre-epiglottic space and hyoid bone are crucial parameters to determine the extent of tongue tumour resection. Our cohort included 173 patients with FOM SCC and 299 patients with tongue SCC. Six patients (3.5%) from the FOM group and eight patients (3%) from the tongue group had involved (R1) margins following surgery. This was associated with local relapse (p<0.05). In conclusion, we demonstrate that image-based planning can aid achieving R0 resections and reduce disease relapse.

Value of preoperative computed tomography in predicting the local staging of cancer of the tongue at primary surgery

Preoperative staging is essential for the planning of treatment of cancer. This study was designed to evaluate the accuracy of computed tomography (CT) in predicting the local stage of tongue cancer by comparing it with the gold standard of histopathology. A total of 233 patients with newly-diagnosed tongue cancer was retrospectively reviewed, and the size of the tumour and the status of the cervical lymph node were compared between CT images and histopathological results. Patients with stage II cancer were followed up to assess the influence of inaccurate preoperative staging on prognosis. The accuracy of local staging by CT was 47.6% (111/233), with 59.7% (139/233) for tumour stage, and 70.4% (164/233) for nodal stage. The greatest dimension of the tumour on the CT image was about 2mm less than that measured by histopathology. The estimated volume of tumour was a quarter smaller. The accuracy of predicting malignant lymph nodes by CT was 68.9% (n=161). Among patients with stage II disease, simultaneous neck dissection was less likely in the understaged group than in the accurately staged one. The reoperation rate was a little higher but not significantly so. We conclude that the accuracy of CT in predicting local staging for tongue cancer was only moderate, because it underestimated the size of the tumour and needed to improve the criteria for detecting malignant lymph nodes. Understaging on CT images may influence the prognosis of patients with early stage tongue cancer.

Sonographicdetectionof a metastatic lingual lymph node in a patient with tongue cancer

A 41-year-old male was diagnosed with tongue cancer. Preoperative imaging revealed no metastatic lymph nodes. Intraoperative sonography detected a minute hypoechoic lesion in the sublingual space. A pea-sized mass was found during dissection of that space and histopathology revealed a metastatic lymph node. The lingual lymph node should routinely be evaluated sonographically prior to surgical resection of tongue cancer.

Applications of a deep learning method for anti-aliasing and super-resolution in MRI

Magnetic resonance (MR) images with both high resolutions and high signal-to-noise ratios (SNRs) are desired in many clinical and research applications. However, acquiring such images takes a long time, which is both costly and susceptible to motion artifacts. Acquiring MR images with good in-plane resolution and poor through-plane resolution is a common strategy that saves imaging time, preserves SNR, and provides one viewpoint with good resolution in two directions. Unfortunately, this strategy also creates orthogonal viewpoints that have poor resolution in one direction and, for 2D MR acquisition protocols, also creates aliasing artifacts. A deep learning approach called SMORE that carries out both anti-aliasing and super-resolution on these types of acquisitions using no external atlas or exemplars has been previously reported but not extensively validated. This paper reviews the SMORE algorithm and then demonstrates its performance in four applications with the goal to demonstrate its potential for use in both research and clinical scenarios. It is first shown to improve the visualization of brain white matter lesions in FLAIR images acquired from multiple sclerosis patients. Then it is shown to improve the visualization of scarring in cardiac left ventricular remodeling after myocardial infarction. Third, its performance on multi-view images of the tongue is demonstrated and finally it is shown to improve performance in parcellation of the brain ventricular system. Both visual and selected quantitative metrics of resolution enhancement are demonstrated.

Detection of Recurrence in Retromolar Trigone using "Puffed Cheek" F-18 FDG PET-CT Scan

Fluorine-18 Fluorodeoxyglucose positron emission tomography computed tomography (18FDG PET-CT) is routinely used for post treatment monitoring of oral cavity cancers. In addition to physiological uptake of 18FDG, apposition of anatomical structures poses a big challenge in accurate localization of pathology in oral cavity that is further complicated by surgery and/or radiotherapy related distortion. Case of recurrent Squamous Cell Carcinoma (SCC) of tongue is being presentedin a patient who underwent 18FDG PET-CT for post treatment monitoring. Simple 'Puffed Cheek' maneuver helped in accurately localizing tumour recurrence site within the retro-molar trigone.

Reliability of MRI-Derived Depth of Invasion of Oral Tongue Cancer

RATIONALE AND OBJECTIVE

To evaluate the inter-rater reliability of the magnetic resonance imaging (MRI)-derived depth of invasion (DOI) and the agreement between MRI and pathological measurements of oral tongue cancer.

MATERIALS AND METHODS

The institutional review board approved this retrospective study. The study population consisted of 29 patients with clinical T2N0 oral tongue cancer treated by surgery. Routine pretreatment MRI was performed on a 3T superconducting imager. Two raters with 23 and 18 years of head-and-neck MRI experience, respectively, independently chosen MRI sequences for each patient, then delineate the tumor, and then used three protocols to measure the MRI-derived DOI: the axial reconstructed thickness (method 1), the axial invasive portion (method 2), and the coronal invasive portion (method 3). Then they consensually selected the optimal among the three methods for each patient; it was designated method 4. The Bland-Altman plots, intraclass correlation coefficients (ICCs), and the paired samples test were used. According to the median follow-up of 41 months, the relationship between the MRI-derived DOI and nodal recurrence was also investigated.

RESULTS

The inter-rater reliability of methods 2 and 4 was excellent (ICC of 0.829 and 0.807, respectively). The correlation between MRI and pathological measurements was good for method 4 (ICC of 0.611), however, all measurements recorded on MRI were 2-3 mm larger than on pathology. No patients whose MRI-derived DOI was less than 5 mm suffered nodal recurrence.

CONCLUSION

The MRI-derived DOI was valuable for the preoperative staging. The optimal measurement method should be selected on a case-by-case basis.

MRI prognostic factors of tongue cancer: potential predictors of cervical lymph nodes metastases

Background This study aimed to evaluate the efficacy of three MR imaging parameters, which are tumour thickness, para-lingual distance and apparent diffusion coefficient (ADC) value for prediction of cervical lymph nodes metastasis in cancer tongue patients. Patients and methods Fifty patients with proved cancer tongue by histopathological examination underwent MRI examination. T1 and T2- weighted MRI, diffusion-weighted images and post-contrast T1 fat suppression sequences were used. Results The patients were classified according to lymph nodes involvement as seen by MRI into two groups. Significant differences between positive and negative nodes groups were observed regarding tumour thickness and para-lingual distance (p-values = 0.008 and 0.003 respectively). ROC curve analyses revealed cut-off values >13.8 mm and ≤ 3.3 mm for tumour thickness and para-lingual distance respectively for prediction of nodes involvement. No significant differences between patients with and without cervical lymph nodes metastasis were found regarding corresponding ADC value of the tumour (p-value = 0.518). Conclusions Para-lingual distance and tumour thickness are factors that could influence pre-operative judgment and prognosis of tongue cancer patients. ADC value of the tumour itself seem not to be a reliable index of cancer progression to regional lymph nodes.

Initial experience with image-guided surgical navigation in transoral surgery

BACKGROUND

Surgical navigation using image guidance may improve the safety and efficacy of transoral surgery (TOS); however, preoperative imaging cannot be accurately registered to the intraoperative state due to deformations resulting from placement of the laryngoscope or retractor. This proof of concept study explores feasibility and registration accuracy of surgical navigation for TOS by utilizing intraoperative imaging.

METHODS

Four patients undergoing TOS were recruited. Suspension laryngoscopy was performed with a CT-compatible laryngoscope. An intraoperative contrast enhanced CT scan was obtained and registered to fiducials placed on the neck, face, and laryngoscope.

RESULTS

All patients were successfully scanned and registered. Registration accuracy within the pharynx and larynx was 1 mm or less. Target registration was confirmed by localizing endoscopic and surface structures to the CT images. Successful tracking was performed in all 4 patients.

CONCLUSION

For surgical navigation during TOS, although a high level of registration accuracy can be achieved by utilizing intraoperative imaging, significant limitations of the existing technology have been identified. These limitations, as well as areas for future investigation, are discussed.

Detection and delineation of squamous neoplasia with hyperspectral imaging in a mouse model of tongue carcinogenesis

Hyperspectral imaging (HSI) holds the potential for the noninvasive detection of cancers. Oral cancers are often diagnosed at a late stage when treatment is less effective and the mortality and morbidity rates are high. Early detection of oral cancer is, therefore, crucial in order to improve the clinical outcomes. To investigate the potential of HSI as a noninvasive diagnostic tool, an animal study was designed to acquire hyperspectral images of in vivo and ex vivo mouse tongues from a chemically induced tongue carcinogenesis model. A variety of machine-learning algorithms, including discriminant analysis, ensemble learning, and support vector machines, were evaluated for tongue neoplasia detection using HSI and were validated by the reconstructed pathological gold-standard maps. The diagnostic performance of HSI, autofluorescence imaging, and fluorescence imaging were compared in this study. Color-coded prediction maps were generated to display the predicted location and distribution of premalignant and malignant lesions. This study suggests that hyperspectral imaging combined with machine-learning techniques can provide a noninvasive tool for the quantitative detection and delineation of squamous neoplasia.

Sonographic detection of a lingual node metastasis from early squamous cell cancer of the tongue

Lingual node metastasis from early-stage squamous cell carcinoma of the tongue is uncommon and difficult to detect. We present a case of a lingual lymph node metastasis from an early tongue cancer. Sonography demonstrated a small nonpalpable lingual node on the floor of the mouth. The patient underwent complete surgical resection and had no evidence of recurrence 5 years later. © 2017 Wiley Periodicals, Inc. J Clin Ultrasound 46:69-72, 2018.

[Diagnostic value of diffusion-weighted magnetic resonance imaging for solid tumors affecting the base of tongue and glossopharynx]

PURPOSE

The objective of the present study was to evaluate the potential application value of diffusion-weighted magnetic resonance imaging (DW-MRI) to differentiate solid benign from malignant tumors affecting the base of tongue and glossopharynx.

METHODS

Sixty-one patients who presented with solid tumors affecting the base of tongue and glossopharynx underwent conventional MRI and DW-MRI before pathologic verification. Based on pathologic findings, the tumors were classified into 4 groups: Group 1, solid benign tumor (n=10); Group 2, epithelial carcinoma (n=35); Group 3: non- epithelial malignant tumor (n=16); and Group 4: malignant tumor (n=51, a combination of Group 2 and 3). The mean apparent diffusion coefficients (ADCs) were computed from DW-MRI scans obtained with b factors of 0 and 1000 s/mm2. SAS 9.1 software package was used for Wilcoxon test.

RESULTS

The mean ADCs were significantly different (P<0.05) between group 1 and the other 3 groups, and between group 2 and group 3, respectively. Among all kinds of pathological types of lesions, non-Hodgkin lymphoma had a lowest mean ADC value, and schwannoma had a highest mean ADC value. In addition, the mean ADCs among epithelial carcinoma subtypes were not significantly different(P>0.05).

CONCLUSIONS

DW-MRI has differential diagnostic value of solid tumors affecting the base of tongue and glossopharynx, especially in distinguishing between benign solid tumors and malignant tumors, and between epithelial carcinomas and non-epithelial malignant tumors.

Report: Small cell neuroendocrine carcinoma (SCNEC) of the tongue: A case report

Small cell neuroendocrine carcinoma (SCNEC) of the tongue is very rare. We here present a SCNEC impatient with distant metastasis. A 74-year-old Chinese male went to hospital to treat a tongue tumor, which was founded at a conventional physical examination in Weifang Stomatology Hospital. The check of positron emission tomography-computer tomography (PET-CT) by Weifang people's hospital revealed a tumor in the right root of tongue, and distant metastasis in the right submandibular area, neck, mediastinum, right hilar, abdominal, retroperitoneal multiple lymph nodes, left thyroid, right lower lung, right scapula and bilateral adrenal. The patient was diagnosed tongue SCNEC by the pathological analysis of the tissue section. Conforming to the diagnosis of tongue SCNEC, the patient received adjuvant chemotherapy for 6 cycles with etoposide and carboplatin, and is alive now 9 months after the diagnosis.

Locoregional control of tongue base adenoid cystic carcinoma with primary resection and radial forearm free flap reconstruction

Adenoid cystic carcinoma of the minor salivary glands can be challenging and marked by high rates of local recurrence despite appropriate surgical resection. Management of this pathology in the base of the tongue is particularly difficult given the poor functional outcomes traditionally associated with an aggressive surgical approach. This article presents a case series of patients who underwent up-front surgical resection followed by free tissue transfer reconstruction. A retrospective analysis was performed of patients with adenoid cystic carcinoma of the base of the tongue who underwent composite resection and reconstruction with a radial forearm free flap. Three patients met inclusion criteria and underwent analysis. All patients achieved locoregional control after at least 4 years of surveillance. In addition, all patients were decannulated and were swallowing without the need for gastrostomy tube feeding. This series demonstrates that for select patients with adenoid cystic carcinoma of the base of the tongue, excellent locoregional control can be achieved with acceptable functional outcomes and prolonged survival when appropriate reconstructive measures are employed.

Doppler sonography of the deep lingual artery

Purpose: To clarify the Doppler sonographic features of the lingual artery in normal subjects and to evaluate those of patients with cancer of the tongue.

Material and Methods: Sixty-seven volunteers and 12 patients with cancer and/or leukoplakia of the tongue were examined with an intraoral sonographic probe. The visibility of the deep lingual artery was determined on transverse and anteroposterior images. On the transverse images, the vascular index, which was defined as the number of colored pixels, was measured on bilateral lingual arteries. Thereafter, the degree of symmetry was evaluated for normal subjects and patients.

Results: In normal subjects, between younger and older volunteers, there were no significant differences in visibility of the trunk but differences were found between the two groups for the dorsal branches. The vascular indices of the right and left sides were not different. The characteristic Doppler sonographic feature was vasculature in and around the tumors in the patients with cancer of the tongue. The symmetry indices of the cancer patients were significantly different from those of normal subjects.

Conclusion: Doppler sonography should be an important procedure for evaluation of tongue neoplasms.

Quantitative Three-Dimensional Ultrasound Imaging of Partially Resected Tongues

OBJECTIVE: We sought to assess the impact of partial lateral glossectomies on tongue function and speech.

STUDY DESIGN AND SETTING: Tongue shapes of 12 patients with lateral tumors of the tongue were recorded with three-dimensional ultrasound before their surgery and two months after. Twelve normal participants served as controls. Speech acceptability was also assessed.

RESULTS: Principal component analyses demonstrated that the flap reconstructions led to a stiffening of the operated side of the tongue. A concavity index and an asymmetry index demonstrated that the glossectomy patients exhibited decreased midsagittal grooving and increased lingual asymmetry. The change in midsagittal grooving correlated moderately with the decrease in speech acceptability.

CONCLUSION: A lateral partial glossectomy affects the tongue's intrinsic deformation, in particular midsagittal grooving and symmetry. This can have a detrimental effect for speech.

SIGNIFICANCE: A lateral resection affects the tongue's symmetry and midsagittal groove. The change in midsagittal grooving correlates with a decrease in speech acceptability.