Dataset of weekly intra-treatment diffusion weighted imaging in head and neck cancer patients treated with MR-Linac

Radiation therapy (RT) is a crucial treatment for head and neck squamous cell carcinoma (HNSCC); however, it can have adverse effects on patients' long-term function and quality of life. Biomarkers that can predict tumor response to RT are being explored to personalize treatment and improve outcomes. While tissue and blood biomarkers have limitations, imaging biomarkers derived from magnetic resonance imaging (MRI) offer detailed information. The integration of MRI and a linear accelerator in the MR-Linac system allows for MR-guided radiation therapy (MRgRT), offering precise visualization and treatment delivery. This data descriptor offers a valuable repository for weekly intra-treatment diffusion-weighted imaging (DWI) data obtained from head and neck cancer patients. By analyzing the sequential DWI changes and their correlation with treatment response, as well as oncological and survival outcomes, the study provides valuable insights into the clinical implications of DWI in HNSCC.

Anterior Cervical Cystic Lymphangioma in an Adult Patient

N/a.

Prognostic Value of a Combined Nomogram Model Integrating 3-Dimensional Deep Learning and Radiomics for Head and Neck Cancer

OBJECTIVE

The preoperative prediction of the overall survival (OS) status of patients with head and neck cancer (HNC) is significant value for their individualized treatment and prognosis. This study aims to evaluate the impact of adding 3D deep learning features to radiomics models for predicting 5-year OS status.

METHODS

Two hundred twenty cases from The Cancer Imaging Archive public dataset were included in this study; 2212 radiomics features and 304 deep features were extracted from each case. The features were selected by univariate analysis and the least absolute shrinkage and selection operator, and then grouped into a radiomics model containing Positron Emission Tomography /Computed Tomography (PET/CT) radiomics features score, a deep model containing deep features score, and a combined model containing PET/CT radiomics features score +3D deep features score. TumorStage model was also constructed using initial patient tumor node metastasis stage to compare the performance of the combined model. A nomogram was constructed to analyze the influence of deep features on the performance of the model. The 10-fold cross-validation of the average area under the receiver operating characteristic curve and calibration curve were used to evaluate performance, and Shapley Additive exPlanations (SHAP) was developed for interpretation.

RESULTS

The TumorStage model, radiomics model, deep model, and the combined model achieved areas under the receiver operating characteristic curve of 0.604, 0.851, 0.840, and 0.895 on the train set and 0.571, 0.849, 0.832, and 0.900 on the test set. The combined model showed better performance of predicting the 5-year OS status of HNC patients than the radiomics model and deep model. The combined model was shown to provide a favorable fit in calibration curves and be clinically useful in decision curve analysis. SHAP summary plot and SHAP The SHAP summary plot and SHAP force plot visually interpreted the influence of deep features and radiomics features on the model results.

CONCLUSIONS

In predicting 5-year OS status in patients with HNC, 3D deep features could provide richer features for combined model, which showed outperformance compared with the radiomics model and deep model.

Performance evaluation of commercial and non-commercial shear wave elastography implementations for vascular applications

BACKGROUND

Shear wave elastography (SWE) is mainly used for stiffness estimation of large, homogeneous tissues, such as the liver and breasts. However, little is known about its accuracy and applicability in thin (∼0.5-2 mm) vessel walls. To identify possible performance differences among vendors, we quantified differences in measured wave velocities obtained by commercial SWE implementations of various vendors over different imaging depths in a vessel-mimicking phantom. For reference, we measured SWE values in the cylindrical inclusions and homogeneous background of a commercial SWE phantom. Additionally, we compared the accuracy between a research implementation and the commercially available clinical SWE on an Aixplorer ultrasound system in phantoms and in vivo in patients.

METHODS

SWE measurements were performed over varying depths (0-35 mm) using three ultrasound machines with four ultrasound probes in the homogeneous 20 kPa background and cylindrical targets of 10, 40, and 60 kPa of a multi-purpose phantom (CIRS-040GSE) and in the anterior and posterior wall of a homogeneous polyvinyl alcohol vessel-mimicking phantom. These phantom data, along with in vivo SWE data of carotid arteries in 23 patients with a (prior) head and neck neoplasm, were also acquired in the research and clinical mode of the Aixplorer ultrasound machine. Machine-specific estimated phantom stiffness values (CIRS phantom) or wave velocities (vessel phantom) over all depths were visualized, and the relative error to the reference values and inter-frame variability (interquartile range/median) were calculated. Correlations between SWE values and target/vessel wall depth were explored in phantoms and in vivo using Spearman's correlations. Differences in wave velocities between the anterior and posterior arterial wall were assessed with Wilcoxon signed-rank tests. Intra-class correlation coefficients were calculated for a sample of ten patients as a measure of intra- and interobserver reproducibility of SWE analyses in research and clinical mode.

RESULTS

There was a high variability in obtained SWE values among ultrasound machines, probes, and, in some cases, with depth. Compared to the homogeneous CIRS-background, this variation was more pronounced for the inclusions and the vessel-mimicking phantom. Furthermore, higher stiffnesses were generally underestimated. In the vessel-mimicking phantom, anterior wave velocities were (incorrectly) higher than posterior wave velocities (3.4-5.6 m/s versus 2.9-5.9 m/s, p ≤ 0.005 for 3/4 probes) and remarkably correlated with measurement depth for most machines (Spearman's ρ = -0.873-0.969, p < 0.001 for 3/4 probes). In the Aixplorer's research mode, this difference was smaller (3.3-3.9 m/s versus 3.2-3.6 m/s, p = 0.005) and values did not correlate with measurement depth (Spearman's ρ = 0.039-0.659, p ≥ 0.002). In vivo, wave velocities were higher in the posterior than the anterior vessel wall in research (left p = 0.001, right p < 0.001) but not in clinical mode (left: p = 0.114, right: p = 0.483). Yet, wave velocities correlated with vessel wall depth in clinical (Spearman's ρ = 0.574-0.698, p < 0.001) but not in research mode (Spearman's ρ = -0.080-0.466, p ≥ 0.003).

CONCLUSIONS

We observed more variation in SWE values among ultrasound machines and probes in tissue with high stiffness and thin-walled geometry than in low stiffness, homogeneous tissue. Together with a depth-correlation in some machines, where carotid arteries have a fixed location, this calls for caution in interpreting SWE results in clinical practice for vascular applications.

Adding a Small Field of View for Increased Resolution in Contouring: An Effective and Quick Method

Image resolution is paramount when contouring complex anatomy, as in head and neck planning. It is notable that when diagnostic radiologists perform a computed tomography scan of the neck, they always use a small field of view because it gives the best image resolution. When planning for radiation treatment, however, it is also necessary to have a large field of view to provide a comprehensive external contour that allows for radiation dose calculation. We present a simple method to obtain both small and large fields of view at the time of computed tomography simulation.

Anatomical variations in the superficial venous system of the neck: an image-based study using contrast-enhanced computed tomography

PURPOSE

The superficial venous system (SVS) of the neck receives blood from the face and oral cavity. The SVS comprises the anterior jugular vein (AJV), external jugular vein (EJV), and facial vein (FV). Comprehensive knowledge of the normal anatomy and potential variations in the venous system is valuable in surgical and radiological procedures. This study aimed to update the anatomic knowledge of the SVS using a radiographic approach, which is a beneficial data source in clinical practice.

METHODS

Contrast-enhanced computed tomography images of the neck of patients with head and neck cancer treated between 2017 and 2020 were retrospectively evaluated. Each side of the neck was counted separately. A total of 302 necks of 151 patients were enrolled in this study.

RESULTS

The medial AJV was absent in 49.7% (75/151) of the patients on the left side, which was significantly greater than the 19.2% (29/151) on the right (p < 0.001). The left AJV drained into the right venous system in 6.6% (10/151) of the necks. In 48.3% (146/302) of the necks, the FV did not flow into the internal jugular vein but rather into the EJV or AJV; these findings were significantly more frequent than those reported in previous studies. The diameters of the veins were significantly larger when they received blood from the FV than when they were not connected to the FV.

CONCLUSION

These findings indicate that the AJV has a rightward preference during its course. The course of the FV is diverse and affects the diameter of connected veins.

The Diagnostic Performance of Magnetic Resonance Imaging in the Categorization of Pediatric Neck Lymph Nodes: Radiologic and Pathologic Correlations

BACKGROUND/AIM

To present MRI features of neck lymph nodes in benign and malignant conditions in the pediatric population.

MATERIALS AND METHODS

MRIs of the neck of 51 patients 1 to 18 years old (40 boys, 11 girls [10.08±4.73]) with lymph node biopsy were retrospectively analyzed. Those were grouped as benign including reactive (27 [52.9%]) and lymphadenitis (11 [21.6%]), and malignant (13 [25.5%]). The groups were evaluated multiparametrically in terms of quantitative and qualitative variables.

RESULTS

The long axis, short axis, area, and apparent diffusion coefficient (ADC) values of the largest lymph node were 21 (17 to 24) mm, 14 (12 to 18) mm, 228.60 (144.79 to 351.82) mm 2 , 2531 (2457 to 2714) mm 2 /s for reactive, 24 (19 to 27) mm, 15 (11 to 20) mm, 271.80 (231.43 to 412.20) mm 2 , 2534 (2425 to 2594) mm 2 /s for lymphadenitis, 27 (23.50 to 31.50) mm, 20 (15 to 22) mm, 377.08 (260.47 to 530.94) mm 2 , 2337 (2254 to 2466) mm 2 /s for malignant, respectively. Statistical analysis of our data suggests that the following parameters are associated with a higher likelihood of malignancy: long axis >22 mm, short axis >16 mm, area >319 cm 2 , ADC value <2367 mm 2 /s, and supraclavicular location. Perinodal and nodal heterogeneity, posterior cervical triangle location are common in lymphadenitis ( P <0.001). Reactive lymph nodes are distributed symmetrically in both neck halves ( P <0.001).

CONCLUSION

In the MRI-based approach to lymph nodes, not only long axis, short axis, surface area, and ADC, but also location, distribution, perinodal, and nodal heterogeneity should be used.

Predictive value of local control by 4'-[methyl-11C]-thiotymidine PET volume parameters in p16-negative oropharyngeal, hypopharyngeal, and supraglottic squamous cell carcinoma

PURPOSE

We investigated the potential of baseline 4'-[methyl- 11 C]-thiothymidine ([ 11 C]4DST) PET for predicting loco-regional control of head and neck squamous cell carcinoma (HNSCC).

METHODS

A retrospective analysis was performed using volumetric parameters, such as SUVmax, proliferative tumor volume (PTV), and total lesion proliferation (TLP), of pretreatment [ 11 C]4DST PET for 91 patients with HNSCC with primary lesions in the oral cavity, hypopharynx, supraglottis, and oropharynx, which included p16-negative patients. PTV and TLP were calculated for primary lesions and metastatic lymph nodes combined. We examined the association among the parameters and relapse-free survival and whether case selection focused on biological characteristics improved the accuracy of prognosis prediction.

RESULTS

The area under the curves (AUCs) using PTV and TLP were high for the oropharyngeal/hypopharyngeal/supraglottis groups (0.91 and 0.87, respectively), whereas that of SUVmax was 0.66 ( P  < 0.01). On the other hand, the oral group had lower AUCs for PTV and TLP (0.72 and 0.77, respectively). When all cases were examined, the AUCs using PTV and TLP were 0.84 and 0.83, respectively.

CONCLUSION

Baseline [ 11 C]4DST PET/CT volume-based parameters can provide important prognostic information with p16-negative oropharyngeal, hypopharyngeal, and supraglottic cancer patients.

Image denoising and model-independent parameterization for IVIM MRI

Objective. To improve intravoxel incoherent motion imaging (IVIM) magnetic resonance Imaging quality using a new image denoising technique and model-independent parameterization of the signal versusb-value curve.Approach. IVIM images were acquired for 13 head-and-neck patients prior to radiotherapy. Post-radiotherapy scans were also acquired for five of these patients. Images were denoised prior to parameter fitting using neural blind deconvolution, a method of solving the ill-posed mathematical problem of blind deconvolution using neural networks. The signal decay curve was then quantified in terms of several area under the curve (AUC) parameters. Improvements in image quality were assessed using blind image quality metrics, total variation (TV), and the correlations between parameter changes in parotid glands with radiotherapy dose levels. The validity of blur kernel predictions was assessed by the testing the method's ability to recover artificial 'pseudokernels'. AUC parameters were compared with monoexponential, biexponential, and triexponential model parameters in terms of their correlations with dose, contrast-to-noise (CNR) around parotid glands, and relative importance via principal component analysis.Main results. Image denoising improved blind image quality metrics, smoothed the signal versusb-value curve, and strengthened correlations between IVIM parameters and dose levels. Image TV was reduced and parameter CNRs generally increased following denoising.AUCparameters were more correlated with dose and had higher relative importance than exponential model parameters.Significance. IVIM parameters have high variability in the literature and perfusion-related parameters are difficult to interpret. Describing the signal versusb-value curve with model-independent parameters like theAUCand preprocessing images with denoising techniques could potentially benefit IVIM image parameterization in terms of reproducibility and functional utility.

BAF-Net: bidirectional attention-aware fluid pyramid feature integrated multimodal fusion network for diagnosis and prognosis

Objective. To go beyond the deficiencies of the three conventional multimodal fusion strategies (i.e. input-, feature- and output-level fusion), we propose a bidirectional attention-aware fluid pyramid feature integrated fusion network (BAF-Net) with cross-modal interactions for multimodal medical image diagnosis and prognosis.Approach. BAF-Net is composed of two identical branches to preserve the unimodal features and one bidirectional attention-aware distillation stream to progressively assimilate cross-modal complements and to learn supplementary features in both bottom-up and top-down processes. Fluid pyramid connections were adopted to integrate the hierarchical features at different levels of the network, and channel-wise attention modules were exploited to mitigate cross-modal cross-level incompatibility. Furthermore, depth-wise separable convolution was introduced to fuse the cross-modal cross-level features to alleviate the increase in parameters to a great extent. The generalization abilities of BAF-Net were evaluated in terms of two clinical tasks: (1) an in-house PET-CT dataset with 174 patients for differentiation between lung cancer and pulmonary tuberculosis. (2) A public multicenter PET-CT head and neck cancer dataset with 800 patients from nine centers for overall survival prediction.Main results. On the LC-PTB dataset, improved performance was found in BAF-Net (AUC = 0.7342) compared with input-level fusion model (AUC = 0.6825;p< 0.05), feature-level fusion model (AUC = 0.6968;p= 0.0547), output-level fusion model (AUC = 0.7011;p< 0.05). On the H&N cancer dataset, BAF-Net (C-index = 0.7241) outperformed the input-, feature-, and output-level fusion model, with 2.95%, 3.77%, and 1.52% increments of C-index (p= 0.3336, 0.0479 and 0.2911, respectively). The ablation experiments demonstrated the effectiveness of all the designed modules regarding all the evaluated metrics in both datasets.Significance. Extensive experiments on two datasets demonstrated better performance and robustness of BAF-Net than three conventional fusion strategies and PET or CT unimodal network in terms of diagnosis and prognosis.

MRI radiomics in head and neck cancer from reproducibility to combined approaches

The clinical applicability of radiomics in oncology depends on its transferability to real-world settings. However, the absence of standardized radiomics pipelines combined with methodological variability and insufficient reporting may hamper the reproducibility of radiomic analyses, impeding its translation to clinics. This study aimed to identify and replicate published, reproducible radiomic signatures based on magnetic resonance imaging (MRI), for prognosis of overall survival in head and neck squamous cell carcinoma (HNSCC) patients. Seven signatures were identified and reproduced on 58 HNSCC patients from the DB2Decide Project. The analysis focused on: assessing the signatures' reproducibility and replicating them by addressing the insufficient reporting; evaluating their relationship and performances; and proposing a cluster-based approach to combine radiomic signatures, enhancing the prognostic performance. The analysis revealed key insights: (1) despite the signatures were based on different features, high correlations among signatures and features suggested consistency in the description of lesion properties; (2) although the uncertainties in reproducing the signatures, they exhibited a moderate prognostic capability on an external dataset; (3) clustering approaches improved prognostic performance compared to individual signatures. Thus, transparent methodology not only facilitates replication on external datasets but also advances the field, refining prognostic models for potential personalized medicine applications.

Comparison of semi-automatic and manual segmentation methods for tumor delineation on head and neck squamous cell carcinoma (HNSCC) positron emission tomography (PET) images

Objective. Accurate and reproducible tumor delineation on positron emission tomography (PET) images is required to validate predictive and prognostic models based on PET radiomic features. Manual segmentation of tumors is time-consuming whereas semi-automatic methods are easily implementable and inexpensive. This study assessed the reliability of semi-automatic segmentation methods over manual segmentation for tumor delineation in head and neck squamous cell carcinoma (HNSCC) PET images.Approach. We employed manual and six semi-automatic segmentation methods (just enough interaction (JEI), watershed, grow from seeds (GfS), flood filling (FF), 30% SUVmax and 40%SUVmax threshold) using 3D slicer software to extract 128 radiomic features from FDG-PET images of 100 HNSCC patients independently by three operators. We assessed the distributional properties of all features and considered 92 log-transformed features for subsequent analysis. For each paired comparison of a feature, we fitted a separate linear mixed effect model using the method (two levels; manual versus one semi-automatic method) as a fixed effect and the subject and the operator as the random effects. We estimated different statistics-the intraclass correlation coefficient agreement (aICC), limits of agreement (LoA), total deviation index (TDI), coverage probability (CP) and coefficient of individual agreement (CIA)-to evaluate the agreement between the manual and semi-automatic methods.Main results. Accounting for all statistics across 92 features, the JEI method consistently demonstrated acceptable agreement with the manual method, with median values of aICC = 0.86, TDI = 0.94, CP = 0.66, and CIA = 0.91.Significance. This study demonstrated that JEI method is a reliable semi-automatic method for tumor delineation on HNSCC PET images.

Impact Of PET/CT On Treatment In Patients With Head And Neck Squamous Cell Carcinoma

<b><br>Introduction:</b> Although PET/CT is effective for staging HNSCC, its impact on patient management is somewhat controversial. For this reason, we considered it necessary to carry out a study in order to verify whether PET/CT helps to improve the prognosis and treatment in patients. This study was designed to address the impact of PET-FDG imaging when used alongside CT in the staging and therapeutic management of patients with HNSCC.</br> <b><br>Material and methods:</b> Data was collected from 169 patients diagnosed with HNSCC with both CT and PET/CT (performed within a maximum of 30 days of each other). It was evaluated whether discrepancies in the diagnosis of the two imaging tests had impacted the treatment.</br> <b><br>Results:</b> The combined use of CT and PET/CT led to a change in the treatment of 67 patients, who represented 39.7% of the sample. In 27.2% of cases, it entailed a change in the type of treatment which the patient received. In 3.0% of the cases, using both diagnostic tests led to modifications of the therapeutic intention of our patients.</br> <b><br>Conclusions:</b> Using PET/CT in addition to the conventional imaging method in staging resulted in more successful staging and more appropriate therapeutic decision-making.</br>.

Prediction of the Ki-67 expression level in head and neck squamous cell carcinoma with machine learning-based multiparametric MRI radiomics: a multicenter study

BACKGROUND

This study aimed to develop and validate a machine learning (ML)-based fusion model to preoperatively predict Ki-67 expression levels in patients with head and neck squamous cell carcinoma (HNSCC) using multiparametric magnetic resonance imaging (MRI).

METHODS

A total of 351 patients with pathologically proven HNSCC from two medical centers were retrospectively enrolled in the study and divided into training (n = 196), internal validation (n = 84), and external validation (n = 71) cohorts. Radiomics features were extracted from T2-weighted images and contrast-enhanced T1-weighted images and screened. Seven ML classifiers, including k-nearest neighbors (KNN), support vector machine (SVM), logistic regression (LR), random forest (RF), linear discriminant analysis (LDA), naive Bayes (NB), and eXtreme Gradient Boosting (XGBoost) were trained. The best classifier was used to calculate radiomics (Rad)-scores and combine clinical factors to construct a fusion model. Performance was evaluated based on calibration, discrimination, reclassification, and clinical utility.

RESULTS

Thirteen features combining multiparametric MRI were finally selected. The SVM classifier showed the best performance, with the highest average area under the curve (AUC) of 0.851 in the validation cohorts. The fusion model incorporating SVM-based Rad-scores with clinical T stage and MR-reported lymph node status achieved encouraging predictive performance in the training (AUC = 0.916), internal validation (AUC = 0.903), and external validation (AUC = 0.885) cohorts. Furthermore, the fusion model showed better clinical benefit and higher classification accuracy than the clinical model.

CONCLUSIONS

The ML-based fusion model based on multiparametric MRI exhibited promise for predicting Ki-67 expression levels in HNSCC patients, which might be helpful for prognosis evaluation and clinical decision-making.

Identification of new head and neck squamous cell carcinoma molecular imaging targets

OBJECTIVES

Intraoperative fluorescence imaging (FI) of head and neck squamous cell carcinoma (HNSCC) is performed to identify tumour-positive surgical margins, currently using epidermal growth factor receptor (EGFR) as imaging target. EGFR, not exclusively present in HNSCC, may result in non-specific tracer accumulation in normal tissues. We aimed to identify new potential HNSCC FI targets.

MATERIALS AND METHODS

Publicly available transcriptomic data were collected, and a biostatistical method (Transcriptional Adaptation to Copy Number Alterations (TACNA)-profiling) was applied. TACNA-profiling captures downstream effects of CNAs on mRNA levels, which may translate to protein-level overexpression. Overexpressed genes were identified by comparing HNSCC versus healthy oral mucosa. Potential targets, selected based on overexpression and plasma membrane expression, were immunohistochemically stained. Expression was compared to EGFR on paired biopsies of HNSCC, adjacent macroscopically suspicious mucosa, and healthy mucosa.

RESULTS

TACNA-profiling was applied on 111 healthy oral mucosa and 410 HNSCC samples, comparing expression levels of 19,635 genes. The newly identified targets were glucose transporter-1 (GLUT-1), placental cadherin (P-cadherin), monocarboxylate transporter-1 (MCT-1), and neural/glial antigen-2 (NG2), and were evaluated by IHC on samples of 31 patients. GLUT-1 was expressed in 100 % (median; range: 60-100 %) of tumour cells, P-cadherin in 100 % (50-100 %), EGFR in 70 % (0-100 %), MCT-1 in 30 % (0-100 %), and NG2 in 10 % (0-70 %). GLUT-1 and P-cadherin showed higher expression than EGFR (p < 0.001 and p = 0.015).

CONCLUSIONS

The immunohistochemical confirmation of TACNA-profiling results showed significantly higher GLUT-1 and P-cadherin expression than EGFR, warranting further investigation as HNSCC FI targets.

Quarter Century Positron Emission Tomography/Computed Tomography Transformation of Oncology: Head and Neck Cancer

During the last 2 decades, f-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F FDG PET/CT) has transformed the clinical head and neck cancer imaging for patient management and predicting survival outcomes. It is now widely used for staging, radiotherapy planning, posttherapy assessment, and for detecting recurrence in head and neck cancers and is widely included in NCCN and other evidence based clinical practice guidelines. Future Directions would include evaluating the potential value of FAPI PET/CT for head and neck cancers, opportunity to use volumetric and tumor heterogeneity parameters and deploying AI in diagnostic and therapeutic assessments.

Deep Semisupervised Transfer Learning for Fully Automated Whole-Body Tumor Quantification and Prognosis of Cancer on PET/CT

Automatic detection and characterization of cancer are important clinical needs to optimize early treatment. We developed a deep, semisupervised transfer learning approach for fully automated, whole-body tumor segmentation and prognosis on PET/CT. Methods: This retrospective study consisted of 611 18F-FDG PET/CT scans of patients with lung cancer, melanoma, lymphoma, head and neck cancer, and breast cancer and 408 prostate-specific membrane antigen (PSMA) PET/CT scans of patients with prostate cancer. The approach had a nnU-net backbone and learned the segmentation task on 18F-FDG and PSMA PET/CT images using limited annotations and radiomics analysis. True-positive rate and Dice similarity coefficient were assessed to evaluate segmentation performance. Prognostic models were developed using imaging measures extracted from predicted segmentations to perform risk stratification of prostate cancer based on follow-up prostate-specific antigen levels, survival estimation of head and neck cancer by the Kaplan-Meier method and Cox regression analysis, and pathologic complete response prediction of breast cancer after neoadjuvant chemotherapy. Overall accuracy and area under the receiver-operating-characteristic (AUC) curve were assessed. Results: Our approach yielded median true-positive rates of 0.75, 0.85, 0.87, and 0.75 and median Dice similarity coefficients of 0.81, 0.76, 0.83, and 0.73 for patients with lung cancer, melanoma, lymphoma, and prostate cancer, respectively, on the tumor segmentation task. The risk model for prostate cancer yielded an overall accuracy of 0.83 and an AUC of 0.86. Patients classified as low- to intermediate- and high-risk had mean follow-up prostate-specific antigen levels of 18.61 and 727.46 ng/mL, respectively (P < 0.05). The risk score for head and neck cancer was significantly associated with overall survival by univariable and multivariable Cox regression analyses (P < 0.05). Predictive models for breast cancer predicted pathologic complete response using only pretherapy imaging measures and both pre- and posttherapy measures with accuracies of 0.72 and 0.84 and AUCs of 0.72 and 0.76, respectively. Conclusion: The proposed approach demonstrated accurate tumor segmentation and prognosis in patients across 6 cancer types on 18F-FDG and PSMA PET/CT scans.

The utility of ultrasonographic surveillance in management of a presumed branchial cleft cyst later confirmed HPV-associated oropharyngeal cancer

While branchial cleft cysts are often considered benign pathologies, the literature discusses cases of squamous cell carcinoma (SCC) arising from these cystic lesions as either a primary or metastatic tumor. We illustrate our institutional experience and review the current literature to identify recommendations for best diagnostic, surveillance, and treatment guidelines for SCC identified in a branchial cleft cyst. A 61-year-old male presented with a right sided neck mass, with suspicion of a branchial cleft cyst due to benign findings on fine needle aspiration. Following surgical excision, a focus of SCC was found on surgical pathology. Despite PET/CT and flexible laryngoscopy, no primary tumor was identified prompting routine surveillance every 3 months with cervical ultrasonography and flexible nasolaryngoscopy. Two and a half years following his initial presentation, pathologic right level II lymphadenopathy was detected on ultrasound without evidence of primary tumor. Subsequent transoral robotic surgery with right tonsillectomy and partial pharyngectomy, with right lateral neck dissection revealed a diagnosis of pT1N1 HPV-HNSCC and he was referred for adjuvant chemotherapy and radiation. To our knowledge there are less than 10 cases of confirmed HPV-associated oropharyngeal SCC arising from a branchial cleft cyst. Here we demonstrate the utility of ultrasound as a surveillance tool and emphasize a higher index of suspicion for carcinoma in adult patients with cystic neck masses.

The clinical acceptability of short versus long duration acquisitions for head and neck cancer using long-axial field-of-view PET/CT: a retrospective evaluation

PURPOSE

To evaluate the utility of long duration (10 min) acquisitions compared to standard 4 min scans in the evaluation of head and neck cancer (HNC) using a long-axial field-of-view (LAFOV) system in 2-[18F]FDG PET/CT.

METHODS

HNC patients undergoing LAFOV PET/CT were included retrospectively according to a predefined sample size calculation. For each acquisition, FDG avid lymph nodes (LN) which were highly probable or equivocal for malignancy were identified by two board certified nuclear medicine physicians in consensus. The aim of this study was to establish the clinical acceptability of short-duration (4 min, C40%) acquisitions compared to full-count (10 min, C100%) in terms of the detection of LN metastases in HNC. Secondary endpoints were the positive predictive value for LN status (PPV) and comparison of SUVmax at C40% and C100%. Histology reports or confirmatory imaging were the reference standard.

RESULTS

A total of 1218 records were screened and target recruitment was met with n = 64 HNC patients undergoing LAFOV. Median age was 65 years (IQR: 59-73). At C40%, a total of 387 lesions were detected (highly probable LN n = 274 and equivocal n = 113. The total number of lesions detected at C100% acquisition was 439, of them 291 (66%) highly probable LN and 148 (34%) equivocal. Detection rate between the two acquisitions did not demonstrate any significant differences (Pearson's Chi-Square test, p = 0.792). Sensitivity, specificity, PPV, NPV and accuracy for C40% were 83%, 44%, 55%, 76% and 36%, whilst for C100% were 85%, 56%, 55%, 85% and 43%, respectively. The improved accuracy reached borderline significance (p = 0.057). At the ROC analysis, lower SUVmax was identified for C100% (3.5) compared to C40% (4.5).

CONCLUSION

In terms of LN detection, C40% acquisitions showed no significant difference compared to the C100% acquisitions. There was some improvement for lesions detection at C100%, with a small increment in accuracy reaching borderline significance, suggestive that the higher sensitivity afforded by LAFOV might translate to improved clinical performance in some patients.

Hyperspectral imaging and evaluation of surgical margins: where do we stand?

PURPOSE OF REVIEW

To highlight the recent literature on the use of hyperspectral imaging (HSI) for cancer margin evaluation ex vivo, for head and neck cancer pathology and in vivo during head and neck cancer surgery.

RECENT FINDINGS

HSI can be used ex vivo on unstained and stained tissue sections to analyze head and neck tissue and tumor cells in combination with machine learning approaches to analyze head and neck cancer cell characteristics and to discriminate the tumor border from normal tissue. Data on in vivo applications during head and neck cancer surgery are preliminary and limited. Even now an accuracy of 80% for tumor versus nonneoplastic tissue classification can be achieved for certain tasks, within the current in vivo settings.

SUMMARY

Significant progress has been made to introduce HSI for ex vivo head and neck cancer pathology evaluation and for an intraoperative use to define the tumor margins. To optimize the accuracy for in vivo use, larger HSI databases with annotations for head and neck cancer are needed.

RADCURE: An open-source head and neck cancer CT dataset for clinical radiation therapy insights

PURPOSE

This manuscript presents RADCURE, one of the most extensive head and neck cancer (HNC) imaging datasets accessible to the public. Initially collected for clinical radiation therapy (RT) treatment planning, this dataset has been retrospectively reconstructed for use in imaging research.

ACQUISITION AND VALIDATION METHODS

RADCURE encompasses data from 3346 patients, featuring computed tomography (CT) RT simulation images with corresponding target and organ-at-risk contours. These CT scans were collected using systems from three different manufacturers. Standard clinical imaging protocols were followed, and contours were manually generated and reviewed at weekly RT quality assurance rounds. RADCURE imaging and structure set data was extracted from our institution's radiation treatment planning and oncology information systems using a custom-built data mining and processing system. Furthermore, images were linked to our clinical anthology of outcomes data for each patient and includes demographic, clinical and treatment information based on the 7th edition TNM staging system (Tumor-Node-Metastasis Classification System of Malignant Tumors). The median patient age is 63, with the final dataset including 80% males. Half of the cohort is diagnosed with oropharyngeal cancer, while laryngeal, nasopharyngeal, and hypopharyngeal cancers account for 25%, 12%, and 5% of cases, respectively. The median duration of follow-up is five years, with 60% of the cohort surviving until the last follow-up point.

DATA FORMAT AND USAGE NOTES

The dataset provides images and contours in DICOM CT and RT-STRUCT formats, respectively. We have standardized the nomenclature for individual contours-such as the gross primary tumor, gross nodal volumes, and 19 organs-at-risk-to enhance the RT-STRUCT files' utility. Accompanying demographic, clinical, and treatment data are supplied in a comma-separated values (CSV) file format. This comprehensive dataset is publicly accessible via The Cancer Imaging Archive.

POTENTIAL APPLICATIONS

RADCURE's amalgamation of imaging, clinical, demographic, and treatment data renders it an invaluable resource for a broad spectrum of radiomics image analysis research endeavors. Researchers can utilize this dataset to advance routine clinical procedures using machine learning or artificial intelligence, to identify new non-invasive biomarkers, or to forge prognostic models.

A radiomics model to predict γδ T-cell abundance and overall survival in head and neck squamous cell carcinoma

γδ T cells are becoming increasingly popular because of their attractive potential for antitumor immunotherapy. However, the role and assessment of γδ T cells in head and neck squamous cell carcinoma (HNSCC) are not well understood. We aimed to explore the prognostic value of γδ T cell and predict its abundance using a radiomics model. Computer tomography images with corresponding gene expression data and clinicopathological data were obtained from online databases. After outlining the volumes of interest manually, the radiomic features were screened using maximum melevance minimum redundancy and recursive feature elimination algorithms. A radiomics model was developed to predict γδ T-cell abundance using gradient boosting machine. Kaplan-Meier survival curves and univariate and multivariate Cox regression analyses were used for the survival analysis. In this study, we confirmed that γδ T-cell abundance was an independent predictor of favorable overall survival (OS) in patients with HNSCC. Moreover, a radiomics model was built to predict the γδ T-cell abundance level (the areas under the operating characteristic curves of 0.847 and 0.798 in the training and validation sets, respectively). The calibration and decision curves analysis demonstrated the fitness of the model. The high radiomic score was an independent protective factor for OS. Our results indicated that γδ T-cell abundance was a promising prognostic predictor in HNSCC, and the radiomics model could discriminate its abundance levels and predict OS. The noninvasive radiomics model provided a potentially powerful prediction tool to aid clinical judgment and antitumor immunotherapy.

Oxygen-enhanced MRI assessment of tumour hypoxia in head and neck cancer is feasible and well tolerated in the clinical setting

BACKGROUND

Tumour hypoxia is a recognised cause of radiotherapy treatment resistance in head and neck squamous cell carcinoma (HNSCC). Current positron emission tomography-based hypoxia imaging techniques are not routinely available in many centres. We investigated if an alternative technique called oxygen-enhanced magnetic resonance imaging (OE-MRI) could be performed in HNSCC.

METHODS

A volumetric OE-MRI protocol for dynamic T1 relaxation time mapping was implemented on 1.5-T clinical scanners. Participants were scanned breathing room air and during high-flow oxygen administration. Oxygen-induced changes in T1 times (ΔT1) and R2* rates (ΔR2*) were measured in malignant tissue and healthy organs. Unequal variance t-test was used. Patients were surveyed on their experience of the OE-MRI protocol.

RESULTS

Fifteen patients with HNSCC (median age 59 years, range 38 to 76) and 10 non-HNSCC subjects (median age 46.5 years, range 32 to 62) were scanned; the OE-MRI acquisition took less than 10 min and was well tolerated. Fifteen histologically confirmed primary tumours and 41 malignant nodal masses were identified. Median (range) of ΔT1 times and hypoxic fraction estimates for primary tumours were -3.5% (-7.0 to -0.3%) and 30.7% (6.5 to 78.6%) respectively. Radiotherapy-responsive and radiotherapy-resistant primary tumours had mean estimated hypoxic fractions of 36.8% (95% confidence interval [CI] 17.4 to 56.2%) and 59.0% (95% CI 44.6 to 73.3%), respectively (p = 0.111).

CONCLUSIONS

We present a well-tolerated implementation of dynamic, volumetric OE-MRI of the head and neck region allowing discernment of differing oxygen responses within biopsy-confirmed HNSCC.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT04724096 . Registered on 26 January 2021.

RELEVANCE STATEMENT

MRI of tumour hypoxia in head and neck cancer using routine clinical equipment is feasible and well tolerated and allows estimates of tumour hypoxic fractions in less than ten minutes.

KEY POINTS

• Oxygen-enhanced MRI (OE-MRI) can estimate tumour hypoxic fractions in ten-minute scanning. • OE-MRI may be incorporable into routine clinical tumour imaging. • OE-MRI has the potential to predict outcomes after radiotherapy treatment.

Imaging of Tumor Stroma Using 68Ga-FAPI PET/CT to Improve Diagnostic Accuracy of Primary Tumors in Head and Neck Cancer of Unknown Primary: A Comparative Imaging Trial

The low detection rate of primary tumors by current diagnostic techniques remains a major concern for patients with head and neck cancer of unknown primary (HNCUP). Therefore, in this study, we aimed to investigate the potential role of 68Ga-labeled fibroblast activation protein inhibitor (68Ga-FAPI) PET/CT compared with 18F-FDG PET/CT for the detection of primary tumors of HNCUP. Methods: In this prospective comparative imaging trial conducted at Fudan University Shanghai Cancer Center, 91 patients with negative or equivocal findings of a primary tumor by comprehensive clinical examination and conventional imaging were enrolled from June 2020 to September 2022. The presence of a primary tumor was recorded by 3 experienced nuclear medicine physicians. Primary lesions were validated by histopathologic analysis and a composite reference standard. Results: Of the 91 patients (18 women, 73 men; median age, 60 y; age range, 24-76 y), primary tumors were detected in 46 (51%) patients after a thorough diagnostic work-up. 68Ga-FAPI PET/CT detected more primary lesions than 18F-FDG PET/CT (46 vs. 17, P < 0.001) and showed better sensitivity, positive predictive value, and accuracy in locating primary tumors (51% vs. 25%, 98% vs. 43%, and 51% vs. 19%, respectively). Furthermore, 68Ga-FAPI PET/CT led to treatment changes in 22 of 91 (24%) patients compared with 18F-FDG PET/CT. The Kaplan-Meier curve illustrated that patients with unidentified primary tumors had a significantly worse prognosis than patients with identified primary tumors (hazard ratio, 5.77; 95% CI, 1.86-17.94; P = 0.0097). Conclusion: 68Ga-FAPI PET/CT outperforms 18F-FDG PET/CT in detecting primary lesions and could serve as a sensitive, reliable, and reproducible imaging modality for HNCUP patients.

Pencil Beam Scanning Proton Therapy for Pregnant Patients With Brain and Head and Neck Cancers

PURPOSE

When radiation therapy is medically necessary for pregnant patients, photon-based treatments (XRT) have traditionally been used, whereas proton radiation therapy (PRT) is avoided due to concerns about neutron dose. This retrospective study analyzes pregnant patients treated with XRT and models the equivalent dose that would have been delivered to the fetus with proton radiation compared with XRT. The purpose of this work is to provide a comprehensive analysis of pencil beam scanning proton therapy (PBS-PRT) for pregnant patients and to evaluate whether PBS-PRT should be the new standard of practice for treating pregnant patients with brain and head and neck cancers.

METHODS AND MATERIALS

PBS-PRT plans were made for seven pregnant patients who received XRT: four treated for brain tumors and three for head and neck tumors. Measurements were performed with the patient plans using an anthropomorphic phantom and Wendi-2 meter placed at the phantom's abdomen. Patient-specific measurements were used to determine the total fetal equivalent dose from PBS-PRT compared with XRT. Imaging dose was also evaluated with a Fluke 451 dose meter.

RESULTS

The average measured fetal equivalent dose, accounting for photons and neutrons, for the brain plans was 0.4 mSv for PBS-PRT and 7 mSv for XRT. For the head and neck plans, it was 6 mSv and 90 mSv for PBS-PRT and XRT, respectively. The PBS-PRT plans were preferred by the physicians for both tumor coverage and normal-tissue sparing. Daily imaging added between 0.05 and 1.5 mSv to the total dose.

CONCLUSIONS

This retrospective study showed that when treating brain or head and neck cancers in pregnant patients, fetal equivalent dose is reduced by approximately a factor of 10 with PBS-PRT compared with XRT without making any compromises in treatment planning objectives. These results support a change of practice to using PBS-PRT as the new standard for treating pregnant patients with brain or head and neck tumors compared with XRT.

A gradient mapping guided explainable deep neural network for extracapsular extension identification in 3D head and neck cancer computed tomography images

BACKGROUND

Diagnosis and treatment management for head and neck squamous cell carcinoma (HNSCC) is guided by routine diagnostic head and neck computed tomography (CT) scans to identify tumor and lymph node features. The extracapsular extension (ECE) is a strong predictor of patients' survival outcomes with HNSCC. It is essential to detect the occurrence of ECE as it changes staging and treatment planning for patients. Current clinical ECE detection relies on visual identification and pathologic confirmation conducted by clinicians. However, manual annotation of the lymph node region is a required data preprocessing step in most of the current machine learning-based ECE diagnosis studies.

PURPOSE

In this paper, we propose a Gradient Mapping Guided Explainable Network (GMGENet) framework to perform ECE identification automatically without requiring annotated lymph node region information.

METHODS

The gradient-weighted class activation mapping (Grad-CAM) technique is applied to guide the deep learning algorithm to focus on the regions that are highly related to ECE. The proposed framework includes an extractor and a classifier. In a joint training process, informative volumes of interest (VOIs) are extracted by the extractor without labeled lymph node region information, and the classifier learns the pattern to classify the extracted VOIs into ECE positive and negative.

RESULTS

In evaluation, the proposed methods are well-trained and tested using cross-validation. GMGENet achieved test accuracy and area under the curve (AUC) of 92.2% and 89.3%, respectively. GMGENetV2 achieved 90.3% accuracy and 91.7% AUC in the test. The results were compared with different existing models and further confirmed and explained by generating ECE probability heatmaps via a Grad-CAM technique. The presence or absence of ECE has been analyzed and correlated with ground truth histopathological findings.

CONCLUSIONS

The proposed deep network can learn meaningful patterns to identify ECE without providing lymph node contours. The introduced ECE heatmaps will contribute to the clinical implementations of the proposed model and reveal unknown features to radiologists. The outcome of this study is expected to promote the implementation of explainable artificial intelligence-assiste ECE detection.

Deformable image registration for composite planned doses during adaptive radiation therapy

INTRODUCTION

Some patients have significant anatomic changes during radiotherapy, necessitating an adaptive repeat CT-simulation and re-planning. This yields two unique planning datasets that introduce uncertainty into total dose records. This study explored the impact of using deformable image registration (DIR) to spatially align repeat CT-simulation images and calculate total planned dose distributions.

MATERIALS & METHODS

Data from 5 head-and-neck, 5 lung, and 5 sarcoma patients who had unanticipated re-planning during radiotherapy were analyzed in a treatment planning system (RayStation v6.1 RaySearch Laboratories). Total planned doses to normal tissues were calculated using two methods and the previously generated manual contours defined on each CT. The first method, termed 'parameter addition', simply sums the relevant DVH metrics from the initial and re-planned distributions without spatially registering the CTs. The second, termed 'dose accumulation', uses a validated hybrid contour/intensity-based DIR algorithm to deform initial CT and dose distribution onto the repeat CT and re-planning dose distribution. DVH metrics from the summed distribution on the repeat CT are then calculated. Dose differences for organs-at-risk between parameter addition and dose accumulation ≥100 cGy were assumed to be clinically relevant. To elucidate whether relevant differences were due to registration accuracy or contouring variability between CTs, the analysis was repeated using contours on the first CT and the same contours deformed to the repeat CT with DIR.

RESULTS

For all patients, high overall DIR accuracy was verified visually (qualitatively) and numerically (quantitatively) using image similarity and contour-based metrics. All head-and-neck and lung patients, and one sarcoma patient (11 of 15 total) had dose differences between parameter addition and dose accumulation ≥100 cGy, with absolute mean differences of 160 cGy (range 101-436 cGy) seen in 41 of 205 total DVH criteria. In 22 of these 41 criteria, these differences were attributed to contouring variability between CTs. After correcting for contouring variations using DIR, the mean absolute differences in 7 of these 22 criteria with a relevant result (across 6 patients) was 146 cGy (range 100-502 cGy). In only 4 DVH criteria, the DIR mapped contours had higher variations than the original contours. One lung patient had a DVH criteria exceeding the clinical dose constraint by 125 cGy with parameter addition, and with accurate DIR and dose accumulation, the criteria was actually 97 cGy lower than the constraint.

CONCLUSIONS

The use of DIR to generate total planned dose records revealed substantial dose differences in most cases compared to commonly used clinical methods (i.e. parameter addition), and altered the planned acceptance criteria in a minority. DIR is recommended to be used for future adaptive re-plans to generate total planned dose records and facilitate accurate re-contouring. More accurate dose records may also improve our understanding of clinical outcomes.

Pilomatrixoma and its Imitators

Introduction: Pilomatrixomas are benign neoplasms derived from hair follicle matrix cells. They are among the most common soft tissue head and neck tumors of childhood. Pilomatrixomas are typically isolated, slow-growing, firm, nontender masses that are adherent to the epidermis but mobile in the subcutaneous plane. This clinical presentation is so characteristic that many experienced surgeons will excise suspected pilomatrixomas without prior imaging. We reviewed the results of this approach to determine whether physical examination alone differentiates pilomatrixomas from other similar soft tissue lesions of the pediatric head and neck. Methods: Computerized review of all pilomatrixomas over a 20-year period in a single academic pediatric otolaryngology practice. Results: 18 patients presented to our pediatric otolaryngology practice between 2001 and 2021 with historical and physical findings consistent with pilomatrixoma. Of the 18 patients, 7 were male and 11 were female. Ages ranged from 1.5 to 14 years, with a mean of 7.5 years. Most of the lesions (12) were located in the head and face, while the rest (6) were found in the neck. All patients were treated with complete surgical excision. Pathology confirmed pilomatrixoma in 15 patients. The remaining 3 children were found to have an epidermal inclusion cyst, a ruptured trichilemmal cyst, and a giant molluscum contagiosum lesion, respectively. One additional patient presented with a small lesion of the auricular helix that was thought to be a dermoid cyst, but proved to be a pilomatrixoma on histologic examination. Discussion: As pilomatrixomas are common and have a very characteristic presentation, surgical excision without prior diagnostic imaging will lead to correct treatment in the majority of cases. High resolution ultrasonography can help to confirm the diagnosis preoperatively, but is not definitive in large case series. Most of the cystic lesions that imitate pilomatrixoma will ultimately require surgical excision.

Artificial intelligence-based image-domain material decomposition in single-energy computed tomography for head and neck cancer

PURPOSE

While dual-energy computed tomography (DECT) images provide clinically useful information than single-energy CT (SECT), SECT remains the most widely used CT system globally, and only a few institutions can use DECT. This study aimed to establish an artificial intelligence (AI)-based image-domain material decomposition technique using multiple keV-output learning of virtual monochromatic images (VMIs) to create DECT-equivalent images from SECT images.

METHODS

This study involved 82 patients with head and neck cancer. Of these, the AI model was built with data from the 67 patients with only DECT scans, while 15 patients with both SECT and DECT scans were used for SECT testing. Our AI model generated VMI50keV and VMI100keV from VMI70keV equivalent to 120-kVp SECT images. We introduced a loss function for material density images (MDIs) in addition to the loss for VMIs. For comparison, we trained the same model with the loss for VMIs only. DECT-equivalent images were generated from SECT images and compared with the true DECT images.

RESULTS

The prediction time was 5.4 s per patient. The proposed method with the MDI loss function quantitatively provided more accurate DECT-equivalent images than the model trained with the loss for VMIs only. Using real 120-kVp SECT images, the trained model produced precise DECT images of excellent quality.

CONCLUSION

In this study, we developed an AI-based material decomposition approach for head and neck cancer patients by introducing the loss function for MDIs via multiple keV-output learning. Our results suggest the feasibility of AI-based image-domain material decomposition in a conventional SECT system without a DECT scanner.

Posttreatment Head and Neck Cancer Imaging: Anatomic Considerations Based on Cancer Subsites

Posttreatment imaging surveillance of head and neck cancer is challenging owing to complex anatomic subsites and diverse treatment modalities. Early detection of residual disease or recurrence through surveillance imaging is crucial for devising optimal treatment strategies. Posttreatment imaging surveillance is performed using CT, fluorine 18-fluorodeoxyglucose PET/CT, and MRI. Radiologists should be familiar with postoperative imaging findings that can vary depending on surgical procedures and reconstruction methods that are used, which is dictated by the primary subsite and extent of the tumor. Morphologic changes in normal structures or denervation of muscles within the musculocutaneous flap may mimic recurrent tumors. Recurrence is more likely to occur at the resection margin, margin of the reconstructed flap, and deep sites that are difficult to access surgically. Radiation therapy also has a varying dose distribution depending on the primary site, resulting in various posttreatment changes. Normal tissues are affected by radiation, with edema and inflammation occurring in the early stages and fibrosis in the late stages. Distinguishing scar tissue from residual tumor becomes necessary, as radiation therapy may leave behind residual scar tissue. Local recurrence should be carefully evaluated within areas where these postradiation changes occur. Head and Neck Imaging Reporting and Data System (NI-RADS) is a standardized reporting and risk classification system with guidance for subsequent management. Familiarity with NI-RADS has implications for establishing surveillance protocols, interpreting posttreatment images, and management decisions. Knowledge of posttreatment imaging characteristics of each subsite of head and neck cancers and the areas prone to recurrence empowers radiologists to detect recurrences at early stages. ©RSNA, 2024 Test Your Knowledge questions in the supplemental material and the slide presentation from the RSNA Annual Meeting are available for this article.

vOARiability: Interobserver and intermodality variability analysis in OAR contouring from head and neck CT and MR images

BACKGROUND

Accurate and consistent contouring of organs-at-risk (OARs) from medical images is a key step of radiotherapy (RT) cancer treatment planning. Most contouring approaches rely on computed tomography (CT) images, but the integration of complementary magnetic resonance (MR) modality is highly recommended, especially from the perspective of OAR contouring, synthetic CT and MR image generation for MR-only RT, and MR-guided RT. Although MR has been recognized as valuable for contouring OARs in the head and neck (HaN) region, the accuracy and consistency of the resulting contours have not been yet objectively evaluated.

PURPOSE

To analyze the interobserver and intermodality variability in contouring OARs in the HaN region, performed by observers with different level of experience from CT and MR images of the same patients.

METHODS

In the final cohort of 27 CT and MR images of the same patients, contours of up to 31 OARs were obtained by a radiation oncology resident (junior observer, JO) and a board-certified radiation oncologist (senior observer, SO). The resulting contours were then evaluated in terms of interobserver variability, characterized as the agreement among different observers (JO and SO) when contouring OARs in a selected modality (CT or MR), and intermodality variability, characterized as the agreement among different modalities (CT and MR) when OARs were contoured by a selected observer (JO or SO), both by the Dice coefficient (DC) and 95-percentile Hausdorff distance (HD95 $_{95}$ ).

RESULTS

The mean (±standard deviation) interobserver variability was 69.0 ± 20.2% and 5.1 ± 4.1 mm, while the mean intermodality variability was 61.6 ± 19.0% and 6.1 ± 4.3 mm in terms of DC and HD95 $_{95}$ , respectively, across all OARs. Statistically significant differences were only found for specific OARs. The performed MR to CT image registration resulted in a mean target registration error of 1.7 ± 0.5 mm, which was considered as valid for the analysis of intermodality variability.

CONCLUSIONS

The contouring variability was, in general, similar for both image modalities, and experience did not considerably affect the contouring performance. However, the results indicate that an OAR is difficult to contour regardless of whether it is contoured in the CT or MR image, and that observer experience may be an important factor for OARs that are deemed difficult to contour. Several of the differences in the resulting variability can be also attributed to adherence to guidelines, especially for OARs with poor visibility or without distinctive boundaries in either CT or MR images. Although considerable contouring differences were observed for specific OARs, it can be concluded that almost all OARs can be contoured with a similar degree of variability in either the CT or MR modality, which works in favor of MR images from the perspective of MR-only and MR-guided RT.

Head and neck mucosal melanoma: radiological considerations and UK imaging guidelines

PURPOSE

Awareness of head and neck mucosal melanoma (HNMM) is important, as incorrect work-up can impact on the investigation and management of this rare and aggressive cancer. Following on from the 2020 HNMM UK guidelines, we set out the imaging recommendations and their rationale. To illustrate the key imaging characteristics, we also include a case series from our centre.

METHODS

All HNMM cases managed at our institution from January 2016 to January 2021 were identified, and the available imaging for each patient was reviewed. For each patient, the age, gender and location of primary tumour was recorded together with key staging and diagnostic imaging parameters.

RESULTS

A total of 14 patients were identified. The median age was 65 years with a female to male ratio of 1.33:1. Primary tumours were sinonasal in location in 93% of cases, with 7% of patients having metastatic neck nodes at presentation and 21% of cases having distant metastatic disease at presentation.

CONCLUSION

This data set is in general concordance with other published series regarding the sinonasal origin of the vast majority of HNMM tumours along with the proportion of patients with metastatic neck nodes and distant metastases at presentation. We recommend dual-modality imaging with computed tomography (CT) and magnetic resonance imaging (MRI) of primary tumours whenever possible. In the systematic staging of HNMM, positron emission tomography (PET)-CT should be strongly considered, together with MRI of the brain. Pre-biopsy imaging of HNMM tumours is advisable whenever possible.

SwinCross: Cross-modal Swin transformer for head-and-neck tumor segmentation in PET/CT images

BACKGROUND

Radiotherapy (RT) combined with cetuximab is the standard treatment for patients with inoperable head and neck cancers. Segmentation of head and neck (H&N) tumors is a prerequisite for radiotherapy planning but a time-consuming process. In recent years, deep convolutional neural networks (DCNN) have become the de facto standard for automated image segmentation. However, due to the expensive computational cost associated with enlarging the field of view in DCNNs, their ability to model long-range dependency is still limited, and this can result in sub-optimal segmentation performance for objects with background context spanning over long distances. On the other hand, Transformer models have demonstrated excellent capabilities in capturing such long-range information in several semantic segmentation tasks performed on medical images.

PURPOSE

Despite the impressive representation capacity of vision transformer models, current vision transformer-based segmentation models still suffer from inconsistent and incorrect dense predictions when fed with multi-modal input data. We suspect that the power of their self-attention mechanism may be limited in extracting the complementary information that exists in multi-modal data. To this end, we propose a novel segmentation model, debuted, Cross-modal Swin Transformer (SwinCross), with cross-modal attention (CMA) module to incorporate cross-modal feature extraction at multiple resolutions.

METHODS

We propose a novel architecture for cross-modal 3D semantic segmentation with two main components: (1) a cross-modal 3D Swin Transformer for integrating information from multiple modalities (PET and CT), and (2) a cross-modal shifted window attention block for learning complementary information from the modalities. To evaluate the efficacy of our approach, we conducted experiments and ablation studies on the HECKTOR 2021 challenge dataset. We compared our method against nnU-Net (the backbone of the top-5 methods in HECKTOR 2021) and other state-of-the-art transformer-based models, including UNETR and Swin UNETR. The experiments employed a five-fold cross-validation setup using PET and CT images.

RESULTS

Empirical evidence demonstrates that our proposed method consistently outperforms the comparative techniques. This success can be attributed to the CMA module's capacity to enhance inter-modality feature representations between PET and CT during head-and-neck tumor segmentation. Notably, SwinCross consistently surpasses Swin UNETR across all five folds, showcasing its proficiency in learning multi-modal feature representations at varying resolutions through the cross-modal attention modules.

CONCLUSIONS

We introduced a cross-modal Swin Transformer for automating the delineation of head and neck tumors in PET and CT images. Our model incorporates a cross-modality attention module, enabling the exchange of features between modalities at multiple resolutions. The experimental results establish the superiority of our method in capturing improved inter-modality correlations between PET and CT for head-and-neck tumor segmentation. Furthermore, the proposed methodology holds applicability to other semantic segmentation tasks involving different imaging modalities like SPECT/CT or PET/MRI. Code:https://github.com/yli192/SwinCross_CrossModalSwinTransformer_for_Medical_Image_Segmentation.

Frontal and anterior temporal hypometabolism post chemoradiation in head and neck cancer: A real-world PET study

BACKGROUND AND PURPOSE

Adverse neurological effects after cancer therapy are common, but biomarkers to diagnose, monitor, or risk stratify patients are still not validated or used clinically. An accessible imaging method, such as fluorodeoxyglucose positron emission tomography (FDG PET) of the brain, could meet this gap and serve as a biomarker for functional brain changes. We utilized FDG PET to evaluate which brain regions are most susceptible to altered glucose metabolism after chemoradiation in patients with head and neck cancer (HNCa).

METHODS

Real-world FDG PET images were acquired as standard of care before and after chemoradiation for HNCa in 68 patients. Linear mixed-effects voxelwise models assessed changes after chemoradiation in cerebral glucose metabolism quantified with standardized uptake value ratio (SUVR), covarying for follow-up time and patient demographics.

RESULTS

Voxelwise analysis revealed two large clusters of decreased glucose metabolism in the medial frontal and polar temporal cortices following chemoradiation, with decreases of approximately 5% SUVR after therapy.

CONCLUSIONS

These findings provide evidence that standard chemoradiation for HNCa can lead to decreased neuronal glucose metabolism, contributing to literature emphasizing the vulnerability of the frontal and anterior temporal lobes, especially in HNCa, where these areas may be particularly vulnerable to indirect radiation-induced injury. FDG PET shows promise as a sensitive biomarker for assessing these changes.

Radiomic Application for Head and Neck Squamocellular Tumor: Systematic Review

Abstract

Radiomics represents the convergence of artificial intelligence and radiological data analysis, primarily applied in the diagnosis and treatment of cancer. In the head and neck region, squamous cell carcinoma is the most prevalent type of tumor. Recent radiomics research has revealed that specific bio-imaging characteristics correlate with various molecular features of Head and Neck Squamous Cell Carcinoma (HNSCC), particularly Human Papillomavirus (HPV). These tumors typically present a unique phenotype, often affecting younger patients, and show a favorable response to radiation therapy. This study provides a systematic review of the literature, summarizing the application of radiomics in the head and neck region. It offers a comprehensive analysis of radiomics-based studies on HNSCC, evaluating its potential for tumor evaluation, risk stratification, and outcome prediction in head and neck cancer treatment.

Multi-organ segmentation of organ-at-risk (OAR's) of head and neck site using ensemble learning technique

INTRODUCTION

This paper presents a novel approach to automate the segmentation of Organ-at-Risk (OAR) in Head and Neck cancer patients using Deep Learning models combined with Ensemble Learning techniques. The study aims to improve the accuracy and efficiency of OAR segmentation, essential for radiotherapy treatment planning.

METHODS

The dataset comprised computed tomography (CT) scans of 182 patients in DICOM format, obtained from an institutional image bank. Experienced Radiation Oncologists manually segmented seven OARs for each scan. Two models, 3D U-Net and 3D DenseNet-FCN, were trained on reduced CT scans (192 × 192 x 128) due to memory limitations. Ensemble Learning techniques were employed to enhance accuracy and segmentation metrics. Testing was conducted on 78 patients from the institutional dataset and an open-source dataset (TCGA-HNSC and Head-Neck Cetuximab) consisting of 31 patient scans.

RESULTS

Using the Ensemble Learning technique, the average dice similarity coefficient for OARs ranged from 0.990 to 0.994, indicating high segmentation accuracy. The 95% Hausdorff distance (mm) ranged from 1.3 to 2.1, demonstrating precise segmentation boundaries.

CONCLUSION

The proposed automated segmentation method achieved efficient and accurate OAR segmentation, surpassing human expert performance in terms of time and accuracy.

IMPLICATIONS FOR PRACTICE

This approach has implications for improving treatment planning and patient care in radiotherapy. By reducing manual segmentation reliance, the proposed method offers significant time savings and potential improvements in treatment planning efficiency and precision for head and neck cancer patients.

3D-IncNet: Head and Neck (H&N) Primary Tumors Segmentation and Survival Prediction

Cancer begins when healthy cells change and grow out of control, forming a mass called a tumor. Head and neck (H&N) cancers usually develop in or around the head and neck, including the mouth (oral cavity), nose and sinuses, throat (pharynx), and voice box (larynx). 4% of all cancers are H&N cancers with a very low survival rate (a five-year survival rate of 64.7%). FDG-PET/CT imaging is often used for early diagnosis and staging of H&N tumors, thus improving these patients' survival rates. This work presents a novel 3D-Inception-Residual aided with 3D depth-wise convolution and squeeze and excitation block. We introduce a 3D depth-wise convolution-inception encoder consisting of an additional 3D squeeze and excitation block and a 3D depth-wise convolution-based residual learning decoder (3D-IncNet), which not only helps to recalibrate the channel-wise features but adaptively through explicit inter-dependencies modeling but also integrate the coarse and fine features resulting in accurate tumor segmentation. We further demonstrate the effectiveness of inception-residual encoder-decoder architecture in achieving better dice scores and the impact of depth-wise convolution in lowering the computational cost. We applied random forest for survival prediction on deep, clinical, and radiomics features. Experiments are conducted on the benchmark HECKTOR21 challenge, which showed significantly better performance by surpassing the state-of-the-artwork and achieved 0.836 and 0.811 concordance index and dice scores, respectively. We made the model and code publicly available.

Quantitative US Delta Radiomics to Predict Radiation Response in Individuals with Head and Neck Squamous Cell Carcinoma

Purpose To investigate the role of quantitative US (QUS) radiomics data obtained after the 1st week of radiation therapy (RT) in predicting treatment response in individuals with head and neck squamous cell carcinoma (HNSCC). Materials and Methods This prospective study included 55 participants (21 with complete response [median age, 65 years {IQR: 47-80 years}, 20 male, one female; and 34 with incomplete response [median age, 59 years {IQR: 39-79 years}, 33 male, one female) with bulky node-positive HNSCC treated with curative-intent RT from January 2015 to October 2019. All participants received 70 Gy of radiation in 33-35 fractions over 6-7 weeks. US radiofrequency data from metastatic lymph nodes were acquired prior to and after 1 week of RT. QUS analysis resulted in five spectral maps from which mean values were extracted. We applied a gray-level co-occurrence matrix technique for textural analysis, leading to 20 QUS texture and 80 texture-derivative parameters. The response 3 months after RT was used as the end point. Model building and evaluation utilized nested leave-one-out cross-validation. Results Five delta (Δ) parameters had statistically significant differences (P < .05). The support vector machines classifier achieved a sensitivity of 71% (15 of 21), a specificity of 76% (26 of 34), a balanced accuracy of 74%, and an area under the receiver operating characteristic curve of 0.77 on the test set. For all the classifiers, the performance improved after the 1st week of treatment. Conclusion A QUS Δ-radiomics model using data obtained after the 1st week of RT from individuals with HNSCC predicted response 3 months after treatment completion with reasonable accuracy. Keywords: Computer-Aided Diagnosis (CAD), Ultrasound, Radiation Therapy/Oncology, Head/Neck, Radiomics, Quantitative US, Radiotherapy, Head and Neck Squamous Cell Carcinoma, Machine Learning Clinicaltrials.gov registration no. NCT03908684 Supplemental material is available for this article. © RSNA, 2024.

Does Dual-Energy Computed Tomography Material Decomposition Improve Radiomics Capability to Predict Survival in Head and Neck Squamous Cell Carcinoma Patients? A Preliminary Investigation

OBJECTIVE

Our study objective was to explore the additional value of dual-energy CT (DECT) material decomposition for squamous cell carcinoma of the head and neck (SCCHN) survival prognostication.

METHODS

A group of 50 SCCHN patients (male, 37; female, 13; mean age, 63.6 ± 10.82 years) with baseline head and neck DECT between September 2014 and August 2020 were retrospectively included. Primary tumors were segmented, radiomics features were extracted, and DECT material decomposition was performed. We used independent train and validation datasets with cross-validation and 100 independent iterations to identify prognostic signatures applying elastic net (EN) and random survival forest (RSF). Features were ranked and intercorrelated according to their prognostic importance. We benchmarked the models against clinical parameters. Intraclass correlation coefficients were used to analyze the interreader variation.

RESULTS

The exclusively radiomics-trained models achieved similar ( P = 0.947) prognostic performance of area under the curve (AUC) = 0.784 (95% confidence interval [CI], 0.775-0.812) (EN) and AUC = 0.785 (95% CI, 0.759-0.812) (RSF). The additional application of DECT material decomposition did not improve the model's performance (EN, P = 0.594; RSF, P = 0.198). In the clinical benchmark, the top averaged AUC value of 0.643 (95% CI, 0.611-0.675) was inferior to the quantitative imaging-biomarker models ( P < 0.001). A combined imaging and clinical model did not improve the imaging-based models ( P > 0.101). Shape features revealed high prognostic importance.

CONCLUSIONS

Radiomics AI applications may be used for SCCHN survival prognostication, but the spectral information of DECT material decomposition did not improve the model's performance in our preliminary investigation.

Biomechanical imaging biomarker during chemoradiotherapy predicts treatment response in head and neck squamous cell carcinoma

Objective. For response-adapted adaptive radiotherapy (R-ART), promising biomarkers are needed to predict post-radiotherapy (post-RT) responses using routine clinical information obtained during RT. In this study, a patient-specific biomechanical model (BM) of the head and neck squamous cell carcinoma (HNSCC) was proposed using the pre-RT maximum standardized uptake value (SUVmax) of18F-fluorodeoxyglucose (FDG) and tumor structural changes during RT as evaluated using computed tomography (CT). In addition, we evaluated the predictive performance of BM-driven imaging biomarkers for the treatment response of patients with HNSCC who underwent concurrent chemoradiotherapy (CCRT).Approach. Patients with histologically confirmed HNSCC treated with definitive CCRT were enrolled in this study. All patients underwent CT two times as follows: before the start of RT (pre-RT) and 3 weeks after the start of RT (mid-RT). Among these patients, 67 patients who underwent positron emission tomography/CT during the pre-RT period were included in the final analysis. The locoregional control (LC), progression-free survival (PFS), and overall survival (OS) prediction performances of whole tumor stress change (TS) between pre- and mid-RT computed using BM were assessed using univariate, multivariate, and Kaplan-Meier survival curve analyses, respectively. Furthermore, performance was compared with the pre and post-RT SUVmax, tumor volume reduction rate (TVRR) during RT, and other clinical prognostic factors.Main results. For both univariate, multivariate, and survival curve analyses, the significant prognostic factors were as follows (p< 0.05): TS and TVRR for LC; TS and pre-RT FDG-SUVmaxfor PFS; and TS only for OS. In addition, for 2 year LC, PFS, and OS prediction, TS showed a comparable predictive performance to post-RT FDG-SUVmax.Significance. BM-driven TS is an effective prognostic factor for tumor treatment response after CCRT. The proposed method can be a feasible functional imaging biomarker that can be acquired during RT using only routine clinical data and may provide useful information for decision-making during R-ART.

Macroscopy of specimens from the head and neck

Macroscopic examination of surgical resections from the head and neck may be difficult due to the complex anatomy of this area. Recognition of normal anatomical structures is essential for accurate assessment of the extent of a disease process. Communication with the surgical team, correct specimen orientation and sampling are critical for assessment and the importance of radiological and clinical correlation is emphasised. Tumour involvement at each subsite is highlighted with reference to where there are implications on pathological staging and the potential need for adjuvant therapy.

Deep learning for head and neck semi-supervised semantic segmentation

Objective. Radiation therapy (RT) represents a prevalent therapeutic modality for head and neck (H&N) cancer. A crucial phase in RT planning involves the precise delineation of organs-at-risks (OARs), employing computed tomography (CT) scans. Nevertheless, the manual delineation of OARs is a labor-intensive process, necessitating individual scrutiny of each CT image slice, not to mention that a standard CT scan comprises hundreds of such slices. Furthermore, there is a significant domain shift between different institutions' H&N data, which makes traditional semi-supervised learning strategies susceptible to confirmation bias. Therefore, effectively using unlabeled datasets to support annotated datasets for model training has become a critical issue for preventing domain shift and confirmation bias.Approach. In this work, we proposed an innovative cross-domain orthogon-based-perspective consistency (CD-OPC) strategy within a two-branch collaborative training framework, which compels the two sub-networks to acquire valuable features from unrelated perspectives. More specifically, a novel generative pretext task cross-domain prediction (CDP) was designed for learning inherent properties of CT images. Then this prior knowledge was utilized to promote the independent learning of distinct features by the two sub-networks from identical inputs, thereby enhancing the perceptual capabilities of the sub-networks through orthogon-based pseudo-labeling knowledge transfer.Main results. Our CD-OPC model was trained on H&N datasets from nine different institutions, and validated on the four local intuitions' H&N datasets. Among all datasets CD-OPC achieved more advanced performance than other semi-supervised semantic segmentation algorithms.Significance. The CD-OPC method successfully mitigates domain shift and prevents network collapse. In addition, it enhances the network's perceptual abilities, and generates more reliable predictions, thereby further addressing the confirmation bias issue.

Association of circulating tumor HPV16DNA levels and quantitative PET parameters in patients with HPV-positive head and neck squamous cell carcinoma

Circulating tumor DNA (ctDNA), which circulates in the blood after being shed from cancer cells in the body, has recently gained attention as an excellent tumor marker. The purpose of this study was to evaluate whether ct human papillomavirus (HPV) 16 DNA (ctHPV16DNA) levels were associated with quantitative PET parameters in patients with HPV-positive head and neck (HN) squamous cell carcinoma (SCC). Fifty patients with oropharyngeal SCC (OPSCC) and 5 with SCC of unknown primary (SCCUP) before treatment were included. They all underwent blood sampling to test ctHPV16DNA levels and FDG PET-CT examinations. Quantitative PET parameters included SUVmax, metabolic tumor volume (MTV), MTV of whole-body lesions (wbMTV), and 56 texture features. ctHPV16DNA levels were compared to texture features of primary tumors in OPSCC patients (Group A) or the largest primary or metastatic lymph node lesions in OPSCC and SCCUP patients (Group B) and to other PET parameters. Spearman rank correlation test and multiple regression analysis were used to confirm the associations between ctHPV16DNA levels and PET parameters. ctHPV16DNA levels moderately correlated with wbMTV, but not with SUVmax or MTV in Groups A and B. ctHPV16DNA levels exhibited a weak negative correlation with low gray-level zone emphasis in Groups A and B. Multiple regression analysis revealed that wbMTV and high gray-level zone emphasis were the significant factors for ctHPV16DNA levels in Group B. These results were not observed in Group A. This study demonstrated that ctHPV16DNA levels correlated with the whole-body tumor burden and tumor heterogeneity visualized on FDG PET-CT in patients with HPV-positive HNSCC.

Reproducibility of diffusion-weighted magnetic resonance imaging in head and neck cancer assessed on a 1.5 T MR-Linac and comparison to parallel measurements on a 3 T diagnostic scanner

BACKGROUND AND PURPOSE

Before quantitative imaging biomarkers (QIBs) acquired with magnetic resonance imaging (MRI) can be used for interventional trials in radiotherapy (RT), technical validation of these QIBs is necessary. The aim of this study was to assess the reproducibility of apparent diffusion coefficient (ADC) values, derived from diffusion-weighted (DW) MRI, in head and neck cancer using a 1.5 T MR-Linac (MRL) by comparison to a 3 T diagnostic scanner (DS).

MATERIAL AND METHODS

DW-MRIs were acquired on MRL and DS for 15 head and neck cancer patients before RT and in week 2 and rigidly registered to the planning computed tomography. Mean ADC values were calculated for submandibular (SG) and parotid (PG) glands as well as target volumes (TV, gross tumor volume and lymph nodes), which were delineated based on computed tomography. Mean absolute ADC differences as well as within-subject coefficient of variation (wCV) and intraclass correlation coefficients (ICCs) were calculated for all volumes of interest.

RESULTS

A total of 23 datasets were analyzed. Mean ADC difference (DS-MRL) for SG, PG and TV resulted in 142, 254 and 93·10-6 mm2/s. wCVs/ICCs, comparing MRL and DS, were determined as 13.7 %/0.26, 24.4 %/0.23 and 16.1 %/0.73 for SG, PG and TV, respectively.

CONCLUSION

ADC values, measured on the 1.5 T MRL, showed reasonable reproducibility with an ADC underestimation in contrast to the DS. This ADC shift must be validated in further experiments and considered for future translation of QIB candidates from DS to MRL for response adaptive RT.

Automatic planning for head and neck seed implant brachytherapy based on deep convolutional neural network dose engine

BACKGROUND

Seed implant brachytherapy (SIBT) is an effective treatment modality for head and neck (H&N) cancers; however, current clinical planning requires manual setting of needle paths and utilizes inaccurate dose calculation algorithms.

PURPOSE

This study aims to develop an accurate and efficient deep convolutional neural network dose engine (DCNN-DE) and an automatic SIBT planning method for H&N SIBT.

METHODS

A cohort of 25 H&N patients who received SIBT was utilized to develop and validate the methods. The DCNN-DE was developed based on 3D-unet model. It takes single seed dose distribution from a modified TG-43 method, the CT image and a novel inter-seed shadow map (ISSM) as inputs, and predicts the dose map of accuracy close to the one from Monte Carlo simulations (MCS). The ISSM was proposed to better handle inter-seed attenuation. The accuracy and efficacy of the DCNN-DE were validated by comparing with other methods taking MCS dose as reference. For SIBT planning, a novel strategy inspired by clinical practice was proposed to automatically generate parallel or non-parallel potential needle paths that avoid puncturing bone and critical organs. A heuristic-based optimization method was developed to optimize the seed positions to meet clinical prescription requirements. The proposed planning method was validated by re-planning the 25 cases and comparing with clinical plans.

RESULTS

The absolute percentage error in the TG-43 calculation for CTV V100 and D90 was reduced from 5.4% and 13.2% to 0.4% and 1.1% with DCNN-DE, an accuracy improvement of 93% and 92%, respectively. The proposed planning method could automatically obtain a plan in 2.5 ± 1.5 min. The generated plans were judged clinically acceptable with dose distribution comparable with those of the clinical plans.

CONCLUSIONS

The proposed method can generate clinically acceptable plans quickly with high accuracy in dose evaluation, and thus has a high potential for clinical use in SIBT.

Radiomics from dual-energy CT-derived iodine maps predict lymph node metastasis in head and neck squamous cell carcinoma

OBJECTIVE

To develop and validate an iodine maps-based radiomics nomogram for preoperatively predicting cervical lymph node metastasis (LNM) in head and neck squamous cell carcinoma (HNSCC).

MATERIALS AND METHODS

A total of 278 patients who pathologically confirmed as HNSCC were retrospectively recruited from two medical centers between June 2012 and July 2022. The training set (n = 152) and internal set (n = 67) were randomly selected from medical center A, and the patients from medical center B were enrolled as the external set (n = 69). The minority group in the training set was balanced by the adaptive synthetic sampling (ADASYN) approach. Radiomics features were extracted from dual-energy CT-derived iodine maps at arterial phase (AP) and venous phase (VP), respectively. Three radiomics signatures were constructed to predict the LNM by using a random forest algorithm. The independent clinical predictors for LNM were identified by multivariate analysis and combined with radiomics signatures to establish a radiomic-clinical nomogram. The performance of radiomic-clinical nomogram was evaluated with respect to its discrimination and clinical usefulness.

RESULTS

The AP-VP-incorporated radiomics model exhibited a great predictive performance for LNM prediction with an area under curve (AUC) of 0.885 (95% CI, 0.836-0.933) in ADASYN-training set and confirmed in all validation sets. The nomogram that incorporated AP-VP radiomics signatures, CT-reported LN status, and histological grades yielded AUCs of 0.920 (95% CI, 0.881-0.959) in ADASYN-training set, 0.858 (95% CI, 0.771-0.944) in internal validation, and 0.849 (95% CI, 0.752-0.946) in external validation, with good calibration in all cohorts (p > 0.05). Decision curve analyses indicated the nomogram was clinically useful. In addition, the predictive performance of clinical-radiomics nomogram was also validation in combing cohorts. Stratified analysis confirmed the stability of nomogram, particularly in group negative for CT-reported LNM.

CONCLUSION

Clinical-radiomics nomogram based on iodine maps exhibited promising performance in predicting LNM and providing valuable information for making individualized therapy decisions.

Diffusion weighted imaging with reverse encoding distortion correction: Improvement of image quality and distortion for accurate ADC evaluation in in vitro and in vivo studies

PURPOSE

The purpose of this in vivo study was to determine the effect of reverse encoding direction (RDC) on apparent diffusion coefficient (ADC) measurements and its efficacy for improving image quality and diagnostic performance for differentiating malignant from benign tumors on head and neck diffusion-weighted imaging (DWI).

METHODS

Forty-eight patients with head and neck tumors underwent DWI with and without RDC and pathological examinations. Their tumors were then divided into two groups: malignant (n = 21) and benign (n = 27). To determine the utility of RDC for DWI, the difference in the deformation ratio (DR) between DWI and T2-weighted images of each tumor was determined for each tumor area. To compare ADC measurement accuracy of DWIs with and without RDC for each patient, ADC values for tumors and spinal cord were determined by using ROI measurements. To compare DR and ADC between two methods, Student's t-tests were performed. Then, ADC values were compared between malignant and benign tumors by Student's t-test on each DWI. Finally, sensitivity, specificity and accuracy were compared by means of McNemar's test.

RESULTS

DR of DWI with RDC was significantly smaller than that without RDC (p < 0.0001). There were significant differences in ADC between malignant and benign lesions on each DWI (p < 0.05). However, there were no significant difference of diagnostic accuracy between the two DWIs (p > 0.05).

CONCLUSION

RDC can improve image quality and distortion of DWI and may have potential for more accurate ADC evaluation and differentiation of malignant from benign head and neck tumors.

A CT-based Deep Learning Radiomics Nomogram for the Prediction of EGFR Mutation Status in Head and Neck Squamous Cell Carcinoma

RATIONALE AND OBJECTIVES

Accurately assessing epidermal growth factor receptor (EGFR) mutation status in head and neck squamous cell carcinoma (HNSCC) patients is crucial for prognosis and treatment selection. This study aimed to construct and validate a contrast-enhanced computed tomography (CECT)-based deep learning radiomics nomogram (DLRN) to predict EGFR mutation status of HNSCC.

MATERIALS AND METHODS

A total of 300 HNSCC patients who underwent CECT scans were enrolled in this study. Participants from two hospitals were separated into a training set (n = 200, 56 EGFR-negative and 144 EGFR-positive) from one hospital and an external test set from the other hospital (n = 100, 37 EGFR-negative and 63 EGFR-positive). The least absolute shrinkage and selection operator method was used to select the key features from CECT-based manually extracted radiomics (MER) features and features automatically extracted using a deep learning model (DL, extracted using a GoogLeNet model). The selected independent clinical factors, MER features, and DL features were then combined to construct a DLRN. The DLRN's performance was evaluated using receiver operating characteristics curves.

RESULTS

Five MER and six DL features were finally chosen. The DLRN, which includes "gender" and "necrotic areas," along with the selected features, predicted EGFR mutation status of HNSCC (EGFR-negative vs. positive) well in both the training (area under the curve [AUC], 0.901) and test (AUC, 0.875) sets.

CONCLUSION

A DLRN using CECT was built to predict EGFR mutation in HNSCC. The model showed high predictive ability and may aid in treatment selection and patient prognosis.